Liquid biopsy and precision medicine: Characterization of circulating biomarkers in patients with advanced lung cancer

INTERNATIONAL DOCTORAL SCHOOL OF THE USC Patricia Mondelo Macía PhD Thesis Liquid Biopsy and Precision Medicine: Characterization of Circulating Biomarkers in Patients with Advanced Lung Cancer Santiago de Compostela, 2023 Doctoral Programme in Molecular Medicine DOCTORAL THESIS LIQUID BIOPSY AND PRECISION MEDICINE: CHARACTERIZATION OF CIRCULATING BIOMARKERS IN PATIENTS WITH ADVANCED LUNG CANCER Patricia Mondelo Macía INTERNATIONAL DOCTORAL SCHOOL OF THEUNIVERSITY OF SANTIAGO DE COMPOSTELA DOCTORAL PROGRAMME IN MOLECULAR MEDICINE SANTIAGO DE COMPOSTELA 2023 DECLARACIÓN DA AUTORA DA TESE Dna. Patricia Mondelo Macía Título da tese: “Liquid biopsy and precision medicine: Characterization of circulating biomarkers in patients with advanced lung cancer” Presento a miña tese, seguindo o procedemento axeitado ao Regulamento, e declaro que: 1) A tese abarca os resultados da elaboración do meu traballo. 2) De ser o caso, na tese faise referencia ás colaboracións que tivo este traballo. 3) Confirmo que a tese non incorre en ningún tipo de plaxio doutros autores nin de traballos presentados por min para a obtención doutros títulos. 4) A tese é a versión definitiva presentada para a súa defensa e coincide a versión impresa coa presentada en formato electrónico. E comprométome a presentar o Compromiso Documental de Supervisión no caso de que o orixinal non estea na Escola. En Santiago de Compostela, 12 de Abril de 2023. Fdo. Patricia Mondelo Macía AUTORIZACIÓN DO DIRECTOR / TITOR DA TESE “Liquid biopsy and precision medicine: Characterization of circulating biomarkers in patients with advanced lung cancer” Dna. Laura Muinelo Romay D. Roberto Díaz Peña D. Rafael López López INFORMAN: Que a presente tese, correspóndese co traballo realizado por Dna. Patricia Mondelo Macía, baixo a miña dirección/titorización, e autorizo a súa presentación, considerando que reúne os requisitos esixidos no Regulamento de Estudos de Doutoramento da USC, e que como director desta non incorre nas causas de abstención establecidas na Lei 40/2015. De acordo co indicado no Regulamento de Estudos de Doutoramento, declara tamén que a presente tese de doutoramento é idónea para ser defendida en base á modalidade de Monográfica con reprodución de publicacións, nos que a participación do/a doutorando/a foi decisiva para a súa elaboración e as publicacións se axustan ao Plan de Investigación. En Santiago de Compostela, 12 de Abril de 2023 Fdo. Laura Muinelo Romay (Directora) Fdo. Roberto Díaz Peña (Director) Fdo. Rafael López López (Titor) I, Patricia Mondelo Macía, author of this thesis, declare that I am the author of one patent developed in the context of the present thesis: MAPRI: In vitro method for predicting cancer patient response to PD-1 and/or PD-L1 inhibitors; EP No. 23 382 377.2. I, Patricia Mondelo Macía, author of this thesis, have no other conflicts of interest to declare. I, Patricia Mondelo Macía, author of this thesis, declare that, unless clarified, all the images presented in this thesis have been produced by the author. In the case of images reused or adapted from other manuscripts, permission has been granted by the publishers, and the mode of legal use has been clarified at the bottom of the corresponding figures. AGRADECIMIENTOS 17 AGRADECIMIENTOS ¡¡¡Finito!!! Después de casi 6 años, ha llegado el día. El día con el que se cierra esta gran etapa, la cual no habría sido posible ni tan increíble de no ser por todos vosotros. En primer lugar, quiero agradecer a los pacientes y a sus familias, los que luchan cada día contra esta enfermedad y a pesar de sus circunstancias, apuestan por la investigación, haciendo posible nuestro trabajo. Esto va por vosotros. Gracias también a todas y cada una de las personas que participaron en la Campaña de Biopsia Líquida, por creer que el progreso y la investigación son herramientas clave para luchar contra esta patología. Gracias al Dr. López, jefe de Servicio de Oncología. Rafa, en primer lugar, gracias por la oportunidad de dejarme formar parte de ONCOMET. Gracias por ser el motor de este grupo, apostando por la investigación, convirtiéndonos así en un centro de referencia. A Laura y a Roberto, mis directores y los guías en este proceso de aprendizaje. Laura, gracias por estar siempre, por ver siempre el lado bueno de las cosas y por preocuparte por mi presente y mi futuro. Roberto, gracias por enseñarme desde el minuto uno y contar siempre conmigo, por esas reuniones en las que entro con mil dudas y salgo con cero. Infinitas gracias. Gracias también a los miembros de la Unidad Mixta, en especial a Gloria y a Ana Dávila, por la ayuda recibida estos últimos meses, junto con Susana Bravo, responsable de la unidad de proteómica del IDIS. A Mercedes, nuestra odontóloga de excelencia. Gracias por tu actitud positiva y tener siempre tiempo para recibirnos. Gracias al grupo de ONCOMET en general, pero especialmente a Luis León y a Jorge García, por ayudarme en cualquier duda clínica y echar una mano siempre que es posible. Gracias a enfermer@s, auxiliares y oncólog@s, que hacen posible este trabajo. Gracias al grupo de ensayos clínicos, en primer lugar, por estar siempre dispuestos a sacarnos sangre, y segundo, por haberme acogido tan bien en estos últimos meses. 18 Volevo ringraziare anche i colleghi con cui ho lavorato a Milano. Grazie a Daidone e Vera per avermi aperto le porte del loro laboratorio. A Marta, Rosita, Patrizia, Silvia, Elisa, eccetera per avermi aiutato e fatto sentire a casa. In particolare, ringrazio Carolina per avermi ospitato il primo mese e per avermi aperto le porte della sua casa e del suo mondo. Gracias también a los colaboradores externos, por su ayuda tanto de recursos como de conocimientos. Al laboratorio 17, mi labo. Gracias a Nasas group Alba, Lore y Jorge por estar dispuestos siempre a ayudar, y a enseñarnos lo que haga falta. En especial, gracias a Jorge por haber sacado tiempo para mostrarnos a los predocs todas las posibles vidas que existen después de la tesis. Gracias a Alicia y a Ramón, mis primeritos profesores junto con Manuel Abreu, por acompañarme en mis primeros días de incertidumbre y torpeza. Ali, aunque los inicios sean duros (jeje), gracias por enseñarme una y mil cosas, por conseguir lo que sea cuando sea y por ser una fuente inagotable de caramelos. Ramón, gracias por todo el aprendizaje como profesional pero también como persona, eres grande (y no solo literalmente). Gracias incluso por tus chistes malos que alegran días regulinchis. Andrea, Aroa, Yoel, Ana Torres, los recién llegados, gracias por vuestras sonrisas constantes y a por todas. Mimí, el último gran descubrimiento, no cambies y ¡a tope con el código! Rachel, Anita gracias por ese aire fresco que aportáis. Anita, sigue con esa fuerza y energía que tú tienes, llegarás a todo lo que te propongas. Rachel, mi chula. Te has convertido en una gran compañera y amiga a pesar de tenerte desde hace poquito tiempo. Eternamente agradecida de haber coincidido contigo. Gracias por acompañarme en estos últimos meses a planes tan de jóvenes como gimnasio, biblioteca y música del 2000. A tope con la tesis, no dudes que estaré ahí para verla, tú puedes. Carlos, Aida, Aitor, Óscar, para vosotros no tengo las suficientes palabras para agradeceros lo vivido en esta etapa. Carlos, gracias por tu ayuda constante, por Grecia, Barcelona, por no dejarte pegar la COVID, esos viajes de post trabajo a la playa, cafés, 19 cenas, etc. Gracia por ser apoyo, compañero y sobre todo, amigo. Aida, mi coruñesa fav. Gracias por ayudarme siempre, por los desahogos, por los cotilleos, por Sanxenxo, Málaga o Santiago. Estoy segura de que este camino continuará contigo cerca. Aitor, gracias por tu buen humor constante, por tu ayuda en cualquier momento, y estar siempre dispuesto a una buena caña after-work. Gracias también a Manuel, un Epi más. Y no me olvido de Nico, el tercer integrante de Epi, dispuesto siempre a una buena orquesta (si no hay fútbol). Gracias al doctor Rapado, Oscarín por ser siempre lugar de paz y tranquilidad. Gracias por ser como eres Óscar. Nos vemos en Oporto!! Quiero agradecer también a los pedazos de doctores con los que empezó esta aventura. Thais, Nuria, Inés, Pablo, Carol. Gracias a todos. Pablo, nuestro murciano favorito, gracias por todo lo que nos diste y por seguir escapándote cuando es posible. Carol, mi Caroline. Gracias por todos esos años de compi de mesa, de escritura de inicio de tesis, de bodorrio y por convertirte en una amiga. Se te echa de menos. Saínza, la nano doctora. Gracias por tus consejos de diseño profesional y ser la persona más organizada del mundo. Estoy segura de que llegarás lejos. Gracias a Sandra Alijas, por esos primeros años. Gracias a las mini nanos, María C, Soraya, Emma, Laura, Jen, por vuestra sonrisa permanente. Sandra Díez, a ti, gracias por demasiado. Por ser una amiga fuera y dentro, por tu ayuda tanto profesional como personal. Por todos esos “si” ante cualquier plan. Por más surf de desconexión, más furgonetas, y más ver mundo. Gracias por aparecer y seguir. Tengo claro que no podría haberme tropezado con unos compañeros mejores, “El club de Wendoline”, “El club de Oncoline”. Sois la magia y la fuerza que tanto se necesita en días grises. Gracias por hacerme formar una segunda familia con la que compartir esta experiencia, con todo lo bueno que hemos vivido y esas cosas malas que al final no lo son tanto. Por responder a todos los “qué preferirías” y hacer que estar 24/7 con la tesis en la mente se lleve mejor. Se cierra una etapa chicos. Eternamente agradecida de encontraros en ella. Os llevo conmigo. 20 No puedo terminar sin darle las gracias a mis amigos de siempre, a mi grupo “Topo the coldmind”, por ser únicos, escuchar mis discursos, aportar otras visiones y ayudarme a desconectar en nuestro querido valle. Dani, Óscar, Vasco, Xabi, Ine, Vero, gracias. Especialmente, agradecer a Noe, Yaiza, Nani y Judi, las de siempre, por estar y ser como siempre. Por muchos kilómetros que nos separen, siempre llevo un pedacito de vosotras conmigo. Gracias también a mis compañeros de piso. Desde los inicios hasta ahora. Sergio, Nani, Maye, Lucas y Patri “la nueva”. Gracias, por compartir las diferentes etapas conmigo, siempre escuchando y haciendo de Santiago Chile un dulce hogar. A Peter, por estar, pero sobre todo por escuchar y apaciguar mis momentos de crisis. A mi familia, en especial a mis padrinos, Merce y Moncho por interesarse por cada paper. A mi abuelo Baltasar, gracias por alegrarte de cada paso que doy, escucharme y ser mi confidente y amigo. Eres mi alegría, avó. A mis abuelos, Elena, Inés y Manolo. Aunque no hayáis podido ver este final de camino, gracias por alegraros más que yo cuando empecé esta etapa, por interesaros cuando no entendíais ni lo que hacía. Gracias por todo lo que me disteis. A mis padres. Papá, mamá, gracias por todo lo que me habéis dado para poder llegar hasta aquí. Gracias por apoyarme en todo, animarme y hacerme saber que pasara lo que pasara vosotros estaríais ahí. Esto es solo gracias a vosotros. A mi hermana, Paula y a Carlos, el otro ya casi integrante de la familia. Pauli, mi otra mitad con la que no podría vivir. Gracias por estar a mi lado siempre. Por tu ayuda en lo que sea, cuando sea y cómo sea. En definitiva, gracias a todos y cada una de las personas que se han cruzado en este camino. Esta tesis no podría haber sido posible sin todos vosotros. Nos vemos en los bares y, como diría mi abuelo: ¡Sigan bien! Xuntos avanzamos. 21 INDEX 23 INDEX ABBREVIATIONS AND ACRONYMS 27 RESUMEN in extenso 33 RESUMO in extenso 53 SUMMARY 73 INTRODUCTION 83 1.LUNG CANCER 83 1.1 EPIDEMIOLOGY AND AETIOLOGY 83 1.2 LUNG CANCER CLASSIFICATION 82 1.3 ADVANCED NON-SMALL CELL LUNG CANCER 85 1.4 ADVANCED SMALL CELL LUNG CANCER 96 2.LIQUID BIOPSY 100 2.1 CIRCULATING FREE DNA 103 2.2 CIRCULATING TUMOUR CELLS 110 2.3 OTHER CIRCULATING BIOMARKERS 116 OBJECTIVES 121 CHAPTER I 125 CHAPTER IA: EGFR mutations: simple and rapid blood-based qPCR test to select and guide treatment of advanced NSCLC patients 127 PATRICIA MONDELO MACÍA 24 CHAPTER IB: Development of a blood assay to detect MET alterations using circulating DNA and circulating tumour cells in NSCLC patients 157 CHAPTER II 187 Clinical potential of circulating free DNA and circulating tumour cells in patients with metastatic non-small cell lung cancer treated with pembrolizumab CHAPTER III 229 Plasma circulating free DNA and circulating tumour cells as prognostic biomarkers in small cell lung cancer patients CHAPTER IV 265 Circulating proteins as predictive biomarkers for immunotherapy in metastatic non-small cell lung cancer patients OVERALL DISCUSSION 307 CONCLUSIONS 319 REFERENCES 323 ANNEX 1: Index figures and tables 363 ANNEX 2: List of publications 379 ANNEX 3: Ethical considerations 391 RESUMEN in extenso 33 RESUMEN in extenso El cáncer se define como un conjunto de enfermedades en las que las células crecen de manera descontrolada, provocando fallos en los tejidos a los que afecta. Estas células de crecimiento descontrolado pueden crecer y propagarse a otros tejidos y órganos a través del sistema circulatorio o linfático, y pueden interferir con el funcionamiento normal del cuerpo. Hoy en día, el cáncer es una de las principales causas de muerte a nivel mundial, superado sólo por las enfermedades cardiovasculares. El cáncer de pulmón es el tipo de cáncer que más muertes provoca al año en todo el mundo, con más de 1.8 millones de muertes en 2020. En el 57% de los casos, su diagnóstico se produce en etapas avanzadas en el que el pronóstico de los pacientes es malo, ya que menos del 5.3% sobrevive 5 años. Histológicamente, el cáncer de pulmón se puede dividir en dos tipos: el cáncer de pulmón de células no pequeñas (NSCLC, del inglés non-small cell lung cancer) que constituye el 85% de los casos, y el cáncer de pulmón de células pequeñas (SCLC, del inglés small cell lung cancer), que constituye un 15% de los casos. Este último tipo se caracteriza por su mayor agresividad, siendo su tasa de supervivencia a 5 años de solo el 2.8%. Durante muchos años, la estrategia terapéutica más frecuente para hacer frente al cáncer de pulmón se basaba en el uso de la quimioterapia, la cual bloquea la división celular de las células tumorales. Sin embargo, la quimioterapia también afecta negativamente a células sanas, produciendo efectos secundarios no deseados que disminuyen la calidad de vida de los pacientes. Además, la resistencia al tratamiento con este tipo de fármacos PATRICIA MONDELO MACÍA 34 puede ocurrir con frecuencia. En las últimas décadas, un mayor conocimiento de los mecanismos moleculares del cáncer de pulmón ha permitido el desarrollo y la aprobación de nuevos fármacos específicos contra el cáncer, mejorando éstos la supervivencia y la calidad de vida de los pacientes. En concreto, en el NSCLC el descubrimiento de alteraciones en genes drivers ha permitido el uso de terapias dirigidas donde el fármaco se dirige solamente hacia las células tumorales que presentan diferentes alteraciones. Recientemente la inmunoterapia ha emergido como una terapia alternativa cuando no hay mutaciones en genes drivers. Por otro lado, en el SCLC actualmente no existen terapias dirigidas hacia genes drivers mientras que la inmunoterapia ha sido incorporada al manejo clínico de los pacientes hace relativamente pocos años. En ambos tipos de cáncer, seleccionar el tratamiento adecuado e identificar a aquellos pacientes que se beneficiarán de las diferentes terapias que existen en la actualidad para mejorar las tasas de supervivencia, es esencial. Sin embargo, los actuales biomarcadores no siempre son suficientemente eficaces para seleccionar a aquellos pacientes que responderán a estos tratamientos. Actualmente, la identificación de las alteraciones tumorales en la práctica clínica para seleccionar el mejor tratamiento debe realizarse en el tejido tumoral. Sin embargo, esta aproximación no está exenta de desventajas. En primer lugar, las biopsias sólidas pueden ser invasivas y dolorosas, lo que puede causar malestar y complicaciones en algunos pacientes. En segundo lugar, la accesibilidad al tumor no siempre está disponible en los casos de cáncer de pulmón o la cantidad de muestra obtenida no es suficiente para realizar un análisis molecular extenso. Por último, estas biopsias pueden no ser representativas de todo el tumor, ya que solo se toma una pequeña muestra Resumen in extenso 35 del tejido, lo que puede conducir a errores de diagnóstico y/o una selección de un tratamiento inadecuado. Además, la heterogeneidad tumoral es un factor clave en los pacientes con cáncer avanzado donde el tumor primario no siempre es representativo de la imagen total de la enfermedad. Por último, la monitorización de la enfermedad en tiempo real durante el tratamiento es fundamental para determinar la eficacia de éste y prevenir o adelantarse a la aparición de resistencias que produzcan el avance de la enfermedad. Dado el carácter invasivo de las biopsias sólidas, realizar éstas a diferentes puntos del tratamiento, con el fin de monitorizar la respuesta, no es adecuado. Por todo ello, la necesidad de utilizar alternativas no invasivas y repetitivas para mejorar el diagnóstico y permitir la monitorización de la enfermedad durante el tratamiento del cáncer de pulmón, destaca la importancia de desarrollar técnicas que puedan complementar o reemplazar a las biopsias sólidas. Dentro de este contexto, hace aproximadamente 20 años nace la biopsia líquida, una herramienta alternativa para el manejo de varias enfermedades, entre ellas, el cáncer. La biopsia líquida es una técnica no invasiva que permite detectar y analizar células tumorales y material genético del tumor en muestras de sangre u otros fluidos corporales. Esta técnica revolucionaria ofrece ventajas en comparación con las desventajas previamente mencionadas de las biopsias sólidas: no requiere procedimientos invasivos, se puede realizar con mayor frecuencia durante el tratamiento para monitorizar la progresión del cáncer y detectar posibles resistencias y, además, ofrece información más precisa y completa sobre la heterogeneidad molecular que caracteriza a muchos tumores. El ADN libre circulante (cfDNA, del inglés circulating free DNA), las células tumorales circulantes (CTCs, del inglés circulating tumour cells) y PATRICIA MONDELO MACÍA 36 otros componentes de la biopsia líquida han demostrado su utilidad como biomarcadores predictivos y pronósticos en el campo de la oncología (Figura 1). Sin embargo, no está claro cuál es el mejor abordaje para analizar las alteraciones tumorales utilizando muestras de sangre, cómo se deben cuantificar los diferentes biomarcadores y qué técnicas son las más apropiadas en cada caso. Figura 1. Principales componentes de la biopsia líquida estudiados a lo largo de la presente tesis. En vista de los antecedentes mencionados previamente, las necesidades actuales, y el potencial de la biopsia líquida para el abordaje clínico del cáncer de pulmón, el principal objetivo de esta tesis se centra en el desarrollo y validación de estrategias que nos permitan realizar la caracterización de biomarcadores circulantes de forma eficaz con el fin de mejorar el manejo de pacientes con cáncer de pulmón avanzado en distintos contextos de tratamiento. Para ello, el primer capítulo de esta tesis se centra en el estudio y desarrollo de diferentes metodologías para el análisis del ADN tumoral circulante (ctDNA, del inglés circulating tumour cell) y las CTCs para Células tumorales circulantes ADN libre circulante Proteínas circulantes Resumen in extenso 37 seleccionar y monitorizar diferentes terapias dirigidas en pacientes con NSCLC avanzado. Durante los últimos años, la detección de alteraciones en diferentes receptores tirosina quinasa (RTKs, del inglés receptors tyrosine kinase) como el receptor del factor de crecimiento epidérmico (EGFR) y el receptor del factor de crecimiento hepático (MET o c-MET) en los pacientes con NSCLC se ha convertido en una parte habitual de la práctica clínica. La detección de estas alteraciones permite seleccionar a aquellos pacientes que se beneficiarán de tratamientos dirigidos contra estos genes y las vías que regulan, permitiendo un aumento en la supervivencia de los pacientes. Con este fin, en primer lugar (CHAPTER IA; Figura 2), se han analizado las ventajas que tendría la implementación en la práctica clínica de un ensayo simple y rápido de PCR cuantitativa (qPCR, del inglés quantitative PCR) para detectar alteraciones en el gen EGFR a partir de muestras de sangre. Para ello, 40 pacientes con NSCLC fueron incluidos en el estudio, en los cuales se recogieron muestras de sangre antes de iniciar el tratamiento y durante diferentes puntos de este. En total, 53 muestras de plasma fueron analizadas mediante el ensayo de qPCR “ctEGFR de Idylla™ mutational assay” y los datos obtenidos mediante este sistema fueron comparados con los datos obtenidos en las muestras de biopsia sólida, actualmente el gold estándar en la práctica clínica. La concordancia entre ambas muestras fue del 71.8%. Respecto a los casos discordantes entre ambas muestras, es importante resaltas que en más de la mitad de los casos (6/11) la diferencia de tiempo entre la muestra sólida y la muestra de sangre era superior a 3 meses. Seguidamente, con el fin de determinar cuán sensible era el sistema a estudio, se compararon los resultados obtenidos de la técnica Idylla™ con otras dos técnicas de análisis de alteraciones a partir de muestras de plasma: la tecnología PATRICIA MONDELO MACÍA 38 BEAMing (una técnica basada en PCR digital (ddPCR, del inglés digital droplet PCR)) y técnicas de secuenciación masiva (NGS, del inglés nextgeneration sequencing) empleando un panel dirigido (AVENIO ctDNA Expanded panel). Los resultados obtenidos mostraron una gran concordancia de la estrategia de qPCR con la tecnología BEAMing y el panel de secuenciación AVENIO (88.9% y 93.3%, respectivamente). Sin embargo, a pesar de la buena concordancia obtenida, aquellas alteraciones que se encontraban con una frecuencia alélica igual o menor del 0.13% no fueron detectables mediante la tecnología Idylla™. Por otro lado, muestras procedentes de 10 pacientes con NSCLC se analizaron antes y durante el tratamiento con el sistema Idylla™ con el fin de monitorizar la respuesta al tratamiento y detectar de forma temprana la aparición de posibles resistencias. A pesar de que en nuestra cohorte no se detectó la aparición de resistencias al tratamiento, cambios en los niveles de cfDNA y variaciones en las mutaciones sugieren que el sistema Idylla™ constituye un buen enfoque para monitorizar la respuesta a la terapia en pacientes bajo terapias con inhibidores de tirosina quinasa (TKIs, del inglés tyrosine kinase inhibitors). En general, los resultados obtenidos en este estudio piloto permiten concluir que el ensayo de “Idylla™ ctEGFR mutational assay” es una herramienta rápida y sencilla que podría emplearse como una primera prueba para detectar mutaciones de EGFR en la rutina clínica habitual, obteniendo el resultado el mismo día de consulta del paciente, evitando así largos tiempos de espera. Sin embargo, es importante destacar que en aquellos casos que no se detecte ninguna mutación debido al bajo contenido de ctDNA en plasma, se debe emplear otra técnica más sensible como una estrategia basada en ddPCR o NGS para confirmar el resultado negativo. En relación con esto, Resumen in extenso 39 algunos aspectos deben tenerse en cuenta para la selección de los pacientes que se beneficiarán de esta plataforma. Por ejemplo, los tumores con múltiples ubicaciones metastásicas y altamente vascularizados tendrán niveles más altos de ctDNA y, por lo tanto, el genotipado de mutaciones clínicamente relevantes será más fácil de detectar mediante la técnica Idylla™. Figura 2. Resumen gráfico del trabajo realizado en el CHAPTER IA. Abreviaturas: NSCLC, del inglés non-small cell lung cancer, cáncer de pulmón no microcítico; ctDNA, del inglés circulating tumour DNA, DNA tumoral circulante; qPCR, PCR cuantitativa. Siguiendo con las diversas mutaciones que se pueden encontrar en el cáncer de pulmón, es ampliamente reconocido que distintas mutaciones en el gen MET pueden causar resistencia a ciertas terapias, especialmente a TKIs. En la actualidad, las terapias con inhibidores de MET permiten aumentar el pronóstico de los pacientes con alteraciones en este gen, sin embargo, no hay métodos estandarizados para confirmar dichas alteraciones moleculares. Por ello, otro de los objetivos de la presente tesis se centró en desarrollar herramientas basadas en biopsia líquida para evaluar las alteraciones de MET en pacientes con cáncer y determinar su utilidad, centrándonos particularmente en pacientes con NSCLC (CHAPTER IB; Figura 3). Para alcanzar este objetivo, se reclutaron un total de 174 pacientes con diferentes PATRICIA MONDELO MACÍA 40 tipos de cáncer y 49 controles sanos. En esta cohorte de pacientes, se aplicó un ensayo basado en ddPCR para analizar el número de copias de MET en muestras de cfDNA en pacientes con tumores refractarios. Tras validar técnicamente el ensayo, se analizaron 77 pacientes con NSCLC y un 12.99% de los pacientes presentó amplificación de MET, siendo posibles candidatos a fármacos inhibidores de MET. Por otro lado, complementando nuestra estrategia no invasiva para caracterizar el estado de MET, estudiamos la población de CTCs aplicando dos enfoques diferentes (sistemas CellSearch® y Parsortix) para determinar su número y el nivel de expresión de la proteína c-MET en 16 pacientes con NSCLC metastásico. Los resultados obtenidos en ambos sistemas mostraron una concordancia del 62.5% a la hora de detectar la presencia de CTCs, sin embargo, el sistema Parsortix mostró una mayor eficiencia ya que permitió la detección de un mayor número de CTCs (en el 56.2% de los pacientes se detectaron CTCs frente al 31.25% usando el sistema CellSearch®). Respecto a la detección de la sobreexpresión de c-MET, los datos obtenidos sugieren que el sistema Parsortix es capaz de aislar más CTCs con sobreexpresión de c-MET (un 37.1% de las CTCs detectadas mostraron sobreexpresión) que el sistema CellSearch® (10.9%), resultados que concuerdan con el fenotipo mesenquimal propio de estas CTCs. Estos resultados preliminares indican que el sistema Parsortix sería más eficaz a la hora de detectar la sobreexpresión de c-MET en CTCs de pacientes con NSCLC tras varias líneas de tratamiento, sin embargo, estudios con una cohorte mayor de pacientes son necesarios para obtener resultados más robustos. Finalmente, dos casos de pacientes con NSCLC evidenciaron cómo la detección de las alteraciones de MET (amplificación de MET y/o sobreexpresión de c-MET mediante el sistema CellSearch® o Parsortix) puede ser informativa para la selección de tratamiento, así como en la monitorización Resumen in extenso 47 al tumor, incluyendo su composición inmune o el estado inflamatorio del mismo. Con el objetivo de identificar proteínas plasmáticas con interés para guiar el tratamiento con inmunoterapia en pacientes con NSCLC avanzado, se diseñó un estudio en el que se incluyeron 64 pacientes tratados con pembrolizumab. Las muestras de sangre fueron recogidas en diferentes puntos del tratamiento: antes de iniciar el tratamiento, a las 6 y 12 semanas tras el inicio de la terapia y en el momento de la progresión de la enfermedad. En total se recogieron 171 muestras en las cuales se analizaron las proteínas presentes en el plasma mediante la tecnología SWATH-MS (del inglés, sequential window acquisition of all theoretical fragment ion mass spectra), un método de adquisición independiente de datos que permite cuantificar cientos de proteínas en un solo ensayo. Un análisis comparativo entre muestras de pacientes respondedores y no respondedores permitió identificar 324 proteínas diferencialmente expresadas entre ambos grupos. De ellas, 172 proteínas presentaban niveles más altos en los pacientes respondedores mientras que 152 proteínas, presentaban niveles más altos en los no respondedores. A partir del global de proteínas diferencialmente expresadas, se identificó una firma proteómica de 7 proteínas (ATG9A, HPS5, DCDC2, PGTA, FIL1L, LZTL1 y SPTN2) que permite discriminar a los pacientes que responden al tratamiento con inmunoterapia. Individualmente las proteínas mostraban un poder de discriminación alto, con áreas bajo la curva (AUC, del inglés área under the curve) entre 0.72 y 0.78, mejorando así el poder predictivo de la expresión de PD-L1 en tejido (AUC= 0.638). El uso combinado de las 7 proteínas mejoró estos valores, obteniendo una AUC=1. Por otro lado, análisis de supervivencia revelaron que niveles bajos de ATG9A se asociaron de manera significativa con una mejor PFS, mientras que los valores altos de SPTN2 mostraron valor pronóstico positivo para PFS. Así PATRICIA MONDELO MACÍA 48 mismo, niveles bajos de HPS5 y DCDC2 se asociaron de manera significativa con una mayor OS. Finalmente, la cuantificación de estas 7 proteínas a diferentes puntos del tratamiento reveló que el modelo propuesto permite predecir la respuesta tanto a las 6 semanas como a las 12 semanas tras el inicio del tratamiento con pembrolizumab. En resumen, los resultados derivados del capítulo IV de esta tesis, nos permitieron identificar una firma proteómica de 7 proteínas plasmáticas con potencial para seleccionar aquellos pacientes que se beneficiarán de inmunoterapia basada en anti-PD1. Figura 6. Resumen gráfico del trabajo realizado en el CHAPTER IV. Abreviaturas: NSCLC, del inglés non-small cell lung cancer, cáncer de pulmón no microcítico. En resumen, el cáncer de pulmón metastásico es una enfermedad compleja y heterogénea que requiere un enfoque multidisciplinario para su diagnóstico y tratamiento. Las “ómicas”, incluyendo la genómica y proteómica, son herramientas esenciales para identificar biomarcadores y dianas terapéuticas para abordar el tratamiento del cáncer. Además, el estudio de biopsias líquidas, incluyendo el análisis de CTCs, cfDNA y otras moléculas presentes en la sangre, ha surgido como una herramienta no invasiva y altamente sensible para la detección del cáncer y el seguimiento de la Resumen in extenso 49 respuesta al tratamiento, donde cada componente proporciona información única sobre el tumor y su microambiente. La combinación de estos marcadores circulantes con las diferentes “ómicas” y datos clínicos como el evolutivo de los pacientes o las imágenes radiológicas de seguimiento, ofrece un enfoque poderoso para comprender la complejidad del cáncer de pulmón metastásico y desarrollar estrategias de tratamiento personalizadas. Dadas las diferencias individuales y la heterogeneidad molecular del cáncer de pulmón, las distintas estrategias de análisis de biopsia líquida aplicadas en la presente tesis nos aportan una visión más integral de la enfermedad. Centrando nuestros estudios en una muestra no invasiva de sangre hemos podido comprender mejor la biología del cáncer de pulmón y sobre todo, desarrollar herramientas moleculares con potencial para guiar el tratamiento y mejorar la esperanza y calidad de vida en pacientes con cáncer de pulmón metastásico. En concreto, hemos demostrado que: a) podemos genotipar el estado de EGFR y MET en cfDNA mediante una técnica de qPCR y ddPCR de forma robusta en tumores con alta carga tumoral, b) el análisis del cfDNA representa una técnica efectiva para pronosticar la evolución de los pacientes con NSCLC o SCLC avanzado y monitorizar la respuesta a inmunoterapia en los primeros, c) podemos caracterizar la expresión de proteínas clínicamente relevantes como MET o PD-L1 en la población de CTCs de pacientes con cáncer de pulmón, d) el número de CTCs combinado con los niveles de cfDNA aporta información pronóstica en pacientes con NSCLC y, e) la caracterización del proteoma plasmático nos puede proporcionar biomarcadores con valor para identificar a los pacientes que van a presentar una mejor respuesta al tratamiento con inmunoterapia. PATRICIA MONDELO MACÍA 50 Finalmente, es importante destacar que nuestros resultados han permitido estandarizar y validar estrategias de análisis que representan actualmente una herramienta diagnóstica de utilidad real en la práctica clínica de nuestro hospital, pero también identificar nuevos biomarcadores cuyo valor clínico deberá ser determinado en futuros estudios prospectivos. RESUMO in extenso 53 RESUMO in extenso O cancro defínese como un conxunto de enfermidades nas que as células crecen de maneira descontrolada, provocando fallos nos tecidos aos que afecta. Estas células de crecemento descontrolado poden medrar e propagarse a outros tecidos e órganos a través do sistema circulatorio ou linfático, e poden interferir co funcionamento normal do corpo. Hoxe en día, o cancro é unha das principais causas de morte a nivel mundial, superado só polas enfermidades cardiovasculares. O cancro de pulmón é o tipo de cancro que máis mortes provoca ao ano en todo o mundo, con máis de 1.8 millóns de mortes en 2020. No 57% dos casos, o seu diagnóstico prodúcese en etapas avanzadas, no cal o prognostico dos pacientes é moi baixo, onde menos do 5.3% sobrevive 5 anos. Histolóxicamente, o cancro de pulmón pode dividirse en dous tipos: o cancro de pulmón de células non pequenas (NSCLC, do inglés non-small cell lung cancer) que constitúe o 85% dos casos, e o cancro de pulmón de células pequenas (SCLC, do inglés small cell lung cancer), que constitúe o 15% dos casos. Este último tipo caracterízase pola súa maior agresividade sendo a súa taxa de supervivencia a 5 anos de so o 2.8%. Durante moitos anos, a estratexia terapéutica máis frecuente para facer fronte ao cancro de pulmón baseábase no uso da quimioterapia, a cal bloquea a división celular das células cancerosas. Sen embargo, a quimioterapia tamén afecta negativamente as células sas, producindo efectos secundarios non desexados que diminúen a calidade de vida dos pacientes. Por outro lado, a resistencia a este tipo de fármacos pode ocorrer con frecuencia. Nas últimas décadas, un maior coñecemento dos mecanismos moleculares do cancro de pulmón permitiu o desenvolvemento e a aprobación de novos fármacos máis PATRICIA MONDELO MACÍA 54 específicos contra o cancro, mellorando a supervivencia e a calidade de vida dos pacientes. En concreto, no NSCLC o descubrimento de alteracións en xenes drivers permitiu o uso de terapias dirixidas onde o fármaco diríxese só cara as células canceríxenas que presentan ditas alteracións. Recentemente a inmunoterapia emerxeu como unha terapia alternativa cando as mutacións nos xenes drivers non están presentes. Doutra banda, no SCLC actualmente non existen terapias dirixidas cara xenes drivers mentres que a inmunoterapia foi incorporada ao manexo clínico dos pacientes fai relativamente poucos anos. En ambos tipos de cancro, seleccionar o tratamento adecuado e identificar a aqueles pacientes que se beneficiarán das diferentes terapias que existen na actualidade para mellorar as taxas de supervivencia é esencial. Non obstante, os actuais biomarcadores non sempre son o suficientemente eficaces para seleccionar aqueles pacientes que responderán a eles. Actualmente, a identificación das alteracións tumorais na práctica clínica para seleccionar o mellor tratamento debe realizarse no tecido tumoral. Non obstante, esta aproximación non está exenta de desvantaxes. En primeiro lugar, as biopsias sólidas poden ser invasivas e dolorosas, o que pode causar malestar e complicacións en algúns pacientes. En segundo lugar, a accesibilidade ao tumor non sempre está dispoñible nos casos de cancros de pulmón ou a cantidade de mostra obtida non é suficiente para realizar un análise molecular extenso. Por último, estas biopsias poden non ser representativas de todo o tumor, xa que só se toma unha pequena mostra do tecido, o que pode conducir a erros de diagnóstico e/ou tratamento inadecuado. Ademais, a heteroxeneidade tumoral é un factor clave nos pacientes con cancros avanzados onde o tumor primario non sempre é representativo da imaxe total da enfermidade. Por outra banda, a vixilancia da Resumo in extenso 55 enfermidade en tempo real durante o tratamento é fundamental para determinar a eficacia da terapia e previr ou anticiparse á aparición de resistencias que produzan o avance da enfermidade. Dado o carácter invasivo das biopsias sólidas, realizar estas de maneira repetitiva en diferentes puntos do tratamento non é adecuado. Por todo isto, a necesidade de utilizar alternativas non invasivas e repetitivas para mellorar o diagnóstico e permitir a vixilancia en tempo real da enfermidade, destaca a importancia de desenvolver técnicas que poidan complementar ou substituír as biopsias sólidas. Dentro deste contexto, fai aproximadamente 20 anos nace a biopsia líquida, unha ferramenta alternativa para o manexo de varias enfermidades, entre elas, o cancro. A biopsia líquida é unha técnica non invasiva que permite detectar e analizar células tumorais e material xenético do tumor en mostras de sangue ou outros fluídos corporais. Esta técnica revolucionaria ofrece vantaxes en comparación coas desvantaxes previamente mencionadas das biopsias sólidas: non require procedementos invasivos, pódese realizar con maior frecuencia durante o tratamento do cancro e permite a detección de posibles resistencias. Ademais, ofrece información máis precisa e completa sobre a heteroxeneidade da enfermidade molecular que caracteriza a maioría dos tumores. O ADN libre circulante (cfDNA, do inglés circulating free DNA), as células tumorais circulantes (CTCs, do inglés circulating tumour cells) e outros compoñentes da biopsia líquida demostraron a súa utilidade como biomarcadores preditivos e prognósticos no campo da oncoloxía (Figura 7). Non obstante, non está claro cal é o mellor abordaxe para analizar as alteracións tumorais utilizando mostras de sangue, como se deben cuantificar PATRICIA MONDELO MACÍA 56 os diferentes biomarcadores e que técnicas son as máis apropiadas en cada caso. Figura 7. Principais compoñentes da biopsia líquida estudados ao longo da presente tese. Dado os antecedentes mencionados previamente, as necesidades actuais e o potencial da biopsia líquida para o abordaxe clínico do cancro de pulmón, o principal obxectivo desta tese céntrase no desenrolo e validación de estratexias que nos permitan realizar a caracterización de biomarcadores circulantes de forma eficaz co fin de mellorar o manexo de pacientes con cancro de pulmón avanzado en distintos contextos de tratamento. Para isto, o primeiro capítulo desta tese centrase no estudo e desenvolvemento de diferentes metodoloxías para o análise do ADN tumoral circulante (ctDNA, do inglés circulating tumour cell) e as CTCs para seleccionar e monitorizar diferentes terapias dirixidas utilizadas en pacientes con NSCLC avanzado. Durante os últimos anos, a detección de alteracións en diferentes receptores tirosina quinasa (RTKs, do inglés receptor tyrosine kinases) como o receptor do factor de crecemento epidérmico (EGFR) e o receptor do factor de crecemento hepático (MET ou c-MET) nos pacientes con NSCLC, converteuse nunha parte habitual da práctica clínica. A detección Células tumorais circulantes ADN libre circulante Proteínas circulantes Resumo in extenso 63 death-ligand 1) nesta poboación circulante. O sistema que presentou unha maior taxa de detección de CTCs PD-L1 positivas foi o sistema Parsortix. Sen embargo, a expresión de PD-L1 en CTCs non mostrou ningunha asociación coa supervivencia dos pacientes, con ningunha das dúas técnicas empregadas. En contraste, os resultados obtidos indicaron que os pacientes nos que se detectaron CTCs utilizando o sistema CellSearch® tiveron unha PFS e unha OS significativamente máis curtas que os pacientes que non tiñan CTCs, independentemente do estado de PD-L1. Finalmente, análises de regresión multivariante mostraron que a combinación da presenza de CTCs detectadas mediante o sistema CellSearch® cos niveis de cfDNA antes de iniciar o tratamento nos pacientes con NSCLC avanzado, mostrou un valor como biomarcador prognostico independente da progresión da enfermidade. Así, o segundo capítulo da presente tese revelou que a combinación dos niveis basais de CTCs e cfDNA aporta información relevante sobre o prognóstico e a evolución da resposta á terapia con pembrolizumab en pacientes con NSCLC metastático. Utilizando dito enfoque, é posible identificar un subgrupo de pacientes negativos para CTCs e que presentan niveis baixos de cfDNA, os cales se benefician particularmente do tratamento con inmunoterapia. Por outro lado, os resultados obtidos indican que avaliar os niveis de cfDNA ás 12 semanas de iniciar o tratamento podería permitir aos médicos decidir se o beneficio clínico que experimenta o paciente é suficiente para continuar o tratamento, evitando toxicidades e custos económicos innecesarios nos casos nos que os niveis de cfDNA tiveran sufrido cambios desfavorables. PATRICIA MONDELO MACÍA 64 Figura 10. Resumo gráfico do traballo realizado no CHAPTER II. Abreviaturas: NSCLC, do inglés non-small cell lung cancer, cancro de pulmón non microcítico; cfDNA, do inglés circulanting free DNA, ADN libre circulante; CTCs, do inglés circulating tumour cells, células tumorais circulantes; qPCR, do inglés quantitative PCR, PCR cuantitativa. Baseándonos nestes resultados, o seguinte paso centrouse en investigar se estudios de cfDNA e CTCs poderían proporcionarnos información sobre a evolución clínica de pacientes con SCLC. Neste tipo de cancro de pulmón menos frecuente, a falta de biomarcadores para a selección e o seguimento do tratamento xunto coas limitadas opcións terapéuticas que existen hoxe en día, asociase co mal prognóstico nestes pacientes. Por elo, é importante explorar o valor de novos marcadores para mellorar o manexo dos pacientes con SCLC. Con estas premisas, no capitulo terceiro da presente tese (CHAPTER III; Figura 11), realizouse un estudo cun enfoque similar ao realizado no capítulo II pero en pacientes con SCLC. Así, recrutáronse 46 pacientes con SCLC e 20 controles sans. recollendo un total de 131 mostras de sangue. Os niveis de cfDNA se cuantificaron lonxitudinalmente a diferentes puntos durante a primeira liña de terapia (quimioterapia ou inmunoterapia en combinación con quimioterapia) mediante un ensaio de qPCR. De maneira similar aos resultados obtidos nos pacientes con NSCLC, os niveis elevados de cfDNA antes do inicio da terapia mostraron asociación cunha PFS e unha OS máis Resumo in extenso 65 curta na cohorte de SCLC. Ademais, niveis altos de cfDNA ás 3 semanas e no momento da progresión da enfermidade se asociaron cun peor prognóstico, mostrando independencia respecto a outras variables clínicas. Doutra banda, nunha cohorte de 21 pacientes analizouse a presenza de CTCs antes do inicio do tratamento, empregando a tecnoloxía CellSearch®. Mentres en NSCLC a presenza de CTCs mostrou o seu valor prognóstico para predicir PFS e OS, na cohorte de SCLC só a presenza dun número elevado de CTCs (>150 CTCs) mostrou unha asociación discreta con menores taxas de PFS, pero sen mostrar o seu valor independente. Finalmente, e dados estes resultados, explorouse o valor prognóstico do análises do cfDNA xunto con outras variables clínicas como o estado do paciente segundo a escala ECOG PS (del inglés, eastern cooperative oncology group performance status) e o sexo. Este modelo permitiu estratificar a los pacientes en diferentes grupos de risco identificando aqueles que poderían beneficiarse más del tratamento con quimioterapia o inmunoterapia En resumo, no capítulo III descríbese resultados que avalan o interese do análises dos niveis de cfDNA como axuda aos oncólogos a realizar un prognóstico máis certeiro sobre a evolución da enfermidade e a resposta ao tratamento en pacientes recen diagnosticados con SCLC. En particular, con este estudio, demóstrase que a combinación do análises do cfDNA e características clínicas estándar (ECOG PS e sexo) poden axudar a estratificar mellor aos pacientes e detectar aqueles que poderían beneficiarse especialmente do tratamento. PATRICIA MONDELO MACÍA 66 Figura 11. Resumo gráfico do traballo realizado no CHAPTER III. Abreviaturas: SCLC, do inglés small cell lung cancer, cancro de pulmón microcítico; cfDNA, do inglés circulanting free DNA, ADN libre circulante; CTCs, do inglés circulating tumour cells, células tumorales circulantes; qPCR, do inglés quantitative PCR, PCR cuantitativa; ECOG PS, do inglés Eastern Cooperative Oncology Group Performance Status. Por último, no cuarto capítulo desta tese (CHAPTER IV; Figura 12), centrouse na identificación de marcadores proteicos presentes en plasma como potenciais biomarcadores preditivos de resposta a inmunoterapia en pacientes con NSCLC. Respecto ao cfDNA e ás CTCs, as proteínas circulante son tipicamente máis fáciles de medir, o que as converte nun biomarcador accesible e rendible para o uso nun contexto clínico. Ademais de proporcionar información sobre as células tumorais, o proteoma plasmático pode proporcionar información sobre o microambiente que rodea ao tumor, incluíndo a súa composición inmune e o estado inflamatorio do mesmo. Co obxectivo de identificar proteínas plasmáticas con interese para guiar o tratamento con inmunoterapia en pacientes con NSCLC avanzado, deseñouse un estudo no que se incluíron 64 pacientes tratados con pembrolizumab. As mostras de sangue foron recollidas en diferentes puntos do tratamento: antes de iniciar o tratamento, ás 6 e 12 semanas tras o inicio da terapia e no momento da progresión da enfermidade. En total, foron recollidas 171 mostras. O perfil Resumo in extenso 67 das proteínas presentes no plasma dos pacientes analizáronse empleando unha novedosa tecnoloxía baseada na adquisición independente de datos de numerosas proteínas: a tecnoloxía SWATH-MS (do inglés, sequential window acquisition of all theoretical fragment ion mass spectra). Un análises comparativo entre mostras de pacientes respondedores e non-respondedores mostraron permitiu identificar 324 proteínas diferencialmente expresadas entre ambos grupos. Destas, 172 proteínas presentaban niveis máis altos nos pacientes respondedores mentres que 152 proteínas presentaban niveis máis altos en non-respondedores. A partir do global de proteínas diferencialmente expresadas, identificouse unha firma proteómica de 7 proteínas (ATG9A, HPS5, DCDC2, PGTA, FIL1L, LZTL1 e SPTN2) con valor para discriminar aos pacientes que responderon ao tratamento con inmunoterapia. Individualmente, as proteínas identificadas mostraban un poder discriminatorio alto, con areas baixo a curva (AUC, do inglés area under the curve) entre 072 e 0.78, mellorando así o poder preditivo da expresión de PD- L1 en tecido (AUC= 0.638). O uso combinado das 7 proteínas mellorou estes valores, obtendo una AUC=1. Por outro lado, análises de supervivencia revelaron que niveis baixos de ATG9A asociáronse de maneira significativa cunha mellor PFS, mentres que valores altos de SPTN2 mostraron valor prognóstico positivo para PFS. Así mesmo, niveles baixos de HPS5 e DCDC2 asociáronse de maneira significativa cunha maior OS. Finalmente, a cuantificación destas 7 proteínas a diferentes puntos do tratamento, revelaron que o modelo proposto permite predicir a resposta á inmunoterapia tanto as 6 semanas coma as 12 semanas tras o inicio do tratamento con pembrolizumab. En resumo, os resultados derivados do capítulo IV desta tese, permitíronnos identificar unha firma proteica de 7 proteínas plasmáticas con potencial para PATRICIA MONDELO MACÍA 68 seleccionar aqueles pacientes que se beneficiarán da inmunoterapia baseada en anti-PD1. Figura 12. Resumo gráfico do traballo realizado no CHAPTER IV. Abreviaturas: NSCLC, do inglés non-small cell lung cancer, cancro de pulmón non microcítico. En resumo, o cancro de pulmón metastático é unha enfermidade complexa e heteroxénea que require dun enfoque multidisciplinario para o seu diagnóstico e tratamento. As "ómicas", incluíndo a xenómica e a proteómica, son ferramentas esenciais para identificar biomarcadores e dianas terapéuticas para abordar o tratamento do cancro. Ademais, o estudio de biopsias líquidas, incluíndo a análise de CTCs, cfDNA e outras moléculas presentes na sangue, xurdiu como unha ferramenta non invasiva e altamente sensíbel para a detección do cancro e o seguimento da resposta ao tratamento, onde cada compoñente proporciona información única sobre o tumor e o seu microambiente. A combinación destes marcadores circulantes coas diferentes "ómicas" e datos clínicos, como o evolutivo dos pacientes e as imaxes radiolóxicas de seguimento, ofrece un enfoque poderoso para comprender a complexidade do cancro de pulmón metastático e desenvolver estratexias de tratamento personalizadas. Resumo in extenso 69 Dadas as diferencias individuais e a heteroxeneidade molecular que caracteriza ao cancro de pulmón, as distintas estratexias de análise de biopsia líquida, aplicadas na presente tese, apórtannos unha visión máis integral da enfermidade. Centrando os nosos estudios nunha mostra non invasiva de sangue pódese comprender mellor a bioloxía do cancro de pulmón e sobre todo, deseñar ferramentas moleculares con potencial para guiar o tratamento e mellorar a esperanzado e a calidade de vida dos pacientes con cancro de pulmón avanzado. En concreto púidose demostrar que: a) Podemos xenotipar o estado de EGFR e MET en cfDNA mediante técnicas de qPCR e ddPCR de maneira robusta en tumores con alta carga tumoral; b) o análise do cfDNA representa unha técnica efectiva para prognosticar a evolución dos pacientes con NSCLC o SCLC avanzado e monitorizar a resposta a inmunoterapia nos primeiros.; c) pódemos caracterizar a expresión de proteínas clinicamente relevantes como c-MET ou PD-L1 en poboación de CTCs de pacientes con cancro de pulmón; d) o número de CTCS combinado cos niveis de cfDNA arroxa información prognóstica en pacientes con NSCLC; e) a caracterización do proteoma plasmático pódenos proporcionar biomarcadores con valor para identificar aos pacientes que van a presentar unha mellor resposta ao tratamento con inmunoterapia. Finalmente, é importante destacar que os nosos resultados permitiron estandarizar e validar estratexias de análises que representan actualmente unha ferramenta diagnóstica de utilidade real na práctica clínica do noso centro hospitalario pero tamén, identificar novos biomarcadores cuxo valor clínico deberá ser determinado en futuros estudios prospectivos. SUMMARY 79 INTRODUCTION Section 2.1 has been adapted/extracted from a published review entitled “Circulating Free DNA and Its Emerging Role in Autoimmune Diseases” (1). Section 2.1.1 has been adapted/extracted from a published book chapter entitled “Methods for the Detection of Circulating Biomarkers in Cancer Patients” (2). Section 2.2 has been adapted/extracted from a published book chapter entitled “Current status and future perspectives of liquid biopsy in small cell lung cancer. Biomedicines” (3). 81 INTRODUCTION 1. LUNG CANCER 1.1 Epidemiology and aetiology Lung cancer, with an estimated 1.8 million deaths per year worldwide, is the leading cancer-related cause of death among malignant solid tumours and the second most diagnosed cancer, behind female breast cancer, with more than 2.2 million new cases per year (4) (Figure 13). Incidence and mortality rates are around 2 times higher in men than in women and 3 to 4 times higher in industrialized countries than in transitioning countries (4). Figure 13. Distribution of cases and deaths for the top 3 most common cancers in 2020. Source: GLOBOCAN 2020. Cough, seen in 50-75% of patients, is the most common symptom, followed by haemoptysis, chest pain, and dyspnea (5). The main aetiological factor in all lung cancer types is tobacco consumption. Indeed, its carcinogenic effect on the lung has been described from 1950 to the 1960s (6,7). The World Health Organization (WHO) estimated that 71% of deaths by lung cancer are PATRICIA MONDELO MACÍA 82 caused by smoking. Although in the last years the reduction of smoking rates at global level has contributed to the reduced incidence of lung cancer and improved survival, in transitioning countries prevalence of tobacco consumption is increasing followed by an increase in new lung cancer deaths and diagnoses (8,9). Hence, the most effective intervention for reducing lung cancer mortality remains smoking cessation. However, tobacco is not the unique aetiological factor in lung cancer. One quarter of lung cancer cases worldwide occur in people who have never smoked (10–12) and they are associated with air pollution, poor diet and occupational exposures (13,14). Regarding the familial risk of lung cancer, it has been estimated the heritability at 18% (15), however many of the genetic component remains unidentified. 1.2 Lung cancer classification 1.2.1 Histological classification Histologically, lung cancer is a heterogeneous disease with wide-ranging clinicopathological features (16). Commonly, lung cancer can be divided into two major subtypes: Non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC), which accounts for 85% and 15% of total diagnoses, respectively (17). Both subtypes may share a similar cell of origin, the alveolar type II cell. NSCLC can be subdivided into four major subgroups: adenocarcinoma (ADC), squamous cell carcinoma (SCC), large cell carcinomas (LCC) and others. Oncogenic driver alterations are well reported in NSCLC and targeted therapies have allowed to reach better patients’ outcomes (17). In all subtypes, tobacco consumption remains to be the main cause, however, the association is stronger with SCC than with ADC, being Introduction 83 ADC the most common histology in never smokers (10), associated with environmental exposures that include second-hand smoking, pollution and inherited genetic susceptibility among others (12). In another hand, SCLC is a neuroendocrine tumour highly associated with tobacco consumption (18) and characterised by its rapid growth and aggressiveness. In contrast with NSCLC, the lack of specific somatic mutations and the poor progress of targeted therapies in SCLC makes the management of SCLC patients a real clinical challenge (19). 1.2.2 Clinical classification In both cases, the classification of lung tumours through optimal staging is essential for selecting the most appropriate treatment and management. Thus, stage refers to the extent of the cancer, and indicates how large the tumour is, and if it has spread at the moment of diagnosis. Nowadays, there are many staging systems, however the most clinically useful is the tumour, node and metastasis staging system (TNM), developed by the American Joint Committee on Cancer (AJCC) in collaboration with the Union for International Cancer Control (UICC) (20). The AJCC-TNM staging system has three principal components (Figure 14): PATRICIA MONDELO MACÍA 84 Figure 14. Definitions for T, N and M components in lung cancer tumours. • The T refers to the size and extent of the primary tumour. • The N refers to the number of nearby lymph nodes that present cancer cells. • The M refers to whether the cancer has metastasized and spread from the primary tumour to other parts of the body. Depending on the above terms defined, lung cancer patients can be classified into 4 principal stages: stage I, stage II, stage III and stage IV (Table 1) (21). In SCLC staging with the conventional TNM criteria is recommended, but according to the Veterans Administration of Lung Study Group (VALG) staging, SCLC is commonly classified into two stages: limited disease (LDSCLC), when it is confined to a hemithorax, where curative treatment with radio-chemotherapy is feasible (TNM stages I-III) and extensive disease (EDSCLC), defined as the presence of metastatic disease outside the hemithorax at first diagnosis (TNM stage IV) (22,23). Introduction 85 Table 1. Eight Edition of lung cancer stage grouping according the AJCC-TNM staging system (21) and the correspondence VALG staging for SCLC (23). T/M Label N0 N1 N2 N3 According VALG in SCLC T1 T1 IA IIB IIIA IIIB SCLC Limited Disease T2 T2a >3-4 cm IB IIB IIIA IIIB T2b >4-5 cm IIA IIB IIIA IIIB T3 T3 IIB IIIA IIIB IIIC T4 T4 IIIA IIIA IIIB IIIC M1 M1a-b IVA IVA IVA IVA SCLC Extensive Disease M1c IVB IVB IVB IVB Abbreviations: VALG, Veterans Administration of Lung Study Group. Unfortunately, most lung cancer patients are diagnosed at an advanced stage (stage IV; 57%), when surgery is not possible. Advanced lung cancer patients present a poor prognosis, when only 5.2% of patients survive 5 years after diagnosis (24). These data show the need for finding new strategies and biomarkers to improve the management of advanced lung cancer patients, in both subtypes, although the differences between them should be considered. 1.3 Advanced non-small cell lung cancer NSCLC represents 85% of total lung cancer diagnoses overworld. Approximately 55% of them are diagnosed at an advanced stages, with a 5- year survival of 6.1% (24). Its diagnosis requires a tissue biopsy for histological confirmation. Bronchoscopy or bronchoscopy combined with endobronchial ultrasound (EBUS) are the recommended techniques to obtain this tissue biopsy. In the case of peripheral lesions, transthoracic percutaneous fine needle aspiration and/or core biopsy, under imaging guidance is PATRICIA MONDELO MACÍA 86 proposed, however, there is a high risk of pneumothorax complication (17- 50%) (25). Histological diagnosis of NSCLC, as well as the molecular and genomic profile of the tumour, is crucial to make treatment decisions. Histologically, as described above, NSCLC can be divided into four major groups: ADC, SCC, LCC and others. ADC is the most common subtype, with about 50-60% of all NSCLC cases, followed by SCC (20-30%), both arising from alveolar cells located in the smaller airway epithelium. LCC and other types such as transitional cell carcinoma, sarcomatous carcinoma and other unclassified cancers, represent 10-20% of cases (26) (Figure 15). Figure 15. Molecular landscape of NSCLC. ADC is the most common subtype, with about 50-60% of all NSCLC cases, followed by SCC (20-30%). Most prevalent targetable mutations in ADC includes mutations in EGFR, ALK, RET, ROS1 and MET among others. Adapted from Wang et al. (26). Immunohistochemistry (IHC) is recommended to obtain a correct histological diagnosis of lung cancer. In addition to morphology, Introduction 87 characterization with specific antibodies is employed to confirm the different subtypes. Thus, ADC subtype is associated with thyroid transcription factor 1 (TTF1) positivity and napsin A, while p40 positivity with probable SCC. If neither are positive, the diagnosis remains NSCLC-not otherwise specified (NSCLC-NOS) (25). LCC are typically poorly differentiated and characterized by large cells with abundant cytoplasm and large nucleoli. Molecularly, NSCLC is a heterogeneous disease composed of subpopulations of cells or clones with distinct molecular features. Several alterations in targetable oncogenic pathways in this subtype have been reported in the last decades, allowing a better treatment selection. The common alterations found in NSCLC are also dependent on the histology. Thus, the most frequently mutated genes in ADC include KRAS and EGFR, and the tumour suppressor genes TP53, KEAP1, and NF1, among others. Contrary, common mutated genes in SCC include the tumour suppressor TP53 (90% of cases) and CDKN2A. Unlike ADC, actionable mutations in receptor tyrosine kinase (RTKs) are rarely detected in SCC. For example, only a small subset of SCC cases present mutations in EGFR (27) (Figure 15). 1.3.1 Current therapies applied in advanced NSCLC Over the past decade, treatment options for NSCLC have changed dramatically (27). Importantly, these advances in lung cancer treatment and improvements in understanding the disease in the last years allowed a reduction in its incidence and mortality (28). However, the survival of patients with this type of tumour remains the poorest of all tumour types (29). Briefly, the different approaches for NSCLC management include targeted therapies, PATRICIA MONDELO MACÍA 88 immune checkpoint inhibitors (ICIs), chemotherapy and radiotherapy (Figure 16). Figure 16. Different mechanisms of anticancer therapies, including targeted therapies, ICIs, chemotherapy, and radiotherapy. Abbreviations: MHC, major histocompatibility complex; TKI, tyrosine kinase inhibitor; TCR: T-cell receptor. Targeted therapies The molecular testing of EGFR, BRAF, ALK and ROS1 genes in all NSCLC patients at diagnosis is mandatory. If these genes present any alteration, targeted therapy with tyrosine kinase inhibitors (TKIs) in the first line is recommended (25). TKIs inhibit the activation of kinase signalling pathways in cancer cells, and therefore, induce apoptosis. EGFR mutations are the most common targetable driver mutations in lung cancer. Thus, first and second-generation EGFR-TKIs have significantly improved progressionfree survival (PFS) compared with platinum-doublet chemotherapy in patients Introduction 95 heterogeneity of PD-L1 expression (67). In addition, there are several challenges associated with the technique itself, such as the selection of the correct antibody, among others (68). Recently, microsatellite instability (MSI) has been associated with a higher likelihood of response to immunotherapy in cancer patients (69). Nevertheless, due to the small subset of NSCLC patients who present a MSI-high status (70), MSI testing is not routinely performed in clinical practice, and the value of MSI as an immunotherapy biomarker in lung cancer is unclear. Thus, a significant gap remains in the identification of reliable and predictive response biomarkers that can be utilized in a clinical setting to effectively stratify patients for immunotherapy. In another hand, routine laboratory test including haematology, renal and hepatic function and blood biochemistry are required into the management of NSCLC patients under therapy, however the use of serum markers, such as carcinoembryonic antigen (CEA) is not recommended (71). Finally, to guide therapy and evaluate treatment response, clinical guidelines recommended the use of imaging tools, such as computerized tomography (CT) scan. Thus, first response evaluation is recommended after two to three cycles of therapy and performed every 6-9 weeks during course of disease. Measurement of lesions should follow Response Evaluation Criteria in Solid Tumours (RECIST) v1.1 (25,72). 1.3.3 Challenges and potential biomarkers in advanced NSCLC Above, the different biomarkers employed into clinic routine to select the best therapy and monitor advanced NSCLC patients’ evolution were summarized. Nevertheless, they are limited and, in some cases (for example, the use of PD-L1 determination to select immunotherapy candidates), their PATRICIA MONDELO MACÍA 96 efficacy is low. Therefore, in the last years and with the purpose to find new and better biomarkers into the management of advanced NSCLC, potential biomarkers are being under investigation. Actually, several studies have reported the analysis of the tumour mutational burden (TMB) (73) as useful biomarkers to select patients who benefit from immunotherapy. Indeed, several clinical trials have shown a high TMB associated with an improved ORR and PFS in NSCLC treated with ICIs (74). However, more research is needed to confirm its utility as a reliable biomarker. The tumour-infiltrating lymphocytes (TILs) (75), the neutrophil- to-lymphocyte ratio (NLR) (76) and lactate dehydrogenase (LDH) also could be employed as useful biomarkers to select patients who will benefit from immunotherapy (66,77). Particularly, pre-treatment NLR together LDH has been reported as a potential prognostic marker, but prospective validations are needed (78). In this context, liquid biopsy emerges as a potential tool to search new biomarkers, both at diagnosis and during the disease evolution (see expanded information in section 2). 1.4 Advanced small cell lung cancer SCLC is a high-grade neuroendocrine carcinoma and the most aggressive form of lung cancer. SCLC accounts for approximately 15% of total lung cancer diagnoses worldwide and up to 25% of lung cancer deaths (79). This subtype is highly related to tobacco smoking and approximately, 75% of patients are diagnosed at advanced stages, with a 5-year survival of 2.8% (24). Introduction 97 For pathological diagnosis, histology is preferred over cytology. IHC for analysing neuroendocrine markers, such as synaptophysin, chromogranin or CD56 is commonly used, however, SCLC can be negative or focally positive for only one of these markers. TTF1 is expressed in 90% of cases and 50% of nuclei are positive for Ki-67. Negativity for p40 characterizes this tumour subtype, in contrast with p40 positivity in SCC (80). Regarding their molecular features, large studies of genome-wide analyses have confirmed loss-of-functions mutations of TP53 and RB1 genes in almost all SCLC cases (81–84). Moreover, frequent inactivation of Notch family members was also observed (81). Due to all these alterations are loss- of-function, the ability to target oncogenic mutations therapeutically is limited in SCLC, in contrast with NSCLC. Importantly, recent advances have made it possible to deepen the molecular knowledge of SCLC, which has allowed to distinguish between four different subtypes characterized by distinct gene expression profiles. These subtypes are the neuroendocrine subtypes SCLC-A (ASCL1-positive) and SCLC-N (NEUROD1-positive), and the non-neuroendocrine subtypes SCLC-P (POU2F3-positive) and SCLC-Y (YAP1-positive) (85), that could offer new therapeutic options in the close future. 1.4.1 Current therapies applied in advanced SCLC This section has been adapted/extracted from a published review (3). Systemic treatment of advanced SCLC patients in first line is based on the combination of platinum and etoposide-based chemotherapy, with median survival ranging from 6 to 12 months (86). Commonly, chemotherapy regimens are administered together radiotherapy, in curative as well as in PATRICIA MONDELO MACÍA 98 palliative therapy. The use of prophylactic cranial irradiation (PCI) and thoracic radiotherapy has improved outcomes in patients who have responded to first-line treatment (87), being their use incorporated into SCLC treatment guidelines (80). Recently, the use of immunotherapy has been also incorporated in patients with metastatic disease (88–91). In first-line, the blockade of the PD- 1 receptor or its ligand PD-L1 has been shown to improve the prognosis of patients with ED-SCLC (92,93). The Impower 133 (NCT02763579) and CASPIAN phase III trials (NCT03043872) demonstrated that adding atezolizumab or durvalumab to platinum plus etoposide chemotherapy improves OS and PFS compared to chemotherapy alone (94,95). More recently, the KEYNOTE-604 (NCT03066778) study reported that combining pembrolizumab with platinum and etoposide improves PFS but does not significantly improve OS in patients with ED-SCLC (96) (Figure 20). Figure 20. FDA-approved therapies for the treatment of ED-SCLC in 1st line. Despite the therapeutic efforts and a good initial therapy response, unfortunately the majority of SCLC patients progress within the first 6 months of first line therapy completion (80). Until now, the choice of the second-line treatment is limited to topotecan or cyclophosphamide, doxorubicin and Introduction 99 vincristine (CAV), with poor response rates, typically less than 10% (97). Nowadays, more options can be considered, such as Lurbinectedin, which has been approved by the FDA for salvage treatment of SCLC that has relapsed from first-line chemotherapy (98). On the other hand, nivolumab and pembrolizumab have also been approved by the FDA for refractory SCLC based on the results of the CheckMate-032 (NCT01928394) and Keynote-158 (NCT02628067) clinical trials, respectively (99,100). However, in the Checkmate-331 trial (NCT02481830), nivolumab failed to demonstrate an improvement in OS versus standard treatment in patients with relapsed or refractory SCLC treated with a platinum-based line of chemotherapy (101). Regarding other options and in contrast with NSCLC, in SCLC there are no clear driver genes or kinase targets and, consequently, no approved targeted therapies are being employed to treat these patients (102). Existing drugs have failed to demonstrate an improvement in patients’ outcomes (103) and the identification of new therapeutic targets in SCLC has been challenging, partly because driver mutations are primarily loss-of-function (genes RB1 and TP53) or currently difficult to target (104,105). Nevertheless, new therapies such as abemaciclib (a CDK4/6 inhibitor), veliparib (an oral PARP inhibitor) (106), SLFN11 (107,108), Rovalpituzumab tesirine (Rova-T) (109) and alisertinib (106,110) are being under investigation. 1.4.2 Current biomarkers in advanced SCLC and future challenges Nowadays, there are still no validated biomarkers that can be used for treatment decisions in the context of SCLC. Besides, in contrast with NSCLC, PD-L1 expression and testing is not recommended in routine clinical practice in this tumour (80). Regarding disease monitorization, the actual therapeutic PATRICIA MONDELO MACÍA 100 approach is similar than in NSCLC. Therapy response is evaluated using imaging tools such as CT scan every 2-3 months (80). Like in NSCLC field, some biomarkers are being under investigation to improve SCLC management in advanced patients and new therapeutic approaches have opened new horizons and hopes for improving SCLC outcomes. Indeed, NLR have shown its potential utility as predictive biomarker in advanced SCLC patients (111), however, more research is still required to reach a better understanding of the molecular mechanisms behind SCLC biology and its response to the different therapies to advance in the identification of new predictive biomarkers to apply the most convenient therapeutic strategy for each patient. Notably, the information obtained from the analysis of liquid biopsy components would be crucial for this objective. 2. LIQUID BIOPSY The term liquid biopsy was born in the past years as a new strategy to improve the management of different pathologies or physiologic states such as cancer (112–114), liver diseases (115) and prenatal analyses (116), among others. The analysis of liquid biopsy allows to evaluate different molecules obtained from body fluids and to monitor their changes over the time, which enable to follow the disease evolution (113). Nowadays, blood is the most common fluid employed, although, in the last years, other fluids such as saliva (117) or urine (118) are being employed. In the oncology field, the interest in liquid biopsy has increased in the last decades, providing new opportunities in the management of different stages of the disease: early detection of the disease or tumour recurrence, Introduction 101 individual risk assessment and the treatment monitoring (119). Also, cancer screening is considered another potential application (120). Hence, liquid biopsy has emerged as an alternative to the tumour tissuebased analysis, the current gold standard for the management of patients with cancer. Nowadays, the genetic profile of tumours is obtained from surgical or biopsies specimens, but this procedure cannot be always successfully performed. Indeed, in lung cancer, biopsy samples are often of poor quality or quantity, justifying the need to explore new tools for the molecular characterization (121,122). A tissue biopsy involves several undesirable effects such as potential surgical complications and clinical risk to the patient. Besides, the solid biopsy cannot be obtained repeatedly during the therapy. Of note, the non-invasive nature of liquid biopsy offers the opportunity to obtain samples in a non-invasive way at any moment and, therefore, enables to track the genomic evolution of the tumour in real-time. Other advantage of liquid biopsy is that it can offer a complete image of the tumour biology of different lesions, while the solid biopsies only capture a part of them (113). In addition, metastases remain the principal cause of cancer-related deaths (123) and they can develop different genomic characteristics absent or poorly represented in the biopsy of the primary tumour. Large studies of primary and metastatic tumours (124,125) have shown the relevance of intra-tumoral and intertumoral heterogeneity and how this characteristic contributes to treatment failure in patients with cancer (126,127). Thus, liquid biopsy can capture the genetic landscape of both lesions (primary and metastatic) since the tumoral material present in body-fluids can come from the different tumoral lesions (Figure 21). PATRICIA MONDELO MACÍA 102 Figure 21 . Tissue biopsy versus liquid biopsy: comparison of the advantages and limitations (Mondelo-Macía et al., 2021. Biomedicines). The most common circulating elements present in liquid biopsies are circulating free DNA (cfDNA) and circulating tumour cells (CTCs). cfDNA can be released into the blood by normal and tumour cells. The fraction released from tumour cells constitutes the circulating tumour DNA (ctDNA). Similarly, CTCs can be released to the blood from the primary or metastatic tumour locations. In the past years, the clinical interest of both liquid biopsy components is reflected by an increase in their analysis in several clinical trials. Currently, the database ClinicalTrials.gov (www.clinicaltrials.gov: accessed on 20th February 2023) lists more than 300 active clinical trials that analyse CTCs and more than 100 trials interrogating cfDNA/ctDNA levels in different cancer diseases. In addition, technological advances in the detection and characterization of liquid biopsy components have enable the introduction Introduction 103 of liquid biopsy into the clinic routine to manage some tumour types. Thus, the FDA has already approved several liquid biopsy assays (based on blood or on urine analyses) for their use to help oncologists in the patients’ managing (1,128). Due to their clinical interest the characteristics of CTCs and cfDNA are explained in more detail within the next sections. 2.1 Circulating free DNA The next section has been adapted/extracted from a recently published review (1). CfDNA was first reported in healthy individuals by Mandel and Métais in 1948 (129). However, it was not until 1966 when the discovery of high values of cfDNA in patients with systemic lupus erythaematosus (130) showed the potential of cfDNA as a biomarker for autoimmune diseases. Ten years later, Leon et al. characterized cfDNA for the first time in the field of oncology and reported higher levels in cancer patients than in healthy individuals, suggesting its potential as a diagnostic marker and to characterize tumours in a non-invasive and dynamic way (131). In 1994, cfDNA was recognized as an important tool to detect several mutations in the blood of patients with myeloid disorders (132) and pancreatic adenocarcinomas (133) and currently represents a key element for precision oncology (134). CfDNA consists of a heterogeneous and complex DNA fraction present in free body fluids associated with extracellular vesicles (EVs) or as part of macromolecular complexes such as nucleosomes (135). The size of cfDNA is highly variable depending on the mechanisms involved in its fragmentation, with a normal peak of 166 bp fragments, which corresponds to the length of the DNA bound to a nucleosome (136). Due to this high fragmentation, PATRICIA MONDELO MACÍA 104 cfDNA origin is mainly associated with cell death mechanisms; however, to date, the origin of cfDNA remains partially unclear, and different mechanisms have been suggested in several studies (137–139). Here, we summarize the principal origins and sources of cfDNA described until now (Figure 22): Figure 22. Mechanisms involved in the release of circulating free DNA. (Mondelo-Macía et al., 2021. JPM). Secretion. Most cfDNA release into the circulation is associated with active secretion in EVs, such as exosomes, microparticles, or apoptotic bodies. This cfDNA is protected from nucleases and can be released into circulation through the breakdown of EVs. Some studies have reported that over 90% of cfDNA is associated with this type of release (140). Apoptosis. Apoptosis, also known as programmed cell death, is an essential process to maintain cellular homeostasis. This process allows the removal of damaged cells by caspase activation. When the caspase pathway is activated, the cell starts to suffer morphological and biochemical changes that will result in cell and nuclear retraction, lipid redistribution, and DNA Introduction 111 Nowadays, as it is well known, CTCs are tumour cells originating from the primary or metastatic sites that can enter the circulation and disseminate to distant sites. Peripheral blood offers the possibility to analyse the presence of this circulating tumour population for cancer diagnosis and disease monitoring. The importance of CTCs for diagnostic and prognostic purposes has been well reported in different cancer types, such as metastatic breast (171,172), prostate (173), NSCLC (174) and colorectal cancer (175). Interestingly, in SCLC several studies have described higher CTC levels in comparison to other cancer types (176), supporting the interest of this circulating population as an accessible tumour biopsy. However, the proportion of CTCs in the bloodstream is very low, with approximately 1 CTC per 106–107 leukocytes (177), being a challenge their detection and isolation. CTCs have a very short half-life (1–2.4 hours) (112), and the majority of them are rapidly cleared; however, some of them evade recognition by immune cells, survive into the bloodstream and can reach distant locations to generate metastasis (178–180) into secondary organs. These cells are characterized by a hybrid phenotype in terms of epithelial and mesenchymal markers that favour their survival in circulation (179–182). 2.2.1 Technologies for CTCs analyses The low proportion of CTCs in the bloodstream together with the molecular heterogeneity that characterizes these cells is the principal challenge for CTCs isolation and detection. Several CTCs detection platforms have been developed in the past decades to obtain an enriched sample of CTCs. All technologies isolate these cells focusing on differential features between CTCs and blood cells, such as protein expression, morphology, PATRICIA MONDELO MACÍA 112 volume and biophysical properties, presenting different advantages and limitations. These technologies can be categorized based on the method of isolation as antigen-dependent (affinity-based) or antigen-independent (183) (Figure 24). Antigen-dependent Antigen-dependent isolation approaches are the most common methods employed and they are based on the presence of specific surface markers in CTCs (called positive enrichment) or by blood cells (negative enrichment). A. Positive enrichment, the most employed strategy is usually carried out using antibodies that recognize epithelial cell adhesion molecule (EpCAM) (184) conjugated with magnetic nanoparticles. Among the current EpCAM-based technologies, CellSearch® system (Menarini, Silicon Biosystem, Bologna, Italy) (185) has become the “gold standard” for the CTC- detection methods. CellSearch® system employs anti-EpCAM-coated ferrofluid nanoparticles for the selection of EpCAM positive cells. Next, an immunostaining step discriminates CTCs from leukocytes based on the positive expression of cytokeratins and the absence of CD45 staining together with morphologic criteria. A high number of alternatives that employ magnetic nanoparticles conjugated with anti-EpCAM antibodies are available (186). Recently, new positive enrichment methods are being developed, in which the specific surface markers are immobilized on the surface of microfluidic chips (186) to increase the contact between the cells and, therefore, to enhance capture efficiency. However, the isolation in all these approaches is based on the EpCAM expression, therefore, they are not able to Introduction 113 detect CTCs without EpCAM expression, for example, CTCs of non-epithelial tumours such as sarcomas or CTCs that have undergone epithelial-to- mesenchymal transition (EMT) (187,188). B. Negative enrichment methods employ magnetic nanoparticles conjugated with antibodies against the well-established leukocyte antigen CD45 (189) or other antigens expressed in blood cells and represent a good alternative to avoid the limitations of the EpCAM-dependent isolation. They allow isolating CTCs independently of any membrane marker expression, however due to the low proportion of CTCs and the recent observation that they travel into the bloodstream coated with blood cells (190), the resulting recovery rate is often relatively low (191). Antigen-independent Antigen-independent methods are based on the physical properties of CTCs such as density, electric charges (DEP, dielectrophoresis), size and deformability, among others. The principal advantage and difference with the antigen-dependent methods are that they do not require specific surface markers on CTCs, so they also allow the isolation of CTCs with a low epithelial phenotype. Density-based methods were the first techniques developed. These methods allow the processing of high volumes of blood (about 25 mL) in a short time, however, they generally show a low efficiency and purity of the sample obtained (186). The size-based methods are the most common. They are based on the fact that tumour cells are larger than blood cells (191,192) and, therefore, they can be isolated using filter-based strategies (such as ISET assay (Rarecells Diagnostics, Paris, France) (193), microfluidic chips (such as Parsortix system (Angle, UK))(194) and methods based on PATRICIA MONDELO MACÍA 114 centrifugal forces (195). The different charges between blood cells and CTCs can also be employed in their isolation. DEP field forces are employed to move CTCs independently to other blood cells, and represent a highly specific approach for isolation (196). Antigen-independent methods are generally easy to implement, however they depend on the availability of advanced materials or assistive engineering technologies for better clinical application (20). Interestingly, new methods combining antigen-based capture with the advantages of microfluidics strategies, such as CTC-iChip, are being developed for increasing the isolation efficacy (197). However, nowadays the development of a robust and standardized platform to capture CTCs for clinical application remains a challenge. Finally, it’s important to remark that small volumes processed with the methods here described may be a serious limitation for the detection of these rare events, especially in cancer patients without metastases, in which the number of CTCs is expected to be very low. To solve this problem, some “in vivo” approaches such as GILUPI Nanodetector® (198) or the Diagnostic leukapheresis (DLA) can be employed (199). Introduction 115 Figure 24. Different strategies for CTCs enrichment and detection. Abbreviations: DEP, dielectrophoresis. After enrichment, the CTCs fraction usually still contains a substantial number of leukocytes (184). This background of leukocytes is seen in all CTCs enrichment platforms being the posterior molecular analyses of CTCs a challenge. Therefore, after the first enrichment of CTCs there are some platforms that allow the isolation of pure CTCs at the single level by the use of micromanipulation or via dielectron force manipulation, such as the DEPArray system (Menarini, Silicon Biosystem, Bologna, Italy), among other strategies (200). Analyses of individual CTCs/clusters at the DNA, RNA or protein level provide valuable information about the molecular heterogeneity of these cells and a more precise characterization of the disease (201). In addition, surface proteins, which can be key targets for personalized therapies, can be analyzed by immunofluorescence (IF). Thus, certain protein expression e.g EpCAM e.g CD45 PATRICIA MONDELO MACÍA 116 in CTCs has been studied, such as ER and HER2 in breast cancer, among others (202). 2.2.2 Clinical relevance During the last decades, several platforms have reported the feasibility to detect CTCs and monitor changes during the course of treatment in patients with several metastatic carcinomas (197,203,204). Moreover, the presence of CTCs before the treatment in the blood of patients with cancer has shown prognostic significance and their levels after the treatment can be predictive of response to the therapy. Nowadays, the clinical value of CTC presence remains controversial because CTCs number is very variable between different tumour types (205), and only 2 platforms were approved by the FDA.. CellSearch® system was the first FDA-approved platform for CTCs isolation and enumeration for clinical use in metastatic breast, prostate, and colorectal cancer. First, it was approved in 2004 for use in a clinical setting to predict outcomes for metastatic breast cancer patients (206). Although in the posterior years, CellSearch® system was also granted FDA approval to aid in monitoring metastatic colorectal (175) and prostate cancer patients (207), its real use in clinical routine is scarce. Recently, in May 2022, Parsortix system was approved for clinical use in metastatic breast cancer patients (208). 2.3 Other circulating biomarkers CTCs and cfDNA are not the unique biomarkers employed to improvement the management of cancer patients. Thus, there are other markers in the field of liquid biopsy with great potential for future clinical applications. One example are EVs which are nanosized particles with membranes released by any cell type, including cancer cells. The most Introduction 117 common subtypes of EVs are exosomes, with a size between 50 and 150 nm. Exosomes are very abundant, express membrane proteins on their surface and contain several particles, such as proteins, nucleic acids, and lipids (209). Thus, exosomes offer the possibility to analyse the expression of proteins of interest on their surface, such as PD-L1. Actually, EVs have shown to be valuable tools as biomarkers for longitudinal monitoring, defining tumour type, stage, progression and treatment response (210,211). Other elements, such as circulating microRNA (miRNA), circulating RNA, platelets, and circulating proteins, are at the early stages of investigation (212). In recent years the potential of tumour-educated blood platelets as a non-invasive tumour biomarker has been demonstrated (213,214). Platelets are involved in the progression and spread of various solid cancers, and their RNA molecular signatures can provide specific information about the presence, location, and molecular characteristics of the tumours (215). Regarding circulating proteins, few studies have investigated their possible value as prognostic or predictive biomarker in lung cancer, however, global protein changes may provide an independent biomarker that reflects the tumour evolution (216). OBJECTIVES Chapter I.A 127 EGFR mutations: simple and rapid blood-based qPCR test to select and guide treatment of advanced NSCLC patients. “Rapid Idylla™ mutational testing to detect EGFR mutations in plasma samples and to monitor therapy in advanced NSCLC patients (217)” Patricia Mondelo-Macía1,2,3, Ramón Manuel Lago-Lestón1, Aitor Rodríguez-Casanova2,3,4,5, Alicia Abalo1, Ángel Díaz-Lagares3,4,8, Jorge García-González6,7,8, Luis León-Mateos2,3,6,7,8, Roberto Díaz-Peña3,9*, Laura Muinelo-Romay1,3,8*. 1 Liquid Biopsy Analysis Unit, Translational Medical Oncology (Oncomet), Health Research Institute of Santiago (IDIS), Santiago de Compostela, Spain. 2 Universidade de Santiago de Compostela (USC), Santiago de Compostela, Spain. 3 Galician Precision Oncology Research Group (ONCOGAL), Medicine and Dentistry School, Universidade de Santiago de Compostela (USC), Santiago de Compostela, Spain. 4 Epigenomics Unit, Cancer Epigenomics, Translational Medical Oncology (Oncomet), Health Research Institute of Santiago (IDIS), Santiago de Compostela, Spain. 5 Roche-CHUS Joint Unit, Translational Medical Oncology Group (Oncomet), Health Research Institute of Santiago (IDIS), 15706 Santiago de Compostela, Spain. 6 Department of Medical Oncology, Complexo Hospitalario Universitario de Santiago de Compostela (SERGAS), Santiago de Compostela, Spain. 7 Translational Medical Oncology (Oncomet), Health Research Institute of Santiago (IDIS), Santiago de Compostela, Spain. 8 Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), Madrid, Santiago de Compostela, Spain. 9 Fundación Pública Galega de Medicina Xenómica, SERGAS; Grupo de Medicina Xenómica-USC, Health Research Institute of Santiago (IDIS), Santiago de Compostela, Spain. * Corresponding. Chapter I.A 129 EGFR mutations: Simple and rapid blood-based qPCR test to select and guide treatment of advanced NSCLC patients. ABSTRACT During the last few years, detection of EGFR-activating mutations has become a routine part of clinical practice for NSCLC in order to select the optimal treatment strategy. The use of complemental techniques that allow to analyse EGFR-activating mutations in blood samples, would help improve the management of NSCLC patients. In this study, we investigated the feasibility to use Idylla™ ctEGFR mutation assay to detect EGFR alterations in 40 NSCLC patients before and during the treatment. The concordance between the blood-based test and tissue-based test was investigated. Furthermore, comparison with different liquid biopsy-based approaches (BEAMing technology and NGS method) was performed. Our results suggest that IdyllaTM ctEGFR mutation assay is a fast and optimal tool for therapy selection in advanced NSCLC and also a good approach to monitor the therapy response in patients under TKIs therapies. Keywords: Idylla™, NSCLC, liquid biopsy, EGFR mutations. Chapter I.A 131 1. INTRODUCTION NSCLC represents 85% of total lung cancers, with about 2 million new cases per year (4). Target therapies are the first choice in NSCLC patients with specific mutations profile, such as EGFR mutations, present in 17% of metastatic NSCLC patients (218). Thus, EGFR TKIs have become the standard therapy for patients with EGFR activating mutations, leading to a longer survival (17). To select patients who present EGFR mutations, and offer appropriate EGFR TKI treatment, mutation testing of solid tumour samples is required. However, tissue/cytologic samples are not always available or evaluable. CfDNA analyses represent a good alternative strategy to assess the EGFR status though the use of like NGS or digital PCR based technologies. Nevertheless, these methods require technical expertise and long turn-around time. Idylla™ ctEGFR mutation assay (Biocartis NV, Mechelen, Belgium) is a fully integrated real-time PCR-based test with a rapid and easy protocol that allow an easy implementation of liquid biopsy analyses into the clinical routine. The aim of the present study was to interrogate the accuracy and performance of the Idylla™ solution as a non-invasive test to detect EGFR mutations in plasma samples of advanced NSCLC patients. In this context, we think that the Idylla™ ctEGFR mutation assay could be a good alternative tool to analyse the molecular characteristics of NSCLC patients. With this purpose we evaluated the concordance of EGFR status between cfDNA using Idylla™ and tissue samples. In addition, the concordance among different liquid biopsy-based approaches BEAMing (Sysmex Inostics), AVENIO (Roche Diagnostics) and Idylla™ was investigated to evaluate the performance of Idylla™ solution versus the other PATRICIA MONDELO MACÍA 132 approaches. In addition, we determine the feasibility to use the ctEGFR assay to monitor NSCLC patients during the treatment. 2. MATERIAL AND METHODS 2.1 Synthetic DNAs We calculated the limit of detection (LOD) of Idylla™ solution using four different commercial cfDNA standards that covers 10 EGFR variants with specific variant allele frequencies (VAF) of 5%, 1%, 0.1% and 0% (EGFR wild type) (Cat nº. HD825, Horizon Discovery Ltd., Cambridge, United Kingdom). The commercial cfDNA standards were obtained from human cell lines with EGFR mutations fragmented to 160 bp to mimics cfDNA extracted from human plasma. All standards were quantified using the Qubit dsDNA HS assay kit and Qubit 3.0 Fluorometer instrument (Thermo Fisher Scientific, Waltham, USA). Seven of the total 10 EGFR alterations presents in the reference cfDNA are included in the panel analysed by the Idylla™ ctEGFR Mutation Assay (5 single nucleotide variants (SNVs) including L861Q, L858R, S768I, T790M and G719S; one exon 20 insertion (V769-D770insASV) and one exon 19 deletion (Del15)). The LOD was defined for each of these seven alterations as the lowest mutant allele frequency yielding a positive result. 2.2 Study design Fifty-three blood samples from 40 advanced NSCLC patients treated between January 2018 and March 2022 at the Medical Oncology Service of Complexo Hospitalario Universitario de Santiago de Compostela were retrospectively included in the study. The study was performed in accordance Chapter I.A 133 with the Declaration of Helsinki (as revised in 2013) and all individuals signed informed consent forms approved by Santiago de Compostela and Lugo Ethics Committee (Ref: 2017/538) prior to enrolling in the study. 2.3 Blood collection and sample processing Peripheral blood was obtained by direct venepuncture using CellSave (Menarini, Silicon Biosystems, Bologna, Italy), Streck Cell-Free DNA BCT (Streck Inc, Omaha, NE, USA) or EDTA tubes (219). Plasma and cellular components were separated by two centrifugation steps: one centrifugation at 1,600 g for 10 minutes at room temperature and a second time at 5,500 g for 10 minutes at room temperature to remove any remaining cellular debris. Plasma samples were aliquoted and storage at -80 ºC until posterior analyses. In addition, when high volume of plasma was available, cfDNA was isolated using the QIAamp Circulating Nucleic Acid Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions and quantified by the fluorometric instrument Qubit 4 using the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, USA). 2.4 Idylla™ ctEGFR mutation assay Idylla™ ctEGFR mutation assay (Biocartis NV, Mechelen, Belgium) is a real-time PCR assay that allows the qualitative detection of 49 mutations of the EGFR gene using plasma samples: Detects 4 SNVs of exon 18 (G719A/C/S), 2 SNVs of exon 20 (T790M and S768I) and 4 SNVs of exon 21 (L858R and L861Q); 34 different exon 19 deletions: and 5 exon 20 insertions. Commercial cfDNA standards and a total of 53 samples from patients with NSCLC were analysed using the Idylla™ assay. For the analyses, 2 ml of plasma together with 200 µl of proteinase K (20 mg/ml, PATRICIA MONDELO MACÍA 134 Qiagen, Hilden, Germany) were directly pipetted in the cartridge. The Idylla™ system performs the cfDNA extraction, a real time PCR amplification and detection of EGFR mutations. Finally, Idylla™ console software analyses PCR amplification curves determining the cycle of quantification (Cq) in each sample. To know that sample is suitable in terms of quality and quantity, the assay used a Sample Processing Control (SPC) signal, that corresponded to the amplification of an EGFR wild-type control. Based on the difference between the EGFR-mutant Cq and the SPC (denominated ΔCq), an automatic report with the result is obtained. The total time to obtain a result using Idylla™ ctEGFR mutation assay is less than 3 hours. 2.5 Cobas® EGFR Mutation Test Tissue EGFR status was determined in the primary tumour obtained at diagnosis using Cobas® EGFR Mutation Test v2 (Roche, Basel, CH) in 25 patients, as manufacturer’s guidelines. The test is a real-time PCR-based test that identifies 42 mutations in exons 19, 19, 20 and 21 of the EGFR gene using a tissue biopsy sample. Briefly, formalin-fixed, paraffin-embedded tissue (FFPE) specimens are processed using the Cobas® DNA Sample Preparation Kit (Roche, Basel, CH). After preparation, amplification and detection of EGFR mutations are carried out using real-time PCR test. The total time to obtain a result using Cobas® EGFR mutation test in tumour sample is approximately 8 hours. 2.6 BEAMing technology In parallel, BEAMing method was performed with the OncoBEAM EGFR kit (Sysmex Inostics, Hamburg, Germany) using 2 mL of plasma in 17 samples. cfDNA used for the analyses was isolated employing the QIAamp Chapter I.A 135 Circulating Nucleic Acid Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Once cfDNA was extracted, the assay performs a hybridization step between DNA-coated beads and a sequence specific fluorescein-labelled probe. Then, amplificated samples are analysed by flow cytometry. Finally, results are reported as “no mutation detected” or “mutation detected” (220). The total time to obtain a result using BEAMing technology is approximately 72 hours. 2.7 Next-generation sequencing analyses Next-generation sequencing (NGS) analyses was performed in 15 samples with the AVENIO ctDNA Expanded panel (221) (Roche, Basel, CH), which covers alterations in 77 genes including all coding regions of EGFR. cfDNA used for the analyses was isolated from 4 mL of plasma, following the AVENIO ctDNA Expanded panel (Roche, Basel, CH) protocol. Libraries were sequenced on a NextSeq 500 (Illumina, San Diego, CA) which provided high-depth sequencing with a mean coverage of ~9000x (range: 7621x - 10402x). 2.8 Statistical analyses Statistical analyses were performed using R version 4.1.1. The Kappa test was used to determine the concordance between EGFR mutations based on tissue and plasma analyses and among the three different based-blood platforms (IdyllaTM, BEAMing and AVENIO panel). GraphPad Prism 8 and Venny 2.1.0 (https://bioinfogp.cnb.csic.es/tools/venny/) (222) were employed for the graphical representations. PATRICIA MONDELO MACÍA 136 3. RESULTS 3.1 LOD determination using commercial standards To explore in house the performance of the assay we first explored the technique sensitivity. With this aim, we calculated the LOD of Idylla™ ctEGFR mutation test using commercial cfDNA standards with different VAFs that covers different EGFR variants. Thus, LOD was determined as 5% VAF for L861Q and G719S; 1% for L858R, the insertion of exon 20 and T790M; and 0.1% VAF for Del15 of exon 19 and the point mutation S768I using a total of 160 ng of commercial cfDNA (Table 1.1). Table 1.1. Sensitivity of Idylla™ system using commercial cfDNA (160 ng). EGFR exon Variant AA mutation LOD (%) Exon 18 G719S p.Gly719Ser 5 Exon 19 Deletion 15 p.Glu746_Ala750del 0.1 Exon 20 Insertion exon20 p.Val769_Asp770insAlaSerVal 1 S768I p.Ser768Ile 0.1 T790M p.Thr790Met 1 Exon 21 L858R p.Leu858Arg 1 L861Q p.Leu861Gln 5 Abbreviations: LOD, limit of detection; AA: amino acid. 3.2 Patients’ characteristics Forty NSCLC patients were included in the study. Their clinical and pathological characteristics are presented in detail in Table 1.2. The median age was 63.5 (range 40-81) with 52.5% females and 37.5% of patients were never smokers. Most patients had tumours with adenocarcinoma histology Chapter I.A 143 Figure 1.4. Monitoring EGFR status in NSCLC patients using the ctEGFR IdyllaTM mutation assay. Swimmer’ plot on monitored patients (n=10). Specially in one case (ID13), we observed that analysis using Idylla™ solution allows to detect the progression disease (Figure 1.5). A 62-year-old male was diagnosed in March of 2017 with NSCLC adenocarcinoma stage IIIA. Tissue sample analyses revealed the presence of L861Q mutation in EGFR gene. In May, image analyses reported lesions into the SNC. The patient started chemotherapy and radiotherapy and showed a good response. In March of 2018, the patient had a progression and he started gefitinib treatment (day 0). Before the treatment onset, a blood sample was collected and cfDNA analyses were performed by BEAMing technology and Idylla™ solution. Both results reported a negative result for the mutation EGFR L861Q, probably due to location of the metastatic lesions and the consequent circulating tumour DNA (ctDNA) low levels. Longitudinal analyses during gefitinib treatment (day 49, 103 and day 201) in cfDNA reported the absence of L861Q mutation in plasma samples until day 355, when the patient presented progression disease at different levels by image analyses. At ID15 ID16 ID22 ID13 ID25 ID10 ID11 ID19 ID29 ID41 30 20 10 0 PATRICIA MONDELO MACÍA 144 progression, an increase in cfDNA levels and the presence of the mutation L861Q using Idylla™ solution were determined, in accordance with the disease evolution. Figure 1.5. Evolution of cfDNA levels and EGFR L861Q mutations in patient ID13 during Gefitinib treatment. The L86Q1 mutation appeared at progression disease, which confirmed by image analyses. Abbreviations: SDi: stable disease; PD: progression disease. Other cases maintained the mutation positivity at baseline and progression disease and/or showed increased ctDNA levels such as happened in patients ID19 and ID11 (Figure 1.6A). Patient ID19 and ID11 were followed up during different TKIs treatment and in both cases all determinations were positive for the mutation L858R. Patient ID19 progressed at 12 weeks and an increase in L858R mutation and cfDNA levels was found at this time, supporting the interest of using IdyllaTM solution to detect progression disease. Patient ID11 showed stable disease at 10 weeks by image analyses and a VAF reduction of L858R mutation was found using Idylla™ solution despite of an increase in cfDNA levels, suggesting an active response PDSDi Chapter I.A 145 to gefitinib but probably a progression due to a different tumour clone. One month later, patient showed a clinical deterioration and died. In this type of cases, imaging tests may be of limited value, while liquid biopsy could provide useful information on the disease evolution. A. B. PATRICIA MONDELO MACÍA 146 Figure 1.6 . Evolution of cfDNA levels and EGFR L858R mutation in patient ID19 (A) and patient ID11 (B) during different TKIs treatment. Abbreviations: PD: progression disease; SDi: stable disease. 4. DISCUSSION Regarding the interest of improving the options to interrogate EGFR status in cfDNA, it is important to have in mind that tumour analyses have several limitations in NSCLC patients such as access difficulty (mainly to biopsy metastasis), the impossibility to characterize the disease at real time and the lack of representativity of the tumour heterogeneity. To solve these limitations, in the last years, liquid biopsy has emerged as a clear alternative and one of the main pillars for personalized oncology, reporting promising results mainly focused on the cfDNA analyses (205). In the present study we explored the feasibility of determine EGFR status in cfDNA by Idylla™ solution in NSCLC patients, previously reported in tissue (223,224) and plasma samples (225) and its potential value to monitor the therapy response in NSCLC patients looking for a fast technique not make the tumour genotyping . Of note, a good concordance rate between analyses in tissue and plasma cfDNA with the Idylla™ BRAF Mutation Test has been previously described (226). In our study, a moderate agreement between tissue and plasma results was found. This result can be partially explained due to the time between blood and tissue samples collection in the discordant patients. As higher time between samples collection as bigger probability to find a molecular evolution of the tumour. In another hand, although we reported a low concordance rate of exon 19 deletions in contrast with the high concordance of the point mutation L858R, the small size of our cohort could not allow us to arise solid conclusions regarding this aspect. However, it’s Chapter I.A 147 important to highlight that 2/7 discordant samples presented a VAF lower than 0.13% by NGS or BEAMing assay and in 5/7 discordant samples, the time between plasma sample collection and tissue biopsy was higher than 5 months. Therefore, the lower ctDNA content and the tumour evolution could explain this lower concordance. Furthermore, we investigated the concordance rates among Idylla™ solution and other reference technologies, BEAMing and NGS, and found a good overall agreement determining EGFR status. In line with our study in lung cancer, the automatized system demonstrated a good diagnostic performance for KRAS analyses in patients with colorectal cancer , reaching a high concordance with BEAMing (227). Besides, studies in tissue samples reported the good concordance between Idylla™ assay and NGS panels (224). Of note, we also reported a good correlation of Idylla™ solution and AVENIO panel employing plasma samples but with a lower time-to-result with Idylla that can be crucial for metastatic patients (228). In another hand, it’s important to remark that Idylla™ solution allowed us to analyse the mutational profile of NSCLC patients in longitudinal plasma samples during treatment without the need to perform further tissue biopsies. This advantage is important in whose patients that undergone treatment resistance. For example, the mutation T790M appears as a mechanism of resistance during TKIs therapy. Its detection is crucial to select whose patients that should be treated with a third generation of TKIs (17). In our study, although we only detected the T790M mutation in two plasma samples, our analysis with commercial cfDNA showed that Idylla™ system allows to detect the TKIs resistance mutation with a LOD of 1%. Preliminary studies have reported a moderate T790M detection rate of 66.6-80.0% in plasma samples using Idylla™ solution (225,229), according to our study. In addition, PATRICIA MONDELO MACÍA 148 in our work paired plasma samples analysis showed that T790M mutations with a variant allele frequency lower than 0.06% using BEAMing/NGS cannot be detected by Idylla™ system. However, analyses with a larger cohort of patients harbouring the resistance mutation T790M are needed to validate the clinical value of the kit for specifically detect T790M mutations. Overall, the results obtained in the in this pilot study with a real-world cohort of patients allow us to conclude that Idylla™ ctEGFR Mutation Assay could be employed as a first screen to detect EGFR mutations fast and on demand. If no EGFR mutation is detected due to the low ctDNA content in plasma, another sensitive technique such as a ddPCR-based strategy should be employed to confirm the negative result. To this regard, some aspects should be considered for the selection of an optimal platform or analytic circuit to analyse EGFR alterations in ctDNA. For example, tumours with multiple metastatic locations and highly vascularized will have higher levels of ctDNA and therefore the genotyping of clinically relevant mutation will be easier to detect (150). Regarding the value as a monitoring tool, our preliminary results showed that changes in the mutational profile of EGFR samples using Idylla™ solution, could provide valuable information to monitor the therapy response in patients under TKIs therapies. Chapter I.A CHAPTER I.B Development of a blood-assay to detect MET alterations using circulating free DNA and circulating tumour cells in NSCLC patients. Chapter I.B 151 Development of a blood assay to detect MET alterations using circulating DNA and circulating tumour cells in NSCLC patients. This chapter has been adapted from a recently published article entitled “Detection of MET Alterations Using Cell Free DNA and Circulating Tumor Cells from Cancer Patients” (230). Patricia Mondelo-Macía1#, Carmela Rodríguez-López2,3#, Laura Valiña4,5, Santiago Aguín2,3, Luis León-Mateos2,3, Jorge García-González2,3,6, Alicia Abalo1, Óscar Rapado-González1,7, Mercedes Suárez-Cunqueiro3,6,7, Angel Díaz-Lagares6,8, Teresa Curiel3, Silvia Calabuig-Fariñas6,9,10,11, Aitor Azkárate5,12, Antònia Obrador-Hevia5,13, Ihab Abdulkader14, Laura Muinelo-Romay1,6*, Roberto Diaz-Peña1,15* and Rafael López-López1,2,6. 1 Liquid Biopsy Analysis Unit, Translational Medical Oncology (Oncomet), Health Research Institute of Santiago (IDIS), 15706 Santiago de Compostela, Spain. 2 Department of Medical Oncology, Complexo Hospitalario Universitario de Santiago de Compostela (SERGAS), 15706 Santiago de Compostela, Spain. 3 Translational Medical Oncology (Oncomet), Health Research Institute of Santiago (IDIS), 15706 Santiago de Compostela, Spain. 4 Department of Laboratory Medicine, Hospital Universitari Son Espases, 07120 Palma, Balearic Islands, Spain. 5 Group of Advanced Therapies and Biomarkers in Clinical Oncology, Institut d’Investigació Sanitària de les Illes Balears (IdISBa), 07120 Palma, Balearic Islands, Spain. 6 Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), 28029 Madrid, Spain. 7 Department of Surgery and Medical Surgical Specialties, Medicine and Dentistry School, Universidade de Santiago de Compostela, 15705 Santiago de Compostela, Spain. 8 Cancer Epigenomics, Oncomet, Health Research Institute of Santiago (IDIS), Complexo Hospital Univeritario de Santiago de Compostela (SERGAS), 15706 Santiago de Compostela, Spain. 9 Molecular Oncology Laboratory, Fundación Hospital General Universitario de Valencia, 46014 Valencia, Spain. 10 Department of Pathology, Universitat de València, 46010 València, Spain. 11 TRIAL Mixed Unit, Centro de Investigación Príncipe Felipe-Fundación para la Investigación del Hospital General Universitario de València, 46012 València, Spain. 12 Medical Oncology Department, Hospital Universitari Son Espases, 07120 Palma, Balearic Islands, Spain. 13 Molecular Diagnosis Unit, Hospital Universitari Son Espases, 07120 Palma, Balearic Islands, Spain 14 Department of Pathology, Complexo Hospital Universitario de Santiago de Compostela (SERGAS), Universidade de Santiago de Compostela, 15706 Santiago de Compostela, Spain. 15 Faculty of Health Sciences, Universidad Autónoma de Chile, Talca 3460000, Chile * Corresponding. # These authors have contributed equally to this work. Chapter III 255 Table 3.5. Univariate and multivariate Cox regression analyses of cfDNA levels, CTC counts and clinical parameters. Variable Univariate Multivariate p -value HR (95% CI) p -value HR (95% CI) PFS Baseline log cfDNA (high vs. low cfDNA) 0.001 5.06 (1.89-13.6) 0.005 46.0 (3.16-672) Baseline CTC count, CellSearch® (≥150 vs. <150) 0.028 3.47 (1.14-10.6) ECOG (≥2 vs. <2) <0.001 3.57 (1.72-7.40) 0.04 17.9 (1.11-289) Sex (male vs. female) 0.2 1.76 (0.74-4.22) Age (continue) 0.5 1.01 (0.97-1.05) Stage (IV vs. III) 0.07 3.00 (0.91-9.91) Number of metastasis (>2 vs. ≤2) 0.1 1.70 (0.90-3.19) Liver metastasis (yes vs. no) 0.08 1.74 (0.93-3.23) Smoking (smoker vs. former smoker) 0.8 0.92 (0.49-1.70) 3 weeks log cfDNA (high vs. low cfDNA)* <0.0001 3.50 (1.69-7.23) 0.004 3.49 (1.50-8.12) OS Baseline log cfDNA (high vs. low cfDNA) 0.003 3.32 (1.50-7.37) 0.004 32.4 (3.05-344) Baseline CTC count, CellSearch® (≥150 vs. <150) 0.07 2.71 (0.93-7.88) ECOG (≥2 vs. <2) <0.001 4.54 (2.13-9.68) Sex (male vs. female) 0.2 1.80 (0.75-4.36) Age (continue) 0.5 1.01 (0.97-1.06) Stage (IV vs. III) 0.08 2.86 (0.86-9.47) Number of metastasis (>2 vs. ≤2) 0.5 1.23 (0.65-2.33) Liver metastasis (yes vs. no) 0.3 1.45 (0.77-2.76) Smoking (smoker vs. former smoker) 0.5 0.82 (0.43-1.55) 3 weeks log cfDNA (high vs. low cfDNA)* <0.001 3.67 (1.72-7.82) 0.002 4.35 (1.68-11.3) PD log cfDNA (high vs. low cfDNA)* <0.001 15.2 (3.28-70.7) 0.001 19.5 (3.30-115) *Multivariate Cox regression model including sex, age, Eastern Cooperative Oncology Group Performance Score, stage, number of metastases, presence of liver metastasis and smoking status. The levels of cfDNA were determined as low (<cut-off) or high (≥cut- PATRICIA MONDELO MACÍA 256 off) based on the cut-off obtained from the ROC curve analyses. Abbreviations; HR, hazard ratio; CI, confidence interval. Regarding the cfDNA at 3 weeks, multivariate regression analyses also confirm its independence as a prognostic biomarker of PFS and OS (hazard ratio, 3.49.0; 95% CI, 1.50-8.12; p-value =0.004 and hazard ratio, 4.35; 95% CI, 1.68-11.3; p-value =0.002, respectively) (Table 3.5). In contrast, multivariate analysis in CTCs did not show value as an independent predictive biomarker of PFS and OS. Next, an independent prognostic model for both PFS and OS was developed. Three variables were retained in the final prognostic model: cfDNA levels (high vs. low levels), ECOG PS (<2 vs. ≥2) and sex (male vs. female). The detailed results of the multivariate analyses are shown in Figure 3.12. Figure 3.12. Final multivariate Cox regression prognostic model for (A) PFS and (B) OS. Abbreviations: HR, hazard ratio; CI, confidence interval. Subsequently, we segregated patients into three risk categories: patients with all adverse prognostic factors were classified in the poor-risk category (high cfDNA levels, ECOG PS≥2 and male sex), patients with two adverse Chapter III 257 prognostic factors were classified in the intermediate-risk category, and patients with one or no adverse prognostic factor were classified in the favorable-risk category. The Kaplan–Meier curves representing the three risk categories and median PFS and OS are presented in Figure 3.13. Median PFS ranged from 124 to 289 days based on the number of adverse prognostic factors present before therapy. Median OS ranged from 115 to 514 days. Figure 3.13. Kaplan-Meier survival analysis according to risk-group for PFS (A) and OS (B). Abbreviations: NA, not applicable. Our prognostic model indicates an increased risk of 5-times to present disease recurrence (HR= 5.37, 95% CI, 2.32-12,4; p-value =3x10-5) and 6- PATRICIA MONDELO MACÍA 258 times the risk of death (HR= 6.02, 95% CI, 2.66-13.6; p-value =3x10-6) in the unfavorable category (intermediate and poor groups) than in the favorable group (Table 3.6). Table 3.6. PFS and OS probabilities estimated according two risk groups. Risk Groups PFS OS HR 95% CI p-value HR 95% CI p-value Favourable 1 (Ref.) - 1 (Ref.) - Intermediate-poor 5.37 2.32-12.4 < 0.001 6.02 2.66-13.6 < 0.001 Abbreviations: HR, hazard ratio; CI, confidence interval. 4. DISCUSSION Precision medicine has the objective of optimizing the selection of the best therapy for each patient. In this context, liquid biopsy has born as a promising and minimally invasive tool for this due to its ability to provide a total image of primary and metastatic tumours at different times across therapy (344). Recently, the management of SCLC has changed and new therapies, such as immunotherapy, have been investigated and approved for clinical use (3,345,346). Nevertheless, the necessity to find a prognostic biomarker for helping to select the therapy prescribed and to monitor the evolution of the disease during the treatment, remains a challenge in SCLC patients. In this study, we report for the first time the possibility to employ the cfDNA and its quantification as a prognostic biomarker in SCLC prior to starting therapy and at different time points. Our analyses allowed us to identify a group of lowrisk patients characterized by low cfDNA levels at baseline who probably will benefit from both: chemotherapy in monotherapy or the combination of Chapter III 259 immunotherapy and chemotherapy. The study of another common circulating biomarker, CTCs, also provided us information for the prognostic of patients before starting therapy, although the results were less clear. We found a significant association between presence of a high number of CTCs (≥150 CTCs) and worse PFS. Total cfDNA refers to a heterogeneous and complex DNA fraction free released in body fluids by any cell type (not only tumoral) through several cell death mechanisms such as secretion, apoptosis and necrosis (138,150). Of note, the source of cfDNA is an intriguing question in cancer. According to previous studies, the fraction of ctDNA varies from 0.1–89% of cfDNA but it increases it accordance with the tumour burden. Therefore, although cfDNA content is not tumour specific, it can be assumed that total cfDNA in cancer patients originates from tumour cells, cells in the tumour microenvironment or from cells involved in the antitumour response (137). Therefore, cfDNA can be uses as a surrogate of ctDNA, but always taking into account that they are different entities that can provide us different information. In line with our results, it’s well reported that cancer patients present higher cfDNA levels than healthy controls (347,348), but few studies have investigated the possible prognostic and predictive value of total cfDNA quantification in patients with SCLC. In contrast, the ctDNA, the tumour-derived fraction of this cfDNA, has been reported as a prognostic and predictive biomarker in several works (349– 353). Almodovar et al. reported that changes in the mutant allele frequencies on ctDNA were associated with response to treatment and relapse. Twentyseven patients with SCLC were analysed by next-generation sequencing (NGS) custom panel, however, the lack of driver mutations known in SCLC, limited the number of genes analysed (349). In another work, Devarakonda et PATRICIA MONDELO MACÍA 260 al. analysed 564 patients using a larger NGS panel, including 73 genes, and, according to previous results, RB1 and TP53 were the most frequent mutant, however, no prognostic or predictive value was reported in this study (354). In this way, total cfDNA quantification allows to detect the total DNA released from normal and also tumour cells into the blood. Thus, despite the few known driver mutations found in SCLC, cfDNA quantification allows to quantify the total levels before treatment and monitor the changes during therapy. Recently, our group demonstrated the feasibility to quantify cfDNA levels in NSCLC patients and its association with patients’ outcomes (261), suggesting its possible utility in SCLC. Thus, in the present work, we quantified total cfDNA levels using two different technologies, a fluorometric assay, Qubit and a more specific one, the qPCR assay analysing the hTERT gene. CfDNA quantification by both technologies showed good concordance. Furthermore, the concordance of cfDNA levels at any time point of therapy using both methods also showed a good concordance. Therefore, both methods could be used to robustly measure the cfDNA content. However, to complete our study we chose the qPCR assay, which is a high sensitive and specific assay for cfDNA quantification in SCLC patients, and was previously employed in studies focused on NSCLC (261,270,273,288). Regarding the clinical meaning of cfDNA content, we found that high levels were significantly associated with shorter PFS and OS before therapy onset, being a robust independent prognostic biomarker in newly diagnosed SCLC patients. Also, cfDNA levels at baseline were higher in patients with stage IV that could be a consequence of a more aggressive disease. This can Chapter III 261 be partially explained by an increase of ctDNA levels released from the tumour cells to the bloodstream, increasing the total cfDNA fraction. Moreover, analyses showed that cfDNA levels at 3 weeks are associated with patients’ outcomes, being those patients with high values at 3 weeks, the ones with the worst prognosis. In addition, high cfDNA levels at the time of disease recurrence were associated with a higher risk of death. Of note, multivariate analyses showed the independence of cfDNA levels at 3 weeks and at progression disease as a prognostic biomarker. These results suggest that cfDNA monitoring could provide valuable information for the management of SCLC as our group previously reported in NSCLC (261). Thus, in clinical practice, in those SCLC patients with high levels of cfDNA at the time of disease recurrence, the selection of a more aggressive therapy or the intensification of clinical visits would be considered. Besides cfDNA levels, we investigated the prognostic value of additional biomarkers such as CTCs and clinical characteristics. CTCs were analysed in a cohort of 21 SCLC patients using the CellSearch® system. According to the literature (3), a detection rate of 85.71% was found in our study. Moreover, the CTC count at baseline determined using the CellSearch® system was significantly associated with PFS and OS (334,337,338,355,356). For example, Naito et al. reported that the presence of ≥8 CTCs/7.5mL of blood was associated with worse OS (334), however, another study employed 50 CTCs as cut-off for PFS and OS (338). In fact, a consensus regarding the optimal cut-off of CTCs and the prognostic value remains a challenge (3). In this work, we found a discrete association between the presence of ≥150 CTCs and a shorter PFS, however multivariate analyses did not show independent value for the CTC count. Interestingly, high cfDNA levels and the presence PATRICIA MONDELO MACÍA 262 of CTCs at baseline were significantly associated, reporting the clear association between both circulating biomarkers and disease burden. CTCs release in the bloodstream is related to the intravasation process of potentially metastatic cancer cells. Nevertheless, cfDNA is released by any cell type including tumoral and normal cells, however, how cfDNA release relates to tumour biology is currently unknown. In another hand, we evaluate several factors that could influence the patients’ outcomes. Thus, we proposed a simple model to segregate patients into three categories based on risk of progression and death (considering the cfDNA levels, ECOG PS and sex of patients). We found that patients with one or less adverse prognostic factors were classified in the favourable-risk category and present a longer PFS and OS. Regarding the clinical relevance of these results, some limitations in our design should be considered. First, the sample size of our CTC cohort was relatively small and CTC monitoring during therapy could provide more valuable information. Second, our patient cohort are not homogenous regarding treatment regimen, 71.74% SCLC patients were treated with while 28.26% SCLC patients were treated with chemo-immunotherapy. In addition, a validation study of our prognostic model in a larger independent cohort is needed. In conclusion, we describe an important potential role of cfDNA levels as a prognostic biomarker in newly diagnosed SCLC patients and also as a tool that could provide useful information about disease evolution. Finally, a prognostic model based on cfDNA levels, and some clinical characteristics (ECOG PS and sex) would allow us to stratify patients and detect those who could particularly benefit from the treatment. CHAPTER IV Circulating proteins as predictive biomarkers for immunotherapy in metastatic NSCLC patients. Chapter IV 271 method that combines deep proteome coverage capabilities and allow to identify and quantify circulating proteins with consistency and accuracy (367). In the present study, we hypothesized that plasma proteins can serve as a prognostic and predictive biomarker in NSCLC patients under first-line immunotherapy. We performed a differential proteomic quantitative analysis based on SWATH–MS technology to analyse the proteome in blood samples collected from patients with advanced NSCLC prior to the start therapy and during the treatment. Finally, we reported a proteomic signature of 7 proteins that could predict the response to immunotherapy in advanced NSCLC patients before starting therapy, as well as at 6 and 12 weeks after starting immunotherapy. 2. MATERIAL AND METHODS 2.1 Patients and blood sample collection In total, 64 newly diagnosed advanced NSCLC patients treated with pembrolizumab therapy at the Department of Medical Oncology of Complexo Hospitalario Universitario de Santiago de Compostela, were included in the present study. A first cohort (Discovery cohort) included a total of 48 patients with newly diagnosed advanced NSCLC. A posterior validation cohort enrolled 16 new diagnosed advanced NSCLC patients. The study was performed in accordance with the Declaration of Helsinki (as revised in 2013) and all individuals signed informed consent forms approved by Santiago de Compostela and Lugo Ethics Committee (Ref: 2017/538) prior to enrolling in the study and they could withdraw their consent at any time. Blood samples of 64 patients were collected before the therapy onset. In addition, 107 PATRICIA MONDELO MACÍA 272 longitudinal blood samples were collected at 6 weeks (n=46) and 12 weeks (n=36) after starting the therapy and at time of progression disease (n=25). Best response to pembrolizumab treatment was based on RECIST1.1, according to the following criteria: Complete response (CR), partial response (PR), stable disease (SDi) or progressive disease (PD). Responders were defined as the proportion of patients with CR, PR, or SDi. 2.2 Blood sample processing Ten mL of peripheral blood were obtained by direct venipuncture in CellSave tubes (Menarini, Silicon Biosystems, Bologna, Italy) and processed within 96 hours after blood collection for the discovery cohort. In the case of the validation cohort, peripheral blood was obtained in EDTA tubes and processed within 4 hours. In both cases, plasma and cellular components were separated by centrifugation at 1,600 g for 10 minutes at room temperature. A second centrifugation was performed at 5,500 g for 10 minutes at room temperature to remove any remaining cellular debris. Plasma samples were aliquoted for storage at -80 ºC until use. 2.3 Plasma preparation for mass spectrometry (MS) analysis Plasma abundant proteins depletion Plasma samples were fractionated prior proteolytic digestion to enrich the spectral library employed in SWATH-MS quantification. For this purpose, aliquots of each sample (30 μl) were depleted with dithiothreitol (DTT). Fresh DTT (500 mM) was mixed with the 30 μl of human plasma samples and vortex briefly (368,369). Then, samples were incubated until observe a viscous white precipitate that persisted for 60 minutes, and they were centrifugated at 18,840 g for 20 minutes. Supernatants were transferred to a clean tube. Chapter IV 273 Isolation and proteins digestion To make global protein identification, an equal amount of protein from all samples were loaded on a 10% SDS-PAGE gel. The run was stopped as soon as the front had penetrated 3 mm into the resolving gel (370,371). The protein band was detected by Sypro-Ruby fluorescent staining (Lonza, Switzerland), excised, and processed for in-gel manual tryptic digestion, as described elsewhere (20). Proteins were reduced with DTT 10mM in Ambic 40mM and alkylated with 55 mM iodoacetamide in 50 mM ammonium bicarbonate. Then, the gel pieces were rinsed with 50 mM ammonium bicarbonate in 50% methanol dehydrated by addition of acetonitrile and dried in a SpeedVac. Modified porcine was added to the dry gel pieces at a final concentration of 20 ng/μl in 20mM ammonium bicarbonate, incubating them at 37 °C for 16 hours. Peptides were extracted by carrying out three 20 minutes incubations in 40 μL of 60% acetonitrile dissolved in 0.5% HCOOH. The resulting peptide extracts were pooled, concentrated in a SpeedVac, and stored at −20 °C. 2.4 Protein quantification by SWATH-MS (Sequential Window Acquisition of all Theoretical Mass Spectra) Creation of the spectral library - Data-dependent acquisition (DDA) To construct the MS/MS spectral libraries, the peptide solutions from the discovery cohort were analysed by a shotgun data-dependent acquisition (DDA) approach by micro-liquid chromatography-MS/MS. To get a good representation of the peptides and proteins present in all samples, pooled vials of samples from each group (basal n=48, 6 weeks n=35, 12 weeks n=30 and progression disease, n=21) were prepared using equal mixtures of the original samples for the spectral library building. Four μL (1 µg/μL) of the pool was PATRICIA MONDELO MACÍA 274 separated using Reverse Phase Chromatography. Gradient was created using a micro liquid chromatography system (micro-LC) Ekspert nLC425 (Eksigent Technologies nanoLC 400, Sciex, CA, USA) coupled to high-speed Triple TOF 6600 mass spectrometer (Sciex, CA, USA) with a micro flow source. The chosen analytical column was a silica-based reversed phase column Chrom XP C18 150 × 0.30 mm, 3 mm particle size and 120 Å pore size (Eksigent, Sciex, CA, USA). The trap column was a YMC-TRIART C18 (YMC Technologies, Teknokroma) with a 3 mm particle size and 120 Å pore size, switched on-line with the analytical column. The loading pump delivered a solution of 0.1% formic acid in water at 10 µl/min. The micro-pump generated a flow-rate of 5 µl/min and was operated under gradient elution conditions. Water and acetonitrile, both containing 0.1% formic acid, were used as solvents A and B, respectively. The gradient run consisted of 5% to 95% B for 30 min, 5 min at 90% B and finally 5 min at 5% B for column equilibration, for a total run time of 40 min. When the peptides eluted, they were directly ionized and injected into a hybrid quadrupole-TOF mass spectrometer Triple TOF 6600 (Sciex, CA, USA) operated with a Data dependent acquisition (DDA) system in positive ion mode. A Micro source (Sciex, CA, USA) was used for the interface between micro-LC and MS, with an application of 2600 V voltage. The acquisition mode consisted of a 250 ms survey MS scan from 400 to 1250 m/z followed by an MS/MS scan from 100 to 1500 m/z (25 ms acquisition time) of the top 65 precursor ions from the survey scan, for a total cycle time of 2.8 seconds. The fragmented precursors were then added to a dynamic exclusion list for 15 seconds; any singly charged ions were excluded from the MS/MS analysis. The instrument was automatically calibrated every 4 hours using Chapter IV 275 external calibrant tryptic peptides from PepCalMix solution (AB SCIEX, Framingham, MA, United States). The peptide and protein identifications were performed using Protein Pilot software (version 5.0.2, Sciex, CA, USA) searched against a Human specific Uniprot database (https://www.uniprot.org/), specifying iodoacetamide as Cys alkylation and Trypsin as enzyme used in digestion. The software uses the algorithm ParagonTM for database search and ProgroupTM for data grouping. The false discovery rate (FDR) was set to 1% for both peptides and proteins, using a non-lineal fitting method (372). The MS/MS spectra of the identified peptides were then used to generate the spectral library for SWATH peak extraction using the plug-in MS/MSALL with SWATH Acquisition MicroApp (version 2.0, Sciex) for PeakView Software (version 2.2, Sciex, CA, USA). Peptides with a confidence score above 99% (as obtained from Protein Pilot database search) were included in the spectral library. In addition, a spectral online library called, Human Pan-Human library, which contained 12,046 proteins (https://db.systemsbiology.net/sbeams/cgi/PeptideAtlas/GetDIALibs), was also employed in order to improve and expand the coverage of the identified NSCLC cancer plasma proteome in our library. The sum of the two libraries constitutes our final library, denominated NSClibrary. Relative quantification by SWATH acquisition - DIA SWATH-MS acquisition was performed on a TripleTOF® 6600 LCMS/MS system (AB Sciex, CA, USA). Samples were analysed using a DIA method (n=171 samples). Each sample (4 μL) was analysed using the LC-MS PATRICIA MONDELO MACÍA 276 equipment and LC gradient described above for building the spectral library but instead using the SWATH-MS acquisition method. The method consisted of repeating a cycle that is composed of the acquisition of 100 TOF MS/MS scans (400 to 1500 m/z, high sensitivity mode, 50 ms acquisition time) of overlapping sequential precursor isolation windows of variable width (1 m/z overlap) covering the 400 to 1250 m/z mass range with a previous TOF MS scan (400 to 1500 m/z, 50 ms acquisition time) for each cycle. Total cycle time was 6.3 seconds. For the sample set, the width of the 100 variable windows was optimized according to the ion density found in the DDA runs using a SWATH variable window calculator worksheet from Sciex. SWATH quantification was attempted for all proteins in the ion library that were identified by ProteinPilot with an FDR below 1%. The DIA raw data were firstly converted into mzML format filtered by SWATH Acquisition MicroApp (version 2.0) then analyzed by PeakView (version 2.2), against our spectral NSCLibrary. PeakView computed an FDR and a score for each assigned peptide according to the chromatographic and spectra components; only peptides with an FDR below 1% were used for protein quantitation. The retention time (RT) adjustment were performed using the own peptides of the proteins presents in the library from the peptides that were selected for each protein. Them the RT of all peptides were realigned in each run according to the iRT peptides spiked of the selected peptides in each sample and eluted along the whole time axis. The extracted ion chromatograms were then generated for each selected fragment ion; the peak areas for the peptides were obtained by summing the peak areas from the corresponding fragment ions. Protein quantitation was calculated by adding the peak areas of the corresponding peptides Up to ten peptides per protein Chapter IV 277 and seven fragments per peptide were selected, based on signal intensity; any shared and modified peptides were excluded from the processing. Five-minute windows and 30 ppm widths were used to extract the ion chromatograms. Them this integrated peak areas (processed. mrkvw files from PeakView) were directly exported to the MarkerView 1.3.1 software (AB Sciex, CA, USA) for relative quantitative analysis (373–376). The export data will generate three files containing quantitative information about individual ions, the summed intensity of different ions for a particular peptide and the summed intensity of different peptides for a particular protein. 2.5 Quality control and statistical analysis Next, a quality control was performed. Proteins with over 30% missing values in the sample set were filtered out. Next, missing values of the remaining proteins were imputed using the RandomForest R package (377,378) and protein values were normalized by quantile normalization. Finally, statistical analyses were performed. Continues data were compared using t test for independent samples and categorical variables were compared using the Fisher's exact test. Swimmer plot was provided to visualize every patient's therapy response and the time of survival from the diagnoses. Kaplan-Meier method was used to plot the survival curves applying the log rank test. Student’s t test was applied to identify differentially expressed proteins (DEPs), with a p-value <0.05 and p-value <0.01. To identify the functions and relevant pathways of the DEPs, we performed gene ontology (GO) using Metascape (379). The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway of each group was generated and visualize with proteomaps, using a web tool based on the t-test difference values without PATRICIA MONDELO MACÍA 278 log2 transformation (380). A final predictive model to separate responder group to non-responder group was developed. Comparisons of Cox proportional hazard regression models were made using the AIC technique (341), with a smaller AIC value indicating the better model. A stepwise backward elimination procedure was performed to minimize the AIC. ROC curves were computed based on protein levels, representing the AUC values, and computing the CI at 95% confidence levels. ROC curves were also constructed to evaluate the thresholds of baseline proteins levels for survival analyses. All statistical analyses were performed using GraphPad Prism version 8.0, IBM® SPSS® statistics version 25.0 and R version 4.1.1. The following R packages were used: survival (277), survminer, ggplot2 (342), pROC (276), gtsummary (343), swimplot, RandomForest, MASS and stats. 3. RESULTS 3.1 Patient characteristics In total, 64 newly diagnosed NSCLC patients undergoing first-line pembrolizumab treatment in monotherapy (n=40) or in combination with chemotherapy (n=24) were included in the study (Figure 4.1, Table 4.1). Patients were enrolled in two different sub-cohorts: a discovery cohort with 48 NSCLC patients and a validation cohort with 16 NSCLC patients. The validation cohort was included to evaluate a possible impact of the blood collection tubes. In global terms, the mean age was 64.7 years (range: 45–80), most of patients were males (75%), current or former smokers (87.5%) and had Chapter IV 279 tumours with adenocarcinoma histology (81.25%) (Table 4.1). The ECOG PS of <1 accounted for 17.2% of patients. Table 4.1. Patients’ demographics and clinical characteristics at baseline. Clinical Characteristics Discovery cohort (n=48) Validation cohort (n=16) R (n=30) N (%) NR (n=18) N (%) pvalue R (n=10) N (%) NR (n=6) N (%) pvalue Mean age ± SD, range 64.3 ± 8.2, 50-80 61.9 ± 9.5, 45-78 NS 71.5 ± 5.4, 60-77 63.2 ± 7.6, 51-74 NS Sex Female 6 (20.0) 5 (27.8) NS 1 (9.1) 4 (66.7) 0.03 Male 24 (80.0) 13 (72.2) 9 (90.9) 2 (33.3) Smoking Smoker / Former smoker 25 (83.3) 16 (88.9) NS 9 (90) 6 (100) Never 5 (16.7) 2 (11.1) 1 (10) 0 (0.0) NS ECOG PS 0 8 (26.7) 1 (5.6) NS 1 (10) 0 (0.0) 1 21 (70.0) 13 (72.2) 10 (90) 4 (66.7) NS 2 1 (3.3) 4 (22.2) 0 (0.0) 2 (33.3) Histology Adenocarcinoma 26 (86.7) 16 (88.9) 6 (60) 4 (66.6) NS Squamous cell carcinoma 3 (10.0) 2 (11.1) NS 4 (40) 1 (16.7) Others 1 (3.3) 0 (0.0) 0 (0.0) 1 (16.7) Number of metastatic sites ≤ 2 25 (83.3) 7 (38.9) <0.01 7 (70) 4 (66.6) NS > 2 5 (16.7) 11 (61.1) 3 (30) 2 (33.3) Pembrolizumab treatment Monotherapy 17 (56.7) 15 (83.3) NS 5 (50) 3 (50.0) NS Plus chemotherapy 13 (43.3) 3 (16.7) 5 (50) 3 (50.0) PFS (median days) 19.3 months 2.3 months <0.001 10 months 2.7 months <0.001 OS (median days) 35.7 months 3.4 months <0.001 NA 3.9 months <0.001 Abbreviations: NR, non-responders; R, responders; SD: standard deviation. Regarding the clinical characteristics of R and NR patients it’s important to mention that 61.1% of NR had more than two metastatic sites versus 16.7% for the R group (Fisher test; p-value< 0.01). In addition, the R group showed PATRICIA MONDELO MACÍA 280 significant better PFS (19.27 versus 2.3 months, respectively) and OS (35.7 versus 3.4 months, respectively) than NR (Table 4.1). Figure 4.1. (A) Swimmers’ plot on patients showing the response of therapy. The total length of each bar indicates the duration of survival from the diagnoses. (B-C) Kaplan- Meier plots show highly significant differences between responders and non-responders to pembrolizumab treatment in PFS (B) and OS (C). In the validation cohort, the analysed samples were taken from 16 patients with advanced NSCLC undergoing first-line pembrolizumab Chapter IV 287 Table 4.2. Characteristics of the 7 proteins that allow us to predict immunotherapy response in advanced NSCLC patients. Protein and Uniprot reference Name Gene Up or Down SPTN2_O15020 Spectrin beta chain, nonerythrocytic 2 SPTBN2 Up PGTA_Q92696 Geranylgeranyl transferase type-2 subunit alpha RABGGTA Up FIL1L_Q4L180 Filamin A-interacting protein 1- like FILIP1L Down ATG9A_Q7Z3C6 Autophagy-related protein 9A ATG9A Down LZTL1_Q9NQ48 Leucine zipper transcription factor-like protein 1 LZTL1 Down HPS5_Q9UPZ3 BLOC-2 complex member HPS5 HPS5 Down DCDC2_Q9UHG0 Doublecortin domain-containing protein 2 DCDC2 Down Single proteins showed a good discriminatory power to distinguish responders to non-responders (AUC= 0.72-0.78). In addition, in ROC analyses their combination showed a high AUC (AUC= 1). (Figure 4.8), in contrast with the results obtained with the PD-L1 expression on tumour tissue (AUC= 0.638). PD-L1 expression showed a lower predictive value than our proteins and their combination (Figure 4.8C). PATRICIA MONDELO MACÍA 288 Figure 4.8. (A) Volcano plot showing the 7 proteins of our predictive model. (B) ROCs curves of each protein and the combination in the discovery cohort. (C) ROC curve analyses of the PD-L1 expression on tumour tissue. Abbreviations: AUC, area under the curve; FC, foldchange. Next, an internal validation was performed using an additional cohort of 16 NSCLC patients (Table 4.1), from whom blood samples were collected in EDTA tubes instead CellSave tubes but analysed with the same methodology than the discovery cohort. The protein levels of our proteomic signature were analysed between R (n=10) and NR (n=6). No significant differences were found between both groups, likely due to the small size of the cohort. Nevertheless, ROC analyses were conducted to combine the 7 proteins of the model, which showed a perfect discrimination rate (AUC=1) in the validation cohort (Figure 4.9). Chapter IV 289 Figure 4.9. ROC curves of each protein and the combination in the validation cohort. 3.5 The proteome signature was associated with progression free survival and overall survival in NSCLC patients. To explore the prognostic significance of the 7 selected proteins, we investigated their expression levels in association with PFS and OS in our global cohort. Kaplan-Meier curve analysis showed that low levels of ATG9A and high levels of SPTN2 were significantly associated with longer PFS (logrank p < 0.001 and log rank p = 0.017 for ATG9A and SPTN2, respectively) in NSCLC patients. Additionally, low levels of HPS5 showed a trend towards being associated with longer PFS (log-rank p =0.054) and were significantly associated with longer OS (log-rank p <0.001). Low expression levels of DCDC2 were also significantly associated with longer OS in our global cohort (log-rank p <0.01) (Figure 4.10). No association with the patients’ survival was found for the proteins PGTA, FIL1L, and LZTL1. PATRICIA MONDELO MACÍA 290 Figure 4.10. Kaplan-Meier survival analysis of ATG9A, SPTN2, HPS5 and DCDC2 proteins levels at baseline. Kaplan-Meier plots of PFS (A-C) and OS (D-E) showed significant associations between proteins levels before start therapy and prognosis in metastatic NSCLC under immunotherapy regimens. 3.6 Dynamics of the Proteome Signature during pembrolizumab therapy Next, 107 plasma samples collected at different time points were analysed to investigate whether our protein model undergoes changes during immunotherapy treatment. A total of 46 samples were analysed at 6 weeks, 36 Chapter IV 291 at 12 weeks, and 25 at the time of disease progression. The results obtained showed that proteins levels did not show significant changes during different time points of the immunotherapy regimen, neither in the overall cohort nor when separated by response groups (Figure 4.11). Figure 4.11. Protein levels ATG9A, DCDC2, FIL1L, HPS5, LZTL1, PGTA and SPTN2 were measured at different time points during the immunotherapy regimen. Abbreviations: PD, progression of disease. ROC curves analyses at 6 weeks and at 12 weeks were also performed. In all time points, our model allows us to predict therapy response with an AUC >0.70. Importantly, at 12 weeks after start therapy, a good discriminatory rate was found (AUC=0.935) (Figure 4.12), suggesting the PATRICIA MONDELO MACÍA 292 potential value of the model to determine the response to immunotherapy also during the treatment. Figure 4.12. ROC analyses with the protein model at different time points during the immunotherapy regimen (at 6 weeks and at 12 weeks after start therapy). 4. DISCUSSION Immunotherapy, in particular ICIs, have improved survival rates in cancer patients, including those with NSCLC. However, it remains unclear how to select those patients who will respond to this therapy. PD-L1 expression on tumour tissue has been reported as a possible biomarker, but some patients without PD-L1 expression also respond to ICIs (381,382). TMB, which refers to the number of somatic mutations in tumours, has recently emerged as another possible biomarker in the immunotherapy field. Nevertheless, there are several challenges for clinical implementation of TMB, especially in standardizing detection methods and appropriate thresholds by tumour type (383,384). MSI, usually based on immunohistochemistry analyses for four mismatch repair proteins, may represent a novel biomarker to select patients who will benefit from ICIs Chapter IV 293 therapy in several cancer types (70); however, this alteration is scarcely present in lung cancer (385). The search for new predictive and prognostic biomarkers for ICI therapies will improve patient selection. Proteomics, the study of the entire set of proteins expressed in a cell, tissue, or individual, has become an important field in molecular sciences, as it provides valuable information on the identity, expression levels, and modification of proteins (386). In this work, we hypothesized that plasma proteome analyses could have predictive and prognostic value in newly diagnosed NSCLC patients who started pembrolizumab therapy. We employed SWATH-MS, a new technology that allows detection of tens of thousands of peptides in a single injection and enables identification and quantification of circulating proteins. Previously, several studies in breast (387), bladder (388), and endometrial cancer patients (389), among others, have shown the potential of this emerging technology. Importantly, our study apport a dataset that might serve the scientific community as a resource of clinical proteomic data in lung cancer samples, previously reported in melanoma (390) and ovarian cancer (391,392). More importantly, the analyses had allowed us to identify 324 DEPs between R and NR, showing the potential value of proteomic analysis to identify differences between both groups. Previously our group has employed the methodology here described to study the red blood cells population proteome in breast cancer patients (393). A similar study focuses on breast cancer reported that this approach was useful to determine DEPs in serum samples associated with the response to neoadjuvant chemotherapy. The authors of this study described three proteins specially correlated with resistance to this therapy (387). Some other studies focused on melanoma PATRICIA MONDELO MACÍA 294 patients, also reported the potential of proteomic technologies to identify DEPs and predict therapy response in tissue (390) and plasma samples (365). Regarding the enrichment analyses of biological functions and pathways, and even though the results should be interpreted with caution, we observed a high degree of variability in GO processes among the 324 DEPs between R and NR groups, indicating the significant heterogeneity present in the plasma samples. These samples can reflect both individual characteristics and tumour features, as well as the tumour microenvironment (394). Next, comparing both groups, we found that the process involved in immune system was enriched exclusively in R, as expected given the known role of the immune system in response to ICI therapies. More analyses should be performed to confirm these results. In KEGG pathways, R appear to have higher levels of metabolic proteins than NR patients, but caution should be exercised when interpreting the results due to the lack of statistical significance. Nevertheless, our findings are consistent with those of Harel et al. (390) in their study of melanoma patients. Their research aimed to investigate why most cancer patients do not respond to immunotherapy by profiling the proteome of tumour tissue samples from advanced melanoma patients undergoing either TIL-based or anti-PD1 immunotherapy. They found 414 DEPs and 636 DEPs between responder and non-responder patients in the TIL and anti-PD1 immunotherapy cohort, respectively. Pathway analyses revealed that responders had higher oxidative phosphorylation and lipid metabolism, which increases melanoma immunogenicity and sensitivity to T cell-mediated killing. These findings suggest that the metabolic state of melanoma may play a role in the response to immunotherapy and could lead to improved therapeutic responses in the future. A similar process may be occurring in our NSCLC patients. Chapter IV 295 Next, a predictive model to identify proteins that can be associated with the response to immunotherapy treatment was developed, showing that the combination of 7 proteins can predict the response to immunotherapy with high sensitivity and specificity, in a discovery cohort and, in addition, in a validation cohort. Regarding the proteins, five of them were more expressed in NR than R group: ATG9A, DCDC2, HPS5, FIL1L and LZTL1. The ATG9A protein is a transmembrane protein that plays an important role in the formation and regulation of autophagosomes, which are cellular structures involved in the process of autophagy. This protein is localized on the membranes of the endoplasmic reticulum and Golgi (395). A recent study characterizes novel functions of ATG9A as component of a TNF-induced cell death checkpoint (396). DCDC2 is a protein expressed in the brain that has been linked to the development of cognitive skills such as reading and language processing. DCDC2 has been found to be localized to the cilia of neurons in the developing brain, where is involved in the control of ciliogenesis and ciliary length (397,398). In addition, DCDC2 seems to play a role in the inhibition of canonical Wnt signaling pathway (399). HPS5 is a protein that plays a crucial role in the biogenesis and function of lysosomerelated organelle and may be involved in the regulation of general functions of integrins (400). On the other side FIL1L is a protein that acts as a regulator of the antiangiogenic activity on endothelial cells. The overexpression in endothelial cells leads to inhibition of cell proliferation and migration and increase in apoptosis (401). In oncology field, FIL1L has been previously reported as a protein downregulated in ovarian cancer (402). LZTL1 regulates ciliary localization of the BBSome complex (403) and may have tumour suppressor function in several primary cancer types (404). PATRICIA MONDELO MACÍA 296 In another hand, PGTA and SPTN2, showed higher levels in R than NR group in our study. Both proteins have been less explored. Regarding PGTA it is believed to play a role as catalytic protein (405). In contrast, SPTN2 probably plays an important role in neuronal membrane skeleton. To our knowledge, this is the first time that these proteins have been associated with immunotherapy response in NSCLC patients. Importantly, our proteomic signature (considering the combination of the 7 proteins or the use of single proteins) showed a stronger value to predict immunotherapy response than the actual biomarker, the PD-L1 expression in tumour tissue. Hence, the proposed model could be useful to select which patients can be treated with immunotherapy, and in which patients, an alternative therapy should be considered. In addition, survival analyses showed that low levels of ATG9A, DCDC2, and HPS5 were associated with longer PFS or OS rates, while low levels of SPTN2 were significantly associated with worse OS. In another hand, longitudinal samples were analysed in the global cohort to investigate the value of the predictive model during immunotherapy. Although no significant differences were found between the R and NR groups during therapy, ROC analyses at 6 and 12 weeks showed a good discriminatory rate between the two groups. The determination at 6 weeks could assist clinicians in detecting an early response to therapy in NSCLC patients or in cases where sample extraction was not possible at baseline. Interestingly, at 12 weeks a higher AUC was found with our model, which could help clinicians determine treatment response in cases where patients experience pseudoprogressions through imaging techniques (406), suggesting Overall discussion 303 In 2018, a blood test named CancerSEEK highlighted the importance of using liquid biopsy samples in cancer screening (410). Indeed, the test showed potential value to early detection of eight common cancer types through assessment of the levels of circulating proteins and mutations in cfDNA. In agreement, numerous studies have reported in the last years the potential value of different liquid biopsy components to guide treatment decisions, monitor disease progression, detect minimal residual disease and predict recurrence (411). The present thesis focuses on advanced lung cancer patients, the leading cause of death among solid tumours worldwide. The value of different liquid biopsy components (ctDNA, CTCs, total cfDNA and circulating proteins) to improve the management of these patients was further investigated. Overall, 250 lung cancer patients were enrolled in the current thesis. Figure 26 showed the 250 patients subdivided according to lung cancer type, and treatment regimen. Figure 26. Sankey diagram showing the total lung cancer patients included in the present thesis, subdivided by lung cancer type, stage and therapy. PATRICIA MONDELO MACÍA 304 NSCLC serves as a prime example of the move towards precision medicine in cancer treatment. Fortunately, nowadays there are numerous options and treatments available for the management of NSCLC patients. Although some blood-based technologies have been already approved by the FDA to guide the therapy selection (see Introduction section, point 2), the prioritization of the most helpful for each case is not always clear. In the present thesis, one of the first objectives was to investigate and select the more accurate liquid biopsy-based technology to choose those patients that will benefit from EGFR and MET inhibitors. Despite ctDNA can be found in a variety of fluids, including urine, peritoneal fluid or cerebrospinal fluids (CSF) (412–414), plasma remains the preferred source (415). Thus, two different platforms to analyse specific mutations in ctDNA derived from blood samples were studied (CHAPTER I). In CHAPTER I.A, advanced NSCLC patients were analysed by Idylla™ technology, a qPCR-based method, to detect EGFR mutations in plasma samples. In contrast, a ddPCR-based method was employed in CHAPTER I.B to detect MET amplification. It is well known that the sensibility of a qPCR method such as Idylla™ technology, is lower than other technologies such as ddPCR or some NGS approaches (154). Nevertheless, we hypothesised that, in some cases, a qPCR determination can be useful and present the advantage to be more rapid and simple than other approaches. Results obtained in the first chapter of this thesis showed a good overall concordance with NGS method (93.3%) or BEAMing technology (88.9%) in our cohort of metastatic cancer patients, demonstrating its utility to have a fast and accurate results to genotype the tumour despite the lowest sensitivity. Therefore, our results demonstrate that Overall discussion 305 Idylla™ technology represents an alternative to the classic qPCR or ddPCR to determine the EGFR status in NSCLC patients, always considering its limitation for detecting ctDNA presence in tumours low cfDNA shedders, since cases below 0.5% of VAFs were found as false negatives in our study. Thus, the EGFR characterization on cfDNA in patients with low tumour burden (146) or metastasis located only in brain, which release lower DNA amount, should be attempted with a high-sensitive technology such as the ddPCR. Also, in cases where the input of cfDNA is low the preference should be the most sensitive technology. Our results incise on the importance of standardising the method for the ctDNA genotyping and pointed out to the use of the qPCR assay as a first EGFR screen to avoid costs and long result-times. In addition, thanks to the simple workflow that Idylla™ technology offers, it can be easily implemented into clinical routine. In CHAPTER I.B, blood samples of NSCLC patients were analysed in order to detect MET amplification, an alteration associated with acquired resistance to EGFR inhibitors (416). In this case, due to the detection of CNA needs higher sensitivity and sensibility than a qPCR method, a ddPCR bloodbased test was developed. High detection rate of MET amplification was found in patients with high cfDNA levels suggesting that those patients who present high tumour burden and, in addition, who received several lines of TKIs are likely to present MET amplification, so they are good candidates to make the MET assessment in cfDNA looking for therapy alternatives. Importantly the ddPCR assay employed in the present thesis can represents a transversal clinical tool as it can be applied in different tumour types. Thus, in the first chapter of this thesis (CHAPTER I.A and I.B), we showed the utility of two different blood-based technologies to detect ctDNA PATRICIA MONDELO MACÍA 306 alterations: a qPCR method to detect EGFR mutations and ddPCR to detect MET amplification. These minimally invasive determinations can help to make important therapeutic decisions. Idylla™ assay could be employed as a first screen in metastatic NSCLC patients to select who will benefit from TKI based therapies and to monitor tumour evolution. In the case of MET amplification, ddPCR method represents the most sensitive strategy to identify this resistance mechanism against the activity of EGFR inhibitors in patients who have undergone several lines of treatment. In both cases, bloodbased technologies allow to detect clinically relevant alterations in noninvasive samples and provide a more complete image of the disease, since information from metastatic locations is also evaluated with both strategies. However, although qPCR and ddPCR methods allowed us to detect specific mutations in ctDNA of blood samples, one limitation of these techniques is the requirement of previous information about the tumour and the mutations characterising this tumour. For this reason, PCR-based techniques to interrogate ctDNA are commonly employed to select targeted therapies or detect mutations associated with resistance in clinical context where these mutations are clearly defined. Thus, the lack of clear driver genes or a high genetic heterogeneity make more difficult the clinical translation of the ctDNA assessment in some tumours, including for example SCLC. To face this limitation, in the present thesis we wanted to explore the possible clinical interest of quantifying total cfDNA levels in advanced lung cancer patients as a simpler and cost-effective analysis than ctDNA determination. CfDNA is released into the bloodstream through apoptosis or necrosis (among other mechanisms; see Introduction section, point 2 to expand the information) of both normal and tumour cells. cfDNA from normal cells is found in plasma at Overall discussion 307 low levels (347). In contrast, tumour cell turnover is higher than that of normal tissue, therefore, cancer patients tend to have higher total amounts of cfDNA than healthy patients (347). In CHAPTER II and CHAPTER III, the total cfDNA levels in advanced NSCLC and SCLC patients were quantified using a qPCR-based method. As result, we validated a simple blood-based qPCR tool that allowed us to quantify total cfDNA levels with high sensitivity and specificity and discriminated cancer patients from healthy controls. Interestingly, some studies similar to our approach, reported a good discrimination between melanoma patients (stage III and IV) and healthy controls based on cfDNA levels quantification by Qubit method (AUC=0.72) (417). Similar results were also obtained among bladder cancer patients and healthy controls (418) interrogating cfDNA content. In concordance with bibliography, results derived from this thesis reported that total cfDNA levels (based on the detection of hTERT gene) from lung cancer patients (both advanced NSCLC and SCLC patients) were higher than cfDNA levels in a control population cohort, showing a good discriminatory power (AUC=0.88) (Figure 27). Moreover, in SCLC patients cfDNA levels were higher than in NSCLC patients, in concordance with the aggressiveness that characterized this lung cancer type. PATRICIA MONDELO MACÍA 308 Figure 27. Total cfDNA levels in lung cancer patients and healthy controls. (A) Boxplot of cfDNA levels in NSCLC, SCLC, and healthy controls (Mann-Whitney test). (B) ROC showing the discriminatory value of cfDNA levels to discriminate against lung cancer patients versus healthy controls (AUC=0.8885). In both lung cancer types high cfDNA levels before therapy onset were determined as a negative prognostic factor in terms of PFS and OS in NSCLC patients treated with immunotherapy and in SCLC patients treated with immunotherapy or immunotherapy in combination with chemotherapy. According to our results, several studies have reported high cfDNA levels as a negative prognostic factor in other cancer types such as melanoma (417,419), colorectal (420), prostate (421) and bladder cancer (418), confirming the interest of quantify total cfDNA levels independently of the value of the ctDNA assessment. In addition, we reported that cfDNA dynamics provides predictive and prognostic information in both NSCLC patients under immunotherapy treatment and SCLC under immunotherapy or chemotherapy regimen. In the first case, we described a prognostic and predictive value of cfDNA levels 12 weeks after therapy onset. In SCLC, cfDNA levels analysed 3 weeks after the Overall discussion 309 therapy administration had a prognostic role. The early results obtained in SCLC patients could suggest that due to the aggressiveness of this subtype, molecular changes can be detected earlier than in NSCLC. Results derived from this thesis are in accordance with a recent study, where quantitative changes in cfDNA values between baseline and time of radiological evaluation correlated with responses to therapy and relapse of disease in treatment-naïve patients with advanced NSCLC undergoing TKIs- and immunotherapy-based treatments (422). Another study showed the potential utility of quantifying cfDNA levels at 4 weeks. Indeed, elevated cfDNA levels at 4 weeks were reported as a negative prognostic factor for OS in metastatic castration-resistant prostate cancer treated with androgen receptor pathway inhibitor (421). Together with cfDNA, CTCs constitute the main liquid biopsy biomarkers employed in the oncology field. CTCs can be shed from primary tumour tissue or different metastatic localizations, travel into the blood to future metastatic sites and represent a good option to capture the tumour heterogeneity in metastatic lung cancer patients (423). Moreover, several studies have shown that CTC numbers may serve as useful predictive markers of recurrence and survival in patients with solid tumours, including NSCLC (333,424). However, the widespread adoption of CTCs in routine clinical practice will require strong evidence of their analytic validity, clinical validity, and most importantly, their clinical utility (425,426). In NSCLC, the number of CTCs detected in the blood is usually low, around 1-10 cells. Some studies suggest that this low number can be due to CTCs in NSCLC may be under an EMT (427), therefore, finding a good platform that enables detection of all types of CTCs remains to be a challenge. PATRICIA MONDELO MACÍA 310 Taking this context in mind, in CHAPTER I.B and CHAPTER II, two different platforms were compared to investigate the best CTC isolation method in NSCLC patients: Parsortix system (Angle, Germany), a size-based method and, CellSearch® system (Menarini, Bologna, Italy) an EpCAM- dependent method (Figure 28). In concordance with bibliography, results derived from these chapters do not allow to define which platform is the best, since we obtained a high heterogeneity of the CTCs count determined with both technologies among patients. In another hand, different assays to detect the expression of MET and PD-L1 expression in CTCs were developed in both systems. Figure 28. CTCs detected in 36 NSCLC patients employing two different isolation methods. In addition, in CHAPTER III, we employed the CellSearch® system to detect CTCs in SCLC patients. Comparing the NSCLC cohort (from Chapter I.B and Chapter II) and SCLC cohort, we observed that in SCLC the number of CTCs detected was significantly higher than in NSCLC patients, closely Overall discussion 311 related to the higher stage and poorer prognosis of this tumour type (Figure 29). Indeed, in NSCLC 63.3% of patients did not present CTCs while in SCLC only 14.3% of patients did not present any CTCs. Figure 29. CTCs detected in NSCLC and SCLC cohorts using the CellSearch® system. SCLC patients presented statistically higher number of CTCs than NSCLC patients (pvalue <0.001; Mann-Whitney test). Survival analyses were performed in both subtypes to predict the prognostic role of CTCs in NSCLC under immunotherapy (n=30) and in SCLC under chemotherapy (n=21). Comparing both cohorts, very different cut-offs (≥1 CTCs in NSCLC patients and ≥150 CTCs in SCLC) at baseline were employed, due again to the higher aggressiveness that characterized to SCLC. The presence of CTCs has been associated with worse PFS and OS in NSCLC under immunotherapy regimens (n=30) and multivariate analyses showed their independence value. In contrast, in SCLC the prognostic role of CTCs was less clear. A discrete association between high CTCs levels and PFS was found but multivariate analyses did not show their independent value. PATRICIA MONDELO MACÍA 312 The results derived from these chapters suggest that epithelial CTCs play a more important role in advanced NSCLC patients in contrast with SCLC. A possible explanation could be related to the EMT. It’s important to keep in mind that both in NSCLC and SCLC cohort, we detected CTCs with an epithelial phenotype. However, some CTCs could be not detected due to a more mesenchymal phenotype, correlated in some studies with poor clinical outcomes (428). In our cohort of SCLC patients, the faint prognostic impact found suggests than in this cancer type, mesenchymal CTCs could have a principal role in tumour evolution. In another hand, in CHAPTER II and CHAPTER III, we evaluated the prognostic role of combining CTCs, cfDNA levels and clinical characteristics. We can conclude that the combination of CTCs and cfDNA in NSCLC or cfDNA and clinical variables in SCLC, represents a powerful tool for cancer management, offering valuable information for treatment decision-making and predicting patient outcomes in lung cancer patients, previously described in breast cancer (429) and bladder cancer (430) among others. Multimodality approaches, incorporating several components as those employed in the present thesis, seem to be the correct direction to reach a personalised medicine. Last but not least, in CHAPTER IV, we characterize other liquid biopsy components, circulating proteins, showing their potential utility to select who patients will benefit from immunotherapy and which regimen is the most appropriated. Several benefits of using circulating proteins over CTCs and cfDNA have been proposed in the field of cancer research. Firstly, circulating proteins are typically easier to measure, makes them a more accessible and cost-effective biomarker for clinical use (431). In addition, circulating 319 CONCLUSIONS Overall, results obtained in our work provided relevant information to standardize new protocols to study different circulating biomarkers and demonstrated their potential to help clinicians in therapy decision making and the follow-up of lung cancer patients. The main conclusions drawn from the current PhD thesis research are summarized below. 1. Idylla™ ctEGFR Mutation Assay showed a good concordance with the ddPCR or BEAMing assays in advanced NSCLC patients with VAFs > 0.5%. These results support the use of this qPCR-based assay as a screen tool to detect EGFR mutations and guide treatment with TKIs. 2. We established specific and non-invasive assays to monitor MET status in cfDNA/CTCs and demonstrated its utility as a biomarker for monitoring the appearance of resistance to anti-EGFR therapy in advanced NSCLC patients. 3. Two efficient strategies (EpCAM dependent and independent) to monitor PD-L1 expression on CTCs from patients with advanced NSCLC were developed, however the detection of CTCs PD-L1+, lacked predictive and prognostic significance. 4. Both baseline CTCs, and cfDNA levels, analysed independently, showed prognostic value in metastatic NSCLC patients receiving pembrolizumab PATRICIA MONDELO MACÍA 320 therapy. In addition, combined analysis of CTCs and cfDNA improved the capacity to predict progression-free survival (PFS) rates. 5. The presence of high cfDNA levels in newly diagnosed SCLC patients was associated with a poor evolution of the disease in terms of PFS and overall survival (OS) rates. 6. We identified a model based on cfDNA levels, ECOG PS and the patients’ sex that has value in identifying SCLC patients who particularly benefit from the treatment. 7. Extensive characterization of plasma circulating proteins in metastatic NSCLC patients allowed us to identify a novel protein signature (including ATG9A, DCDC2, HPS5, FIL1L, LZTL1, PGTA, and SPTN2) associated with the response to immunotherapy. REFERENCES 323 REFERENCES 1. Mondelo-Macía P, Castro-Santos P, Castillo-García A, Muinelo-Romay L, Diaz-Peña R. Circulating Free DNA and Its Emerging Role in Autoimmune Diseases. J Pers Med. 2021 Feb;11(2). 2. Mondelo-Macía P, Rodríguez-Ces AM, Suárez-Cunqueiro MM, Romay LM. Methods for the Detection of Circulating Biomarkers in Cancer Patients. Adv Exp Med Biol. 2022;1379:525–52. 3. Mondelo-Macía P, García-González J, León-Mateos L, Castillo-García A, López-López R, Muinelo-Romay L, et al. Current status and future perspectives of liquid biopsy in small cell lung cancer. Biomedicines. 2021;9(1):1–22. 4. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209–49. 5. Kocher F, Hilbe W, Seeber A, Pircher A, Schmid T, Greil R, et al. Longitudinal analysis of 2293 NSCLC patients: a comprehensive study from the TYROL registry. Lung Cancer. 2015 Feb;87(2):193–200. 6. Tobacco Smoking and Cancer of the Lung: Statement of the [British] Medical Research Council. CA Cancer J Clin. 1958 Mar 1;8(2):66–8. 7. Wynder EL. Tobacco as a cause of lung cancer: some reflections. Am J Epidemiol. 1997 Nov;146(9):687–94. 8. Thun M, Peto R, Boreham J, Lopez AD. Stages of the cigarette epidemic on entering its second century. Tob Control. 2012 Mar;21(2):96–101. 9. Ng M, Freeman MK, Fleming TD, Robinson M, Dwyer-Lindgren L, Thomson B, et al. Smoking prevalence and cigarette consumption in 187 countries, 1980- 2012. JAMA. 2014 Jan;311(2):183–92. 10. Sun S, Schiller JH, Gazdar AF. Lung cancer in never smokers - A different disease. Nat Rev Cancer. 2007;7(10):778–90. 11. Subramanian J, Govindan R. Molecular genetics of lung cancer in people who have never smoked. Lancet Oncol. 2008;9(7):676–82. 12. Corrales L, Rosell R, Cardona AF, Martín C, Zatarain-Barrón ZL, Arrieta O. Lung cancer in never smokers: The role of different risk factors other than tobacco smoking. Crit Rev Oncol Hematol. 2020;148. PATRICIA MONDELO MACÍA 324 13. Malhotra J, Malvezzi M, Negri E, La Vecchia C, Boffetta P. Risk factors for lung cancer worldwide. Eur Respir J. 2016;48(3):889–902. 14. Turner MC, Andersen ZJ, Baccarelli A, Diver WR, Gapstur SM, Pope CA 3rd, et al. Outdoor air pollution and cancer: An overview of the current evidence and public health recommendations. CA Cancer J Clin. 2020 Aug; 15. Mucci LA, Hjelmborg JB, Harris JR, Czene K, Havelick DJ, Scheike T, et al. Familial Risk and Heritability of Cancer Among Twins in Nordic Countries. JAMA. 2016 Jan;315(1):68–76. 16. Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, et al. The 2015 World Health Organization Classification of Lung Tumors: Impact of Genetic, Clinical and Radiologic Advances since the 2004 Classification. J Thorac Oncol. 2015;10(9):1243–60. 17. Thai AA, Solomon BJ, Sequist L V., Gainor JF, Heist RS. Lung cancer. Lancet. 2021;398(10299):535–54. 18. Huang R, Wei Y, Hung RJ, Liu G, Su L, Zhang R, et al. Associated Links Among Smoking, Chronic Obstructive Pulmonary Disease, and Small Cell Lung Cancer: A Pooled Analysis in the International Lung Cancer Consortium. EBioMedicine. 2015 Nov;2(11):1677–85. 19. Yu L, Lai Q, Gou L, Feng J, Yang J. Opportunities and obstacles of targeted therapy and immunotherapy in small cell lung cancer. J Drug Target. 2020;0(0):000. 20. Edition S, Edge SB, Byrd DR. AJCC cancer staging manual. AJCC cancer staging Man. 2017; 21. Detterbeck FC, Boffa DJ, Kim AW, Tanoue LT. The Eighth Edition Lung Cancer Stage Classification. Chest. 2017;151(1):193–203. 22. Van Meerbeeck JP, Fennell DA, De Ruysscher DKM. Small-cell lung cancer. Lancet. 2011;378(9804):1741–55. 23. Micke P, Faldum A, Metz T, Beeh KM, Bittinger F, Hengstler JG, et al. Staging small cell lung cancer: Veterans Administration Lung Study Group versus International Association for the Study of Lung Cancer - What limits limited disease? Lung Cancer. 2002 Sep;37(3):271–6. 24. Howlader N, Noone AM, Krapcho M, Miller D, Brest A, Yu M, Ruhl J, Tatalovich Z, Mariotto A, Lewis DR, Chen HS, Feuer EJ CK. SEER Cancer Statistics Review, 1975-2016, National Cancer Institute. Natl Cancer Institute. 2018; 25. Planchard D, Popat S, Kerr K, Novello S, Smit EF, Faivre-Finn C, et al. Metastatic non-small cell lung cancer: ESMO Clinical Practice Guidelines for References 325 diagnosis, treatment and follow-up. Ann Oncol. 2018;29(Supplement 4):iv192– 237. 26. Wang M, Herbst RS, Boshoff C. Toward personalized treatment approaches for non-small-cell lung cancer. Nat Med. 2021;27(8):1345–56. 27. Herbst RS, Morgensztern D, Boshoff C. The biology and management of nonsmall cell lung cancer. Nature. 2018;553(7689):446–54. 28. Howlader N, Forjaz G, Mooradian MJ, Meza R, Kong CY, Cronin KA, et al. The Effect of Advances in Lung-Cancer Treatment on Population Mortality. N Engl J Med. 2020;383(7):640–9. 29. Allemani C, Matsuda T, Di Carlo V, Harewood R, Matz M, Nikšić M, et al. Global surveillance of trends in cancer survival 2000–14 (CONCORD-3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries. Lancet. 2018;391(10125):1023–75. 30. Mok TS, Wu Y-L, Thongprasert S, Yang C-H, Chu D-T, Saijo N, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med. 2009 Sep;361(10):947–57. 31. Zhou C, Wu YL, Chen G, Feng J, Liu X-Q, Wang C, et al. Final overall survival results from a randomised, phase III study of erlotinib versus chemotherapy as first-line treatment of EGFR mutation-positive advanced non-small-cell lung cancer (OPTIMAL, CTONG-0802). Ann Oncol Off J Eur Soc Med Oncol. 2015 Sep;26(9):1877–83. 32. Rosell R, Carcereny E, Gervais R, Vergnenegre A, Massuti B, Felip E, et al. Erlotinib versus standard chemotherapy as first-line treatment for European patients with advanced EGFR mutation-positive non-small-cell lung cancer (EURTAC): a multicentre, open-label, randomised phase 3 trial. Lancet Oncol. 2012 Mar;13(3):239–46. 33. Wu Y-L, Zhou C, Liam C-K, Wu G, Liu X, Zhong Z, et al. First-line erlotinib versus gemcitabine/cisplatin in patients with advanced EGFR mutation-positive non-small-cell lung cancer: analyses from the phase III, randomized, open-label, ENSURE study. Ann Oncol Off J Eur Soc Med Oncol. 2015 Sep;26(9):1883– 9. 34. Yang JJ, Zhou Q, Yan HH, Zhang XC, Chen HJ, Tu HY, et al. A phase III randomised controlled trial of erlotinib vs gefitinib in advanced non-small cell lung cancer with EGFR mutations. Br J Cancer. 2017 Feb;116(5):568–74. 35. Sequist L V, Yang JC-H, Yamamoto N, O’Byrne K, Hirsh V, Mok T, et al. Phase III study of afatinib or cisplatin plus pemetrexed in patients with metastatic lung adenocarcinoma with EGFR mutations. J Clin Oncol Off J Am Soc Clin Oncol. 2013 Sep;31(27):3327–34. PATRICIA MONDELO MACÍA 326 36. Rotow J, Bivona TG. Understanding and targeting resistance mechanisms in NSCLC. Nat Rev Cancer. 2017;17(11):637–58. 37. Yun CH, Mengwasser KE, Toms A V., Woo MS, Greulich H, Wong KK, et al. The T790M mutation in EGFR kinase causes drug resistance by increasing the affinity for ATP. Proc Natl Acad Sci U S A. 2008;105(6):2070–5. 38. Cortot AB, Jänne PA. Molecular mechanisms of resistance in epidermal growth factor receptor-mutant lung adenocarcinomas. Eur Respir Rev an Off J Eur Respir Soc. 2014 Sep;23(133):356–66. 39. Schmid S, Li JJN, Leighl NB. Mechanisms of osimertinib resistance and emerging treatment options. Lung Cancer. 2020 Sep;147:123–9. 40. Offin M, Chan JM, Tenet M, Rizvi HA, Shen R, Riely GJ, et al. Concurrent RB1 and TP53 Alterations Define a Subset of EGFR-Mutant Lung Cancers at risk for Histologic Transformation and Inferior Clinical Outcomes. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer. 2019 Oct;14(10):1784–93. 41. Yu HA, Arcila ME, Rekhtman N, Sima CS, Zakowski MF, Pao W, et al. Analysis of tumor specimens at the time of acquired resistance to EGFR-TKI therapy in 155 patients with EGFR-mutant lung cancers. Clin cancer Res an Off J Am Assoc Cancer Res. 2013 Apr;19(8):2240–7. 42. Lee J-K, Lee J, Kim S, Kim S, Youk J, Park S, et al. Clonal History and Genetic Predictors of Transformation Into Small-Cell Carcinomas From Lung Adenocarcinomas. J Clin Oncol Off J Am Soc Clin Oncol. 2017 Sep;35(26):3065–74. 43. Boumahdi S, de Sauvage FJ. The great escape: tumour cell plasticity in resistance to targeted therapy. Nat Rev Drug Discov. 2020;19(1):39–56. 44. Beatty GL, Gladney WL. Immune escape mechanisms as a guide for cancer immunotherapy. Clin cancer Res an Off J Am Assoc Cancer Res. 2015 Feb;21(4):687–92. 45. Havel JJ, Chowell D, Chan TA. The evolving landscape of biomarkers for checkpoint inhibitor immunotherapy. Nat Rev Cancer. 2019 Mar;19(3):133–50. 46. Morad G, Helmink BA, Sharma P, Wargo JA. Hallmarks of response, resistance, and toxicity to immune checkpoint blockade. Cell. 2021 Oct;184(21):5309–37. 47. Garon EB, Rizvi NA, Hui R, Leighl N, Balmanoukian AS, Eder JP, et al. Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J Med. 2015 May;372(21):2018–28. 48. Vokes EE, Ready N, Felip E, Horn L, Burgio MA, Antonia SJ, et al. Nivolumab versus docetaxel in previously treated advanced non-small-cell lung cancer (CheckMate 017 and CheckMate 057): 3-year update and outcomes in patients References 327 with liver metastases. Ann Oncol Off J Eur Soc Med Oncol. 2018 Apr;29(4):959–65. 49. Rittmeyer A, Barlesi F, Waterkamp D, Park K, Ciardiello F, von Pawel J, et al. Atezolizumab versus docetaxel in patients with previously treated non-small- cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial. Lancet. 2017;389(10066):255–65. 50. Herbst RSR, Baas P, Kim D-WD, Felip E, Pérez-Gracia JL, Han J-Y, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial. Lancet. 2016 Apr;387(10027):1540–50. 51. Reck M, Rodríguez-Abreu D, Robinson AG, Hui R, Csőszi T, Fülöp A, et al. Pembrolizumab versus Chemotherapy for PD-L1–Positive Non–Small-Cell Lung Cancer. N Engl J Med. 2016 Nov 10;375(19):1823–33. 52. Brahmer J, Reckamp KL, Baas P, Crinò L, Eberhardt WEE, Poddubskaya E, et al. Nivolumab versus Docetaxel in Advanced Squamous-Cell Non-Small-Cell Lung Cancer. N Engl J Med. 2015 Jul 9;373(2):123–35. 53. Borghaei H, Paz-Ares L, Horn L, Spigel DR, Steins M, Ready NE, et al. Nivolumab versus Docetaxel in Advanced Nonsquamous Non-Small-Cell Lung Cancer. N Engl J Med. 2015 Oct 22;373(17):1627–39. 54. Carbone DP, Reck M, Paz-Ares L, Creelan B, Horn L, Steins M, et al. First-Line Nivolumab in Stage IV or Recurrent Non–Small-Cell Lung Cancer. N Engl J Med. 2017 Jun;376(25):2415–26. 55. Gandhi L, Rodríguez-Abreu D, Gadgeel S, Esteban E, Felip E, De Angelis F, et al. Pembrolizumab plus Chemotherapy in Metastatic Non–Small-Cell Lung Cancer. N Engl J Med. 2018 May 31;378(22):2078–92. 56. Paz-Ares L, Luft A, Vicente D, Tafreshi A, Gümüş M, Mazières J, et al. Pembrolizumab plus Chemotherapy for Squamous Non–Small-Cell Lung Cancer. N Engl J Med. 2018 Nov 22;379(21):2040–51. 57. Schiller JH, Harrington D, Belani CP, Langer C, Sandler A, Krook J, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med. 2002 Jan;346(2):92–8. 58. Sandler A, Gray R, Perry MC, Brahmer J, Schiller JH, Dowlati A, et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med. 2006 Dec;355(24):2542–50. 59. Paz-Ares LG, de Marinis F, Dediu M, Thomas M, Pujol J-L, Bidoli P, et al. PARAMOUNT: Final overall survival results of the phase III study of maintenance pemetrexed versus placebo immediately after induction treatment PATRICIA MONDELO MACÍA 328 with pemetrexed plus cisplatin for advanced nonsquamous non-small-cell lung cancer. J Clin Oncol Off J Am Soc Clin Oncol. 2013 Aug;31(23):2895–902. 60. Formenti SC, Demaria S. Combining radiotherapy and cancer immunotherapy: a paradigm shift. J Natl Cancer Inst. 2013 Feb;105(4):256–65. 61. Marchetti A, Felicioni L, Malatesta S, Grazia Sciarrotta M, Guetti L, Chella A, et al. Clinical features and outcome of patients with non-small-cell lung cancer harboring BRAF mutations. J Clin Oncol Off J Am Soc Clin Oncol. 2011 Sep;29(26):3574–9. 62. Cardarella S, Ogino A, Nishino M, Butaney M, Shen J, Lydon C, et al. Clinical, pathologic, and biologic features associated with BRAF mutations in non-small cell lung cancer. Clin cancer Res an Off J Am Assoc Cancer Res. 2013 Aug;19(16):4532–40. 63. Awad MM, Oxnard GR, Jackman DM, Savukoski DO, Hall D, Shivdasani P, et al. MET Exon 14 Mutations in Non-Small-Cell Lung Cancer Are Associated With Advanced Age and Stage-Dependent MET Genomic Amplification and c- Met Overexpression. J Clin Oncol Off J Am Soc Clin Oncol. 2016 Mar;34(7):721–30. 64. Kalemkerian GP, Narula N, Kennedy EB, Biermann WA, Donington J, Leighl NB, et al. Molecular Testing Guideline for the Selection of Patients With Lung Cancer for Treatment With Targeted Tyrosine Kinase Inhibitors: American Society of Clinical Oncology Endorsement of the College of American Pathologists/International Association for the. J Clin Oncol Off J Am Soc Clin Oncol. 2018 Mar;36(9):911–9. 65. Lindeman NI, Cagle PT, Aisner DL, Arcila ME, Beasley MB, Bernicker EH, et al. Updated Molecular Testing Guideline for the Selection of Lung Cancer Patients for Treatment With Targeted Tyrosine Kinase Inhibitors: Guideline From the College of American Pathologists, the International Association for the Study of Lung Cancer, and the. Arch Pathol Lab Med. 2018 Mar;142(3):321– 46. 66. Gibney GT, Weiner LM, Atkins MB. Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Vol. 17, The Lancet Oncology. 2016. p. e542– 51. 67. Schoenfeld AJ, Rizvi H, Bandlamudi C, Sauter JL, Travis WD, Rekhtman N, et al. Clinical and molecular correlates of PD-L1 expression in patients with lung adenocarcinomas. Ann Oncol Off J Eur Soc Med Oncol. 2020 May;31(5):599– 608. 68. Lantuejoul S, Sound-Tsao M, Cooper WA, Girard N, Hirsch FR, Roden AC, et al. PD-L1 Testing for Lung Cancer in 2019: Perspective From the IASLC References 335 137. Kustanovich A, Schwartz R, Peretz T, Grinshpun A. Life and death of circulating cell-free DNA. Cancer Biol Ther. 2019;20(8):1057–67. 138. Aucamp J, Bronkhorst AJ, Badenhorst CPS, Pretorius PJ. The diverse origins of circulating cell-free DNA in the human body: a critical re-evaluation of the literature. Biol Rev. 2018;93(3):1649–83. 139. Thierry AR, El Messaoudi S, Gahan PB, Anker P, Stroun M. Origins, structures, and functions of circulating DNA in oncology. Cancer Metastasis Rev. 2016;35(3):347–76. 140. Fernando MR, Jiang C, Krzyzanowski GD, Ryan WL. New evidence that a large proportion of human blood plasma cell-free DNA is localized in exosomes. PLoS One. 2017;12(8):e0183915. 141. Duvvuri B, Lood C. Cell-free DNA as a biomarker in autoimmune rheumatic diseases. Front Immunol. 2019;10(MAR):1–21. 142. Guo H, Xie M, Zhou C, Zheng M. The relevance of pyroptosis in the pathogenesis of liver diseases. Life Sci. 2019 Apr;223:69–73. 143. Fang Y, Tian S, Pan Y, Li W, Wang Q, Tang Y, et al. Pyroptosis: A new frontier in cancer. Biomed Pharmacother. 2020 Jan;121:109595. 144. Celec P, Vlková B, Lauková L, Bábíčková J, Boor P. Cell-free DNA: the role in pathophysiology and as a biomarker in kidney diseases. Expert Rev Mol Med. 2018;20:1–14. 145. Teo YV, Capri M, Morsiani C, Pizza G, Faria AMC, Franceschi C, et al. Cellfree DNA as a biomarker of aging. Aging Cell. 2019;18(1):1–14. 146. Hummel EM, Hessas E, Müller S, Beiter T, Fisch M, Eibl A, et al. Cell-free DNA release under psychosocial and physical stress conditions. Transl Psychiatry. 2018. 147. Vittori LN, Tarozzi A, Latessa PM. Circulating cell-free DNA in physical activities. Methods Mol Biol. 2019;1909:183–97. 148. García Moreira V, De La Cera Martínez T, Gago González E, Prieto García B, Alvarez Menéndez F V. Increase in and clearance of cell-free plasma DNA in hemodialysis quantified by real-time PCR. Clin Chem Lab Med. 2006;44(12):1410–5. 149. Bettegowda C, Sausen M, Leary RJ, Kinde I, Wang Y, Agrawal N, et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med. 2014 Feb;6(224):224ra24. 150. Keller L, Belloum Y, Wikman H, Pantel K. Clinical relevance of blood-based ctDNA analysis: mutation detection and beyond. Br J Cancer. 2020;(November 2019). PATRICIA MONDELO MACÍA 336 151. Diehl F, Schmidt K, Choti MA, Romans K, Goodman S, Li M, et al. Circulating mutant DNA to assess tumor dynamics. Nat Med. 2008 Sep;14(9):985–90. 152. Vidal J, Muinelo L, Dalmases A, Jones F, Edelstein D, Iglesias M, et al. Plasma ctDNA RAS mutation analysis for the diagnosis and treatment monitoring of metastatic colorectal cancer patients. Ann Oncol. 2017;28(6):1325–32. 153. Kilgour E, Rothwell DG, Brady G, Dive C. Liquid Biopsy-Based Biomarkers of Treatment Response and Resistance. Cancer Cell. 2020;37(4):485–95. 154. Elazezy M, Joosse SA. Techniques of using circulating tumor DNA as a liquid biopsy component in cancer management. Comput Struct Biotechnol J. 2018;16:370–8. 155. Kwapisz D. The first liquid biopsy test approved. Is it a new era of mutation testing for non-small cell lung cancer? Ann Transl Med. 2017 Feb;5(3):46. 156. Thress KS, Brant R, Carr TH, Dearden S, Jenkins S, Brown H, et al. EGFR mutation detection in ctDNA from NSCLC patient plasma: A cross-platform comparison of leading technologies to support the clinical development of AZD9291. Lung Cancer. 2015 Dec;90(3):509–15. 157. Bando H, Kagawa Y, Kato T, Akagi K, Denda T, Nishina T, et al. A multicentre, prospective study of plasma circulating tumour DNA test for detecting RAS mutation in patients with metastatic colorectal cancer. Br J Cancer. 2019 May;120(10):982–6. 158. García-Foncillas J, Tabernero J, Élez E, Aranda E, Benavides M, Camps C, et al. Prospective multicenter real-world RAS mutation comparison between OncoBEAM-based liquid biopsy and tissue analysis in metastatic colorectal cancer. Br J Cancer. 2018 Dec;119(12):1464–70. 159. Ou C-Y, Vu T, Grunwald JT, Toledano M, Zimak J, Toosky M, et al. An ultrasensitive test for profiling circulating tumor DNA using integrated comprehensive droplet digital detection. Lab Chip. 2019 Mar;19(6):993–1005. 160. Chen M, Zhao H. Next-generation sequencing in liquid biopsy: cancer screening and early detection. Hum Genomics. 2019 Aug;13(1):34. 161. Malapelle U, Sirera R, Jantus-Lewintre E, Reclusa P, Calabuig-Fariñas S, Blasco A, et al. Profile of the Roche cobas® EGFR mutation test v2 for non-small cell lung cancer. Expert Rev Mol Diagn. 2017 Mar;17(3):209–15. 162. Lamb YN, Dhillon S. Epi proColon(®) 2.0 CE: A Blood-Based Screening Test for Colorectal Cancer. Mol Diagn Ther. 2017 Apr;21(2):225–32. 163. André F, Ciruelos E, Rubovszky G, Campone M, Loibl S, Rugo HS, et al. Alpelisib for PIK3CA-Mutated, Hormone Receptor-Positive Advanced Breast Cancer. N Engl J Med. 2019 May;380(20):1929–40. References 337 164. Laufer-Geva S, Rozenblum AB, Twito T, Grinberg R, Dvir A, Soussan-Gutman L, et al. The Clinical Impact of Comprehensive Genomic Testing of Circulating Cell-Free DNA in Advanced Lung Cancer. J Thorac Oncol. 2018;13(11):1705– 16. 165. Woodhouse R, Li M, Hughes J, Delfosse D, Skoletsky J, Ma P, et al. Clinical and analytical validation of FoundationOne Liquid CDx, a novel 324-Gene cfDNA-based comprehensive genomic profiling assay for cancers of solid tumor origin. PLoS One. 2020;15(9):e0237802. 166. Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017 Jul;357(6349):409–13. 167. Ashworth TR. A case of cancer in which cells similar to those in the tumours were seen in the blood after death. Aust Med J. 1869;14:146. 168. Perez FM, Yonemoto RH. The enigma of circulating malignant cells. Calif Med. 1962;97:166–9. 169. Alexander RF, Spriggs AI. The differential diagnosis of tumour cells in circulating blood. J Clin Pathol. 1960;13:414–24. 170. ROBERTS S, WATNE A, McGRATH R, McGREW E, COLE WH. Technique and results of isolation of cancer cells from the circulating blood. AMA Arch Surg. 1958 Mar;76(3):334–46. 171. Bidard F-C, Peeters DJ, Fehm T, Nolé F, Gisbert-Criado R, Mavroudis D, et al. Clinical validity of circulating tumour cells in patients with metastatic breast cancer: a pooled analysis of individual patient data. Lancet Oncol. 2014 Apr;15(4):406–14. 172. Bao-Caamano A, Rodriguez-Casanova A, Diaz-Lagares A. Epigenetics of Circulating Tumor Cells in Breast Cancer. Adv Exp Med Biol. 2020;1220:117– 34. 173. Pantel K, Hille C, Scher HI. Circulating Tumor Cells in Prostate Cancer: From Discovery to Clinical Utility. Clin Chem. 2019 Jan;65(1):87–99. 174. Lindsay CR, Blackhall FH, Carmel A, Fernandez-Gutierrez F, Gazzaniga P, Groen HJM, et al. EPAC-lung: pooled analysis of circulating tumour cells in advanced non-small cell lung cancer. Eur J Cancer. 2019;117(December 2018):60–8. 175. Cohen SJ, Punt CJA, Iannotti N, Saidman BH, Sabbath KD, Gabrail NY, et al. Relationship of circulating tumor cells to tumor response, progression-free survival, and overall survival in patients with metastatic colorectal cancer. J Clin Oncol Off J Am Soc Clin Oncol. 2008 Jul;26(19):3213–21. PATRICIA MONDELO MACÍA 338 176. Hamilton G, Rath B. Mesenchymal-epithelial transition and circulating tumor cells in small cell lung cancer. Adv Exp Med Biol. 2017;994:229–45. 177. Allan AL, Vantyghem SA, Tuck AB, Chambers AF, Chin-Yee IH, Keeney M. Detection and quantification of circulating tumor cells in mouse models of human breast cancer using immunomagnetic enrichment and multiparameter flow cytometry. Cytometry A. 2005 May;65(1):4–14. 178. Baccelli I, Schneeweiss A, Riethdorf S, Stenzinger A, Schillert A, Vogel V, et al. Identification of a population of blood circulating tumor cells from breast cancer patients that initiates metastasis in a xenograft assay. Nat Biotechnol. 2013 Jun;31(6):539–44. 179. Pantel K, Speicher MR. The biology of circulating tumor cells. Oncogene. 2016;35(10):1216–24. 180. Genna A, Vanwynsberghe AM, Villard A V, Pottier C, Ancel J, Polette M, et al. EMT-Associated Heterogeneity in Circulating Tumor Cells: Sticky Friends on the Road to Metastasis. Cancers (Basel). 2020 Jun;12(6). 181. Mariscal J, Alonso-Nocelo M, Muinelo-Romay L, Barbazan J, Vieito M, Abalo A, et al. Molecular Profiling of Circulating Tumour Cells Identifies Notch1 as a Principal Regulator in Advanced Non-Small Cell Lung Cancer. Sci Rep. 2016 Nov;6:37820. 182. Abreu M, Cabezas-Sainz P, Alonso-Alconada L, Ferreirós A, Mondelo-Macía P, Lago-Lestón RM, et al. Circulating Tumor Cells Characterization Revealed TIMP1 as a Potential Therapeutic Target in Ovarian Cancer. Cells. 2020 May;9(5). 183. Habli Z, Alchamaa W, Saab R, Kadara H, Khraiche ML. Circulating tumor cell detection technologies and clinical utility: Challenges and opportunities. Cancers (Basel). 2020;12(7):1–30. 184. Alix-Panabieres C, Pantel K. Circulating tumor cells: Liquid biopsy of cancer. Clin Chem. 2013;59(1):110–8. 185. Riethdorf S, O’Flaherty L, Hille C, Pantel K. Clinical applications of the CellSearch platform in cancer patients. Adv Drug Deliv Rev. 2018;125:102–21. 186. Rushton AJ, Nteliopoulos G, Shaw JA, Coombes RC. A Review of Circulating Tumour Cell Enrichment Technologies. Cancers (Basel). 2021 Feb 26;13(5). 187. Grover PK, Cummins AG, Price TJ, Roberts-Thomson IC, Hardingham JE. Circulating tumour cells: the evolving concept and the inadequacy of their enrichment by EpCAM-based methodology for basic and clinical cancer research. Ann Oncol Off J Eur Soc Med Oncol. 2014 Aug;25(8):1506–16. 188. Kowalik A, Kowalewska M, Góźdź S. Current approaches for avoiding the limitations of circulating tumor cells detection methods-implications for References 339 diagnosis and treatment of patients with solid tumors. Transl Res. 2017;185:58- 84.e15. 189. Liu Z, Fusi A, Klopocki E, Schmittel A, Tinhofer I, Nonnenmacher A, et al. Negative enrichment by immunomagnetic nanobeads for unbiased characterization of circulating tumor cells from peripheral blood of cancer patients. J Transl Med. 2011 May 19;9:70. 190. Szczerba BM, Castro-Giner F, Vetter M, Krol I, Gkountela S, Landin J, et al. Neutrophils escort circulating tumour cells to enable cell cycle progression. Nature. 2019;566(7745):553–7. 191. Vona G, Sabile A, Louha M, Sitruk V, Romana S, Schütze K, et al. Isolation by size of epithelial tumor cells : a new method for the immunomorphological and molecular characterization of circulatingtumor cells. Am J Pathol. 2000 Jan;156(1):57–63. 192. Hao S-J, Wan Y, Xia Y-Q, Zou X, Zheng S-Y. Size-based separation methods of circulating tumor cells. Adv Drug Deliv Rev. 2018 Feb;125:3–20. 193. Farace F, Massard C, Vimond N, Drusch F, Jacques N, Billiot F, et al. A direct comparison of CellSearch and ISET for circulating tumour-cell detection in patients with metastatic carcinomas. Br J Cancer. 2011 Sep 6;105(6):847–53. 194. Miller MC, Robinson PS, Wagner C, O’Shannessy DJ. The ParsortixTM Cell Separation System-A versatile liquid biopsy platform. Cytometry A. 2018;93(12):1234–9. 195. Hou HW, Warkiani ME, Khoo BL, Li ZR, Soo RA, Tan DS-W, et al. Isolation and retrieval of circulating tumor cells using centrifugal forces. Sci Rep. 2013;3:1259. 196. Gascoyne PRC, Shim S. Isolation of circulating tumor cells by dielectrophoresis. Cancers (Basel). 2014 Mar 12;6(1):545–79. 197. Nagrath S, Sequist L V., Maheswaran S, Bell DW, Irimia D, Ulkus L, et al. Isolation of rare circulating tumour cells in cancer patients by microchip technology. Nature. 2007;450(7173):1235–9. 198. Dizdar L, Fluegen G, van Dalum G, Honisch E, Neves RP, Niederacher D, et al. Detection of circulating tumor cells in colorectal cancer patients using the GILUPI CellCollector: results from a prospective, single-center study. Mol Oncol. 2019;13(7):1548–58. 199. Fischer JC, Niederacher D, Topp SA, Honisch E, Schumacher S, Schmitz N, et al. Diagnostic leukapheresis enables reliable detection of circulating tumor cells of nonmetastatic cancer patients. Proc Natl Acad Sci U S A. 2013 Oct 8;110(41):16580–5. PATRICIA MONDELO MACÍA 340 200. Valihrach L, Androvic P, Kubista M. Platforms for Single-Cell Collection and Analysis. Int J Mol Sci. 2018 Mar;19(3). 201. Lawson DA, Kessenbrock K, Davis RT, Pervolarakis N, Werb Z. Tumour heterogeneity and metastasis at single-cell resolution. Nat Cell Biol. 2018 Dec;20(12):1349–60. 202. Cortés-Hernández LE, Eslami-S Z, Pantel K, Alix-Panabières C. Molecular and Functional Characterization of Circulating Tumor Cells: From Discovery to Clinical Application. Clin Chem. 2019 Dec 6;66(1):97–104. 203. Allard WJ, Matera J, Miller MC, Repollet M, Connelly MC, Rao C, et al. Tumor cells circulate in the peripheral blood of all major carcinomas but not in healthy subjects or patients with nonmalignant diseases. Clin Cancer Res. 2004;10(20):6897–904. 204. Maheswaran S, Sequist L V., Nagrath S, Ulkus L, Brannigan B, Collura C V., et al. Detection of Mutations in EGFR in Circulating Lung-Cancer Cells. N Engl J Med. 2008 Jul 24;359(4):366–77. 205. Siravegna G, Marsoni S, Siena S, Bardelli A. Integrating liquid biopsies into the management of cancer. Nat Rev Clin Oncol. 2017;14(9):531–48. 206. Hayes DF, Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Miller MC, et al. Circulating tumor cells at each follow-up time point during therapy of metastatic breast cancer patients predict progression-free and overall survival. Clin cancer Res an Off J Am Assoc Cancer Res. 2006 Jul;12(14 Pt 1):4218–24. 207. de Bono JS, Scher HI, Montgomery RB, Parker C, Miller MC, Tissing H, et al. Circulating tumor cells predict survival benefit from treatment in metastatic castration-resistant prostate cancer. Clin cancer Res an Off J Am Assoc Cancer Res. 2008 Oct;14(19):6302–9. 208. Cohen EN, Jayachandran G, Moore RG, Cristofanilli M, Lang JE, Khoury JD, et al. A Multi-Center Clinical Study to Harvest and Characterize Circulating Tumor Cells from Patients with Metastatic Breast Cancer Using the Parsortix(®) PC1 System. Cancers (Basel). 2022 Oct;14(21). 209. Pasini L, Ulivi P. Extracellular vesicles in non-small-cell lung cancer: Functional role and involvement in resistance to targeted treatment and immunotherapy. Vol. 12, Cancers. MDPI AG; 2020. p. 40. 210. Zhang H-G, Grizzle WE. Exosomes and cancer: a newly described pathway of immune suppression. Clin cancer Res an Off J Am Assoc Cancer Res. 2011 Mar;17(5):959–64. 211. Logozzi M, De Milito A, Lugini L, Borghi M, Calabrò L, Spada M, et al. High levels of exosomes expressing CD63 and caveolin-1 in plasma of melanoma patients. PLoS One. 2009;4(4):e5219. References 341 212. Pasini L, Ulivi P. Liquid Biopsy for the Detection of Resistance Mechanisms in NSCLC: Comparison of Different Blood Biomarkers. J Clin Med. 2019 Jul 9;8(7):998. 213. Nilsson RJA, Balaj L, Hulleman E, van Rijn S, Pegtel DM, Walraven M, et al. Blood platelets contain tumor-derived RNA biomarkers. Blood. 2011 Sep;118(13):3680–3. 214. Luo C-L, Xu Z-G, Chen H, Ji J, Wang Y-H, Hu W, et al. LncRNAs and EGFRvIII sequestered in TEPs enable blood-based NSCLC diagnosis. Cancer Manag Res. 2018;10:1449–59. 215. Best MG, Vancura A, Wurdinger T. Platelet RNA as a circulating biomarker trove for cancer diagnostics. J Thromb Haemost. 2017 Jul;15(7):1295–306. 216. Velez G, Nguyen H V, Chemudupati T, Ludwig CA, Toral M, Reddy S, et al. Liquid biopsy proteomics of uveal melanoma reveals biomarkers associated with metastatic risk. Vol. 20, Molecular cancer. 2021. p. 39. 217. Mondelo-Macía P, Lago-Lestón RM, Rodríguez-Casanova A, Abalo A, Díaz- Lagares Á, García-González J, et al. Rapid Idylla mutational testing to detect EGFR mutations in plasma samples and to monitor therapy in advanced nonsmall cell lung cancer patients. Pathology. England; 2023. 218. Singal G, Miller PG, Agarwala V, Li G, Kaushik G, Backenroth D, et al. Association of Patient Characteristics and Tumor Genomics With Clinical Outcomes Among Patients With Non-Small Cell Lung Cancer Using a Clinicogenomic Database. Jama. 2019;321(14):1391–9. 219. Kang Q, Henry NL, Paoletti C, Jiang H, Vats P, Chinnaiyan AM, et al. Comparative analysis of circulating tumor DNA stability In K(3)EDTA, Streck, and CellSave blood collection tubes. Clin Biochem. 2016 Dec;49(18):1354–60. 220. Garcia J, Wozny A-S, Geiguer F, Delherme A, Barthelemy D, Merle P, et al. Profiling of circulating tumor DNA in plasma of non-small cell lung cancer patients, monitoring of epidermal growth factor receptor p.T790M mutated allelic fraction using beads, emulsion, amplification, and magnetics companion assay and evaluation in fut. Cancer Med. 2019 Jul;8(8):3685–97. 221. Zulato E, Tosello V, Nardo G, Bonanno L, Del Bianco P, Indraccolo S. Implementation of Next Generation Sequencing-Based Liquid Biopsy for Clinical Molecular Diagnostics in Non-Small Cell Lung Cancer (NSCLC) Patients. Diagnostics (Basel, Switzerland). 2021 Aug;11(8). 222. Oliveros JC. Venny. An interactive tool for comparing lists with Venn’s diagram. 223. Delgado-Garciá M, Weynand B, Gómez-Izquierdo L, Hernández MJ, Blanco ÁM, Varela M, et al. Clinical performance evaluation of the IdyllaTM EGFR PATRICIA MONDELO MACÍA 342 Mutation Test on formalin-fixed paraffin-embedded tissue of non-small cell lung cancer. BMC Cancer. 2020;20(1):1–10. 224. Momeni-Boroujeni A, Salazar P, Zheng T, Mensah N, Rijo I, Dogan S, et al. Rapid EGFR Mutation Detection Using the Idylla Platform: Single-Institution Experience of 1200 Cases Analyzed by an In-House Developed Pipeline and Comparison with Concurrent Next-Generation Sequencing Results. J Mol Diagnostics. 2021 Mar;23(3):310–22. 225. Gilson P, Saurel C, Salleron J, Husson M, Demange J, Merlin JL, et al. Evaluation of the Idylla ctEGFR mutation assay to detect EGFR mutations in plasma from patients with non-small cell lung cancers. Sci Rep. 2021;11(1):1– 11. 226. Janku F, Huang HJ, Claes B, Falchook GS, Fu S, Hong D, et al. BRAF Mutation testing in cell-free DNA from the plasma of patients with advanced cancers using a rapid, automated molecular diagnostics system. Mol Cancer Ther. 2016;15(6):1397–404. 227. Vivancos A, Aranda E, Benavides M, Élez E, Gómez-España MA, Toledano M, et al. Comparison of the Clinical Sensitivity of the Idylla Platform and the OncoBEAM RAS CRC Assay for KRAS Mutation Detection in Liquid Biopsy Samples. Sci Rep. 2019;9(1):1–8. 228. Finall A, Davies G, Jones T, Emlyn G, Huey P, Mullard A. Integration of rapid PCR testing as an adjunct to NGS in diagnostic pathology services within the UK: evidence from a case series of non-squamous, non-small cell lung cancer (NSCLC) patients with follow-up. J Clin Pathol. 2022 Jan; 229. Heeke S, Hofman V, Benzaquen J, Otto J, Tanga V, Zahaf K, et al. Detection of EGFR Mutations From Plasma of NSCLC Patients Using an Automatic Cartridge-Based PCR System. Front Pharmacol. 2021;12(April):1–5. 230. Mondelo-Macía P, Rodríguez-López C, Valiña L, Aguín S, León-Mateos L, García-González J, et al. Detection of MET Alterations Using Cell Free DNA and Circulating Tumor Cells from Cancer Patients. Cells. 2020 Feb;9(2). 231. Maroun CR, Rowlands T. The Met receptor tyrosine kinase: A key player in oncogenesis and drug resistance. Pharmacol Ther. 2014;142(3):316–38. 232. Mo H-N, Liu P. Targeting MET in cancer therapy. Chronic Dis Transl Med. 2017;3(3):148–53. 233. Trusolino L, Bertotti A, Comoglio PM. MET signalling: Principles and functions in development, organ regeneration and cancer. Nat Rev Mol Cell Biol. 2010;11(12):834–48. References 343 234. Drilon A, Cappuzzo F, Ou SHI, Camidge DR. Targeting MET in Lung Cancer: Will Expectations Finally Be MET? Vol. 12, Journal of Thoracic Oncology. 2017. p. 15–26. 235. Tovar EA, Graveel CR. MET in human cancer: germline and somatic mutations. Ann Transl Med. 2017 May;5(10):205–205. 236. Bardelli A, Corso S, Bertotti A, Hobor S, Valtorta E, Siravegna G, et al. Amplification of the MET receptor drives resistance to anti-EGFR therapies in colorectal cancer. Cancer Discov. 2013;3(6):658–73. 237. Duplaquet L, Kherrouche Z, Baldacci S, Jamme P, Cortot AB, Copin M-C, et al. The multiple paths towards MET receptor addiction in cancer. Oncogene. 2018 Jun 19;37(24):3200–15. 238. Kwak EL, Ahronian LG, Siravegna G, Mussolin B, Godfrey JT, Clark JW, et al. Molecular heterogeneity and receptor coamplification drive resistance to targeted therapy in MET-Amplified esophagogastric cancer. Cancer Discov. 2015;5(12):1271–81. 239. Kwak Y, Kim S, Park C, Paek SH, Lee S, Park S. C-MET overexpression and amplification in gliomas. 2015;8(11):14932–8. 240. Jia L, Yang X, Tian W, Guo S, Huang W, Zhao W. Increased Expression of c- Met is Associated with Chemotherapy-Resistant Breast Cancer and Poor Clinical Outcome. Med Sci Monit. 2018 Nov 16;24:8239–49. 241. Li Y, Li W, He Q, Xu Y, Ren X, Tang X, et al. Prognostic value of MET protein overexpression and gene amplification in locoregionally advanced nasopharyngeal carcinoma. Oncotarget. 2015;6(15). 242. Zhang Y, Du Z, Zhang M. Biomarker development in MET-targeted therapy. Oncotarget. 2016;7(24). 243. De Bacco F, Luraghi P, Medico E, Reato G, Girolami F, Perera T, et al. Induction of MET by ionizing radiation and its role in radioresistance and invasive growth of cancer. J Natl Cancer Inst. 2011;103(8):645–61. 244. Lee M, Jain P, Wang F, Ma PC, Borczuk A, Halmos B. MET alterations and their impact on the future of non-small cell lung cancer (NSCLC) targeted therapies. Expert Opin Ther Targets. 2021;25(4):249–68. 245. Parikh PK, Ghate MD. Recent advances in the discovery of small molecule c- Met Kinase inhibitors. Eur J Med Chem. 2018 Jan 1;143:1103–38. 246. Ilie M, Hofman V, Long E, Bordone O, Selva E, Washetine K, et al. Current challenges for detection of circulating tumor cells and cell-free circulating nucleic acids, and their characterization in non-small cell lung carcinoma patients. What is the best blood substrate for personalized medicine? Ann Transl Med. 2014 Nov;2(11):107. PATRICIA MONDELO MACÍA 344 247. Park S, Choi Y La, Sung CO, An J, Seo J, Ahn MJ, et al. High MET copy number and MET overexpression: Poor outcome in non-small cell lung cancer patients. Histol Histopathol. 2012;27(2):197–207. 248. Cappuzzo F, Marchetti A, Skokan M, Rossi E, Gajapathy S, Felicioni L, et al. Increased MET gene copy number negatively affects survival of surgically resected non-small-cell lung cancer patients. J Clin Oncol. 2009;27(10):1667– 74. 249. Kitz J, Lowes LE, Goodale D, Allan AL. Circulating Tumor Cell Analysis in Preclinical Mouse Models of Metastasis. Diagnostics (Basel, Switzerland). 2018 Apr;8(2). 250. Alunni-Fabbroni M, Sandri MT. Circulating tumour cells in clinical practice: Methods of detection and possible characterization. Methods. 2010. p. 289–97. 251. Marrinucci D, Bethel K, Bruce RH, Curry DN, Hsieh B, Humphrey M, et al. Case study of the morphologic variation of circulating tumor cells. Hum Pathol. 2007;38(3):514–9. 252. Zhang T, Boominathan R, Foulk B, Rao C, Kemeny G, Strickler JH, et al. Development of a Novel c-MET-Based CTC Detection Platform. Mol Cancer Res. 2016;14(6):539–47. 253. Ilie M, Szafer-glusman E, Hofman V, Long-mira E. Expression of MET in circulating tumor cells correlates with expression in tumor tissue from advancedstage lung cancer patients. 2017;8(16):26112–21. 254. Zhang Y, Tang ET, Du Z. Detection of MET gene copy number in cancer samples using the droplet digital PCR method. PLoS One. 2016;11(1):1–8. 255. Ikeda S, Schwaederle M, Mohindra M, Fontes Jardim DL, Kurzrock R. MET alterations detected in blood-derived circulating tumor DNA correlate with bone metastases and poor prognosis. J Hematol Oncol. 2018;11(1):76. 256. Wang Q, Yang S, Wang K, Sun SY. MET inhibitors for targeted therapy of EGFR TKI-resistant lung cancer. Vol. 12, Journal of Hematology and Oncology. 2019. 257. Shi P, Oh YT, Zhang G, Yao W, Yue P, Li Y, et al. Met gene amplification and protein hyperactivation is a mechanism of resistance to both first and third generation EGFR inhibitors in lung cancer treatment. Cancer Lett. 2016;380(2):494–504. 258. Kang J, Chen HJ, Wang Z, Liu J, Li B, Zhang T, et al. Osimertinib and Cabozantinib Combinatorial Therapy in an EGFR-Mutant Lung Adenocarcinoma Patient with Multiple MET Secondary-Site Mutations after Resistance to Crizotinib. J Thorac Oncol. 2018;13(4):e49–53. References 351 326. Herath S, Sadeghi Rad H, Radfar P, Ladwa R, Warkiani M, O’Byrne K, et al. The Role of Circulating Biomarkers in Lung Cancer. Front Oncol. 2021;11:801269. 327. Cheng ML, Pectasides E, Hanna GJ, Parsons HA, Choudhury AD, Oxnard GR. Circulating tumor DNA in advanced solid tumors: Clinical relevance and future directions. CA Cancer J Clin. 2021;71(2):176–90. 328. García-Pardo M, Makarem M, Li JJN, Kelly D, Leighl NB. Integrating circulating-free DNA (cfDNA) analysis into clinical practice: opportunities and challenges. Br J Cancer. 2022 Mar; 329. Mathios D, Johansen JS, Cristiano S, Medina JE, Phallen J, Larsen KR, et al. Detection and characterization of lung cancer using cell-free DNA fragmentomes. Nat Commun. 2021 Aug;12(1):5060. 330. Romero A, Nadal E, Serna R, Insa A, Campelo MRG, Benito C, et al. OA20.02 Pre-Treatment Levels of ctDNA for Long-Term Survival Prediction in Stage IIIA NSCLC Treated With Neoadjuvant Chemo-Immunotherapy. J Thorac Oncol. 2021;16(10):S883–4. 331. Moding EJ, Liu Y, Nabet BY, Chabon JJ, Chaudhuri AA, Hui AB, et al. Circulating tumor DNA dynamics predict benefit from consolidation immunotherapy in locally advanced non-small-cell lung cancer. Nat Cancer. 2020;1:176–183. 332. Cheng ML, Lau CJ, Milan MSD, Supplee JG, Riess JW, Bradbury PA, et al. Plasma ctDNA Response Is an Early Marker of Treatment Effect in Advanced NSCLC. JCO Precis Oncol. 2021;5. 333. Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Matera J, Miller MC, et al. Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med. 2004;351(8):781–91. 334. Naito T, Tanaka F, Ono A, Yoneda K, Takahashi T, Murakami H, et al. Prognostic impact of circulating tumor cells in patients with small cell lung cancer. J Thorac Oncol. 2012;7(3):512–9. 335. Igawa S, Gohda K, Fukui T, Ryuge S, Otani S, Masago A, et al. Circulating tumor cells as a prognostic factor in patients with small cell lung cancer. Oncol Lett. 2014 May;7(5):1469–73. 336. Cheng Y, Liu XQ, Fan Y, Liu YP, Liu Y, Liu Y, et al. Circulating tumor cell counts/change for outcome prediction in patients with extensive-stage small-cell lung cancer. Futur Oncol. 2016;12(6):789–99. 337. Salgia R, Weaver RW, McCleod M, Stille JR, Yan SB, Roberson S, et al. Prognostic and predictive value of circulating tumor cells and CXCR4 expression as biomarkers for a CXCR4 peptide antagonist in combination with PATRICIA MONDELO MACÍA 352 carboplatin-etoposide in small cell lung cancer: exploratory analysis of a phase II study. Invest New Drugs. 2017;35(3):334–44. 338. Aggarwal C, Wang X, Ranganathan A, Torigian D, Troxel A, Evans T, et al. Circulating tumor cells as a predictive biomarker in patients with small cell lung cancer undergoing chemotherapy. Lung Cancer. 2017 Oct;112:118–25. 339. Radfar P, Aboulkheyr Es H, Salomon R, Kulasinghe A, Ramalingam N, Sarafraz-Yazdi E, et al. Single-cell analysis of circulating tumour cells: enabling technologies and clinical applications. Trends Biotechnol. 2022 Mar; 340. van Dessel LF, Beije N, Helmijr JCA, Vitale SR, Kraan J, Look MP, et al. Application of circulating tumor DNA in prospective clinical oncology trials - standardization of preanalytical conditions. Mol Oncol. 2017 Mar;11(3):295– 304. 341. Akaike H. A New Look at the Statistical Model Identification. In 1974. p. 215– 22. 342. Hadley W. Ggplot2: Elegrant graphics for data analysis. Springer; 2016. 343. Sjoberg, Daniel D, Whiting K, Curry M, Lavery, Jessica A, Larmarange J. Reproducible Summary Tables with the gtsummary Package. R J. 2021;13(1):570. 344. Oliveira KCS, Ramos IB, Silva JMC, Barra WF, Riggins GJ, Palande V, et al. Current Perspectives on Circulating Tumor DNA, Precision Medicine, and Personalized Clinical Management of Cancer. Mol Cancer Res. 2020 Apr;18(4):517–28. 345. Schwendenwein A, Megyesfalvi Z, Barany N, Valko Z, Bugyik E, Lang C, et al. Molecular profiles of small cell lung cancer subtypes: therapeutic implications. Mol Ther - Oncolytics. 2021;20(March):470–83. 346. Yang S, Zhang Z, Wang Q. Emerging therapies for small cell lung cancer. J Hematol Oncol. 2019 May;12(1):47. 347. Leon SA, Shapiro B, Sklaroff DM, Yaros MJ. Free DNA in the serum of cancer patients and the effect of therapy. Cancer Res. 1977 Mar;37(3):646–50. 348. van der Drift MA, Hol BEA, Klaassen CHW, Prinsen CFM, van Aarssen YAWG, Donders R, et al. Circulating DNA is a non-invasive prognostic factor for survival in non-small cell lung cancer. Lung Cancer. 2010 May;68(2):283– 7. 349. Almodovar K, Iams WT, Meador CB, Zhao Z, York S, Horn L, et al. Longitudinal cell-free DNA analysis in patients with small cell lung cancer reveals dynamic insights into treatment efficacy and disease relapse. 2018;13(1):112–23. References 353 350. Nong J, Gong Y, Guan Y, Yi X, Yi Y, Chang L, et al. Circulating tumor DNA analysis depicts subclonal architecture and genomic evolution of small cell lung cancer. Nat Commun. 2018;9(1):1–8. 351. Du M, Thompson J, Fisher H, Zhang P, Huang CC, Wang L. Genomic alterations of plasma cell-free DNAs in small cell lung cancer and their clinical relevance. Lung Cancer]. 2018;120(April):113–21. 352. Mohan S, Foy V, Ayub M, Leong HS, Schofield P, Sahoo S, et al. Profiling of Circulating Free DNA Using Targeted and Genome-wide Sequencing in Patients with SCLC. J Thorac Oncol. 2020;15(2):216–30. 353. Iams WT, Kopparapu PR, Yan Y, Muterspaugh A, Zhao Z, Chen H, et al. Blood- Based Surveillance Monitoring of Circulating Tumor DNA From Patients With SCLC Detects Disease Relapse and Predicts Death in Patients With Limited- Stage Disease. JTO Clin Res Reports. 2020;1(2):100024. 354. Devarakonda S, Sankararaman S, Herzog BH, Gold KA, Owonikoko TK, Li BT, et al. Circulating Tumor DNA Profiling in Small Cell Lung Cancer Identifies Potentially Targetable Alterations. 2019;25(20):6119–26. 355. Hou JM, Greystoke A, Lancashire L, Cummings J, Ward T, Board R, et al. Evaluation of circulating tumor cells and serological cell death biomarkers in small cell lung cancer patients undergoing chemotherapy. Am J Pathol. 2009;175(2):808–16. 356. Hou JM, Krebs MG, Lancashire L, Sloane R, Backen A, Swain RK, et al. Clinical significance and molecular characteristics of circulating tumor cells and circulating tumor microemboli in patients with small-cell lung cancer. J Clin Oncol. 2012;30(5):525–32. 357. Riley RS, June CH, Langer R, Mitchell MJ. Delivery technologies for cancer immunotherapy. Nat Rev Drug Discov. 2019;18(3):175–96. 358. Bagchi S, Yuan R, Engleman EG. Immune Checkpoint Inhibitors for the Treatment of Cancer: Clinical Impact and Mechanisms of Response and Resistance. Annu Rev Pathol Mech Dis. 2021 Jan 24;16(1):223–49. 359. Spigel D, de Marinis F, Giaccone G, Reinmuth N, Vergnenegre A, Barrios CH, et al. IMpower110: Interim overall survival (OS) analysis of a phase III study of atezolizumab (atezo) vs platinum-based chemotherapy (chemo) as first-line (1L) treatment (tx) in PD-L1-selected NSCLC. Ann Oncol. 2019 Oct;30:v915. 360. Socinski MA, Jotte RM, Cappuzzo F, Orlandi F, Stroyakovskiy D, Nogami N, et al. Atezolizumab for First-Line Treatment of Metastatic Nonsquamous NSCLC. N Engl J Med. 2018;378(24). 361. Kennedy LB, Salama AKS. A review of cancer immunotherapy toxicity. CA Cancer J Clin. 2020;70(2):86–104. PATRICIA MONDELO MACÍA 354 362. Shen P, Han L, Ba X, Qin K, Tu S. Hyperprogressive Disease in Cancers Treated With Immune Checkpoint Inhibitors. Front Pharmacol. 2021;12(July):1–13. 363. Hofman P, Heeke S, Alix-Panabières C, Pantel K. Liquid biopsy in the era of immuno-oncology: Is it ready for prime-time use for cancer patients? Ann Oncol. 2019;30(9):1448–59. 364. Janse van Rensburg HJ, Spiliopoulou P, Siu LL. Circulating Biomarkers for Therapeutic Monitoring of Anti-cancer Agents. Oncologist. 2022 May 6;27(5):352–62. 365. Babačić H, Lehtiö J, Pico De Coaña Y, Pernemalm M, Eriksson H. In-depth plasma proteomics reveals increase in circulating PD-1 during anti-PD-1 immunotherapy in patients with metastatic cutaneous melanoma. J Immunother Cancer. 2020;8(1). 366. Altelaar AFM, Munoz J, Heck AJR. Next-generation proteomics: towards an integrative view of proteome dynamics. Nat Rev Genet. 2013 Jan 4;14(1):35– 48. 367. Ludwig C, Gillet L, Rosenberger G, Amon S, Collins BC, Aebersold R. Dataindependent acquisition‐based SWATH ‐ MS for quantitative proteomics: a tutorial . Mol Syst Biol. 2018;14(8):1–23. 368. de Jesus JR, da Silva Fernandes R, de Souza Pessôa G, Raimundo IM, Arruda MAZ. Depleting high-abundant and enriching low-abundant proteins in human serum: An evaluation of sample preparation methods using magnetic nanoparticle, chemical depletion and immunoaffinity techniques. Talanta. 2017;170(March):199–209. 369. Fernández C, Santos HM, Ruíz-Romero C, Blanco FJ, Capelo-Martínez JL. A comparison of depletion versus equalization for reducing high-abundance proteins in human serum. Electrophoresis. 2011;32(21):2966–74. 370. Bonzon-Kulichenko E, Pérez-Hernández D, Núñez E, Martínez-Acedo P, Navarro P, Trevisan-Herraz M, et al. A robust method for quantitative highthroughput analysis of proteomes by 18O labeling. Mol Cell Proteomics. 2011;10(1):1–14. 371. Perez-Hernandez D, Gutiérrez-Vázquez C, Jorge I, López-Martín S, Ursa A, Sánchez-Madrid F, et al. The intracellular interactome of tetraspanin-enriched microdomains reveals their function as sorting machineries toward exosomes. J Biol Chem. 2013;288(17):11649–61. 372. Shilov I V., Seymourt SL, Patel AA, Loboda A, Tang WH, Keating SP, et al. The paragon algorithm, a next generation search engine that uses sequence temperature values sequence temperature values and feature probabilities to identify peptides from tandem mass spectra. Mol Cell Proteomics. 2007;6(9):1638–55. References 355 373. Luo Y, Mok TS, Lin X, Zhang W, Cui Y, Guo J, et al. SWATH-based proteomics identified carbonic anhydrase 2 as a potential diagnosis biomarker for nasopharyngeal carcinoma. Sci Rep. 2017;7(January):1–11. 374. Meyer JG, Schilling B. Clinical applications of quantitative proteomics using targeted and untargeted data-independent acquisition techniques. Expert Rev Proteomics. 2017 May 4;14(5):419–29. 375. Ortea I, Ruiz-Sánchez I, Cañete R, Caballero-Villarraso J, Cañete MD. Identification of candidate serum biomarkers of childhood-onset growth hormone deficiency using SWATH-MS and feature selection. J Proteomics. 2018;175(2017):105–13. 376. Tan HT, Chung MCM. Label-Free Quantitative Phosphoproteomics Reveals Regulation of Vasodilator-Stimulated Phosphoprotein upon Stathmin-1 Silencing in a Pair of Isogenic Colorectal Cancer Cell Lines. Proteomics. 2018;18(8):1–13. 377. McGurk KA, Dagliati A, Chiasserini D, Lee D, Plant D, Baricevic-Jones I, et al. The use of missing values in proteomic data-independent acquisition mass spectrometry to enable disease activity discrimination. Bioinformatics. 2020 Apr;36(7):2217–23. 378. Jin L, Bi Y, Hu C, Qu J, Shen S, Wang X, et al. A comparative study of evaluating missing value imputation methods in label-free proteomics. Sci Rep. 2021 Jan;11(1):1760. 379. Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, et al. Metascape provides a biologist-oriented resource for the analysis of systemslevel datasets. Nat Commun. 2019 Apr 3;10(1):1523. 380. Liebermeister W, Noor E, Flamholz A, Davidi D, Bernhardt J, Milo R. Visual account of protein investment in cellular functions. Proc Natl Acad Sci. 2014 Jun 10;111(23):8488–93. 381. Davis AA, Patel VG. The role of PD-L1 expression as a predictive biomarker: an analysis of all US Food and Drug Administration (FDA) approvals of immune checkpoint inhibitors. J Immunother Cancer. 2019 Dec 26;7(1):278. 382. Banchereau R, Leng N, Zill O, Sokol E, Liu G, Pavlick D, et al. Molecular determinants of response to PD-L1 blockade across tumor types. Nat Commun. 2021 Dec 25;12(1):3969. 383. Jardim DL, Goodman A, de Melo Gagliato D, Kurzrock R. The Challenges of Tumor Mutational Burden as an Immunotherapy Biomarker. Cancer Cell. 2021 Feb;39(2):154–73. PATRICIA MONDELO MACÍA 356 384. Chan TA, Yarchoan M, Jaffee E, Swanton C, Quezada SA, Stenzinger A, et al. Development of tumor mutation burden as an immunotherapy biomarker: utility for the oncology clinic. Ann Oncol. 2019 Jan;30(1):44–56. 385. Yanagawa N, Yamada N, Sugimoto R, Osakabe M, Uesugi N, Shiono S, et al. The Frequency of DNA Mismatch Repair Deficiency Is Very Low in Surgically Resected Lung Carcinoma. Front Oncol. 2021 Oct 6;11. 386. Kwon YW, Jo H-S, Bae S, Seo Y, Song P, Song M, et al. Application of Proteomics in Cancer: Recent Trends and Approaches for Biomarkers Discovery. Front Med. 2021 Sep 22;8. 387. Chantada-Vázquez M del P, Conde-Amboage M, Graña-López L, Vázquez- Estévez S, Bravo SB, Núñez C. Circulating Proteins Associated with Response and Resistance to Neoadjuvant Chemotherapy in HER2-Positive Breast Cancer. Cancers (Basel). 2022 Feb 21;14(4):1087. 388. Gómez BB, López-Cortés R, Casas-Nebra FJ, Vázquez-Estévez S, Pérez-Fentes D, Chantada-Vázquez M del P, et al. Detection of Circulating Serum Protein Biomarkers of Non-Muscle Invasive Bladder Cancer after Protein Corona-Silver Nanoparticles Analysis by SWATH-MS. Nanomaterials. 2021 Sep 13;11(9):2384. 389. Janacova L, Faktor J, Capkova L, Paralova V, Pospisilova A, Podhorec J, et al. SWATH-MS Analysis of FFPE Tissues Identifies Stathmin as a Potential Marker of Endometrial Cancer in Patients Exposed to Tamoxifen. J Proteome Res. 2020 Jul 2;19(7):2617–30. 390. Harel M, Ortenberg R, Varanasi SK, Mangalhara KC, Mardamshina M, Markovits E, et al. Proteomics of Melanoma Response to Immunotherapy Reveals Mitochondrial Dependence. Cell. 2019;179(1):236-250.e18. 391. Coscia F, Lengyel E, Duraiswamy J, Ashcroft B, Bassani-Sternberg M, Wierer M, et al. Multi-level Proteomics Identifies CT45 as a Chemosensitivity Mediator and Immunotherapy Target in Ovarian Cancer. Cell. 2018;175(1):159-170.e16. 392. Yu KH, Levine DA, Zhang H, Chan DW, Zhang Z, Snyder M. Predicting ovarian cancer patients’ clinical response to platinum-based chemotherapy by their tumor proteomic signatures. J Proteome Res. 2016;15(8):2455–65. 393. Pereira-Veiga T, Bravo S, Gómez-Tato A, Yáñez-Gómez C, Abuín C, Varela V, et al. Red Blood Cells Protein Profile Is Modified in Breast Cancer Patients. Mol Cell Proteomics. 2022 Dec;21(12):100435. 394. Hanash SM, Pitteri SJ, Faca VM. Mining the plasma proteome for cancer biomarkers. Nature. 2008 Apr 2;452(7187):571–9. References 357 395. Maeda S, Yamamoto H, Kinch LN, Garza CM, Takahashi S, Otomo C, et al. Structure, lipid scrambling activity and role in autophagosome formation of ATG9A. Nat Struct Mol Biol. 2020 Dec 26;27(12):1194–201. 396. Huyghe J, Priem D, Van Hove L, Gilbert B, Fritsch J, Uchiyama Y, et al. ATG9A prevents TNF cytotoxicity by an unconventional lysosomal targeting pathway. Science (80- ). 2022 Dec 16;378(6625):1201–7. 397. Girard M, Bizet AA, Lachaux A, Gonzales E, Filhol E, Collardeau-Frachon S, et al. DCDC2 Mutations Cause Neonatal Sclerosing Cholangitis. Hum Mutat. 2016 Oct;37(10):1025–9. 398. Grati M, Chakchouk I, Ma Q, Bensaid M, Desmidt A, Turki N, et al. A missense mutation in DCDC2 causes human recessive deafness DFNB66, likely by interfering with sensory hair cell and supporting cell cilia length regulation. Hum Mol Genet. 2015 May 1;24(9):2482–91. 399. Schueler M, Braun DA, Chandrasekar G, Gee HY, Klasson TD, Halbritter J, et al. DCDC2 mutations cause a renal-hepatic ciliopathy by disrupting Wnt signaling. Am J Hum Genet. 2015 Jan 8;96(1):81–92. 400. Huizing M, Hess R, Dorward H, Claassen DA, Helip-Wooley A, Kleta R, et al. Cellular, Molecular and Clinical Characterization of Patients with Hermansky- Pudlak Syndrome Type 5. Traffic. 2004 Sep;5(9):711–22. 401. Kwon M, Hanna E, Lorang D, He M, Quick JS, Adem A, et al. Functional Characterization of Filamin A Interacting Protein 1–Like, a Novel Candidate for Antivascular Cancer Therapy. Cancer Res. 2008 Sep 15;68(18):7332–41. 402. Mok SC, Wong K-K, Chan RKW, Lau CC, Tsao S-W, Knapp RC, et al. Molecular Cloning of Differentially Expressed Genes in Human Epithelial Ovarian Cancer. Gynecol Oncol. 1994 Feb;52(2):247–52. 403. Seo S, Zhang Q, Bugge K, Breslow DK, Searby CC, Nachury M V., et al. A Novel Protein LZTFL1 Regulates Ciliary Trafficking of the BBSome and Smoothened. Barsh GS, editor. PLoS Genet. 2011 Nov 3;7(11):e1002358. 404. Wei Q, Zhou W, Wang W, Gao B, Wang L, Cao J, et al. Tumor-Suppressive Functions of Leucine Zipper Transcription Factor–Like 1. Cancer Res. 2010 Apr 1;70(7):2942–50. 405. Farnsworth CC, Seabra MC, Ericsson LH, Gelb MH, Glomset JA. Rab geranylgeranyl transferase catalyzes the geranylgeranylation of adjacent cysteines in the small GTPases Rab1A, Rab3A, and Rab5A. Proc Natl Acad Sci. 1994 Dec 6;91(25):11963–7. 406. Jia W, Gao Q, Han A, Zhu H, Yu J. The potential mechanism, recognition and clinical significance of tumor pseudoprogression after immunotherapy. Cancer Biol Med. 2019 Nov;16(4):655–70. PATRICIA MONDELO MACÍA 358 407. Society AC. Global Cancer Facts & Figures 4th Edition. Vol. 29, Atlanta: American Cancer Society. 2018. 138–144 p. 408. Sociedad Española de Oncología Médica (SEOM). Las cifras del cáncer en España 2023. 2023. 409. König IR, Fuchs O, Hansen G, von Mutius E, Kopp M V. What is precision medicine? Eur Respir J. 2017 Oct;50(4). 410. Cohen JD, Li L, Wang Y, Thoburn C, Afsari B, Danilova L, et al. Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science. 2018 Feb;359(6378):926–30. 411. Alix-Panabières C, Pantel K. Liquid biopsy: From discovery to clinical application. Cancer Discov. 2021;11(4):858–73. 412. Reckamp KL, Melnikova VO, Karlovich C, Sequist L V, Camidge DR, Wakelee H, et al. A Highly Sensitive and Quantitative Test Platform for Detection of NSCLC EGFR Mutations in Urine and Plasma. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer. 2016 Oct;11(10):1690–700. 413. Villatoro S, Mayo-de-Las-Casas C, Jordana-Ariza N, Viteri-Ramírez S, Garzón- Ibañez M, Moya-Horno I, et al. Prospective detection of mutations in cerebrospinal fluid, pleural effusion, and ascites of advanced cancer patients to guide treatment decisions. Mol Oncol. 2019 Dec;13(12):2633–45. 414. Zheng M-M, Li Y-S, Tu H-Y, Jiang B-Y, Yang J-J, Zhou Q, et al. Genotyping of Cerebrospinal Fluid Associated With Osimertinib Response and Resistance for Leptomeningeal Metastases in EGFR-Mutated NSCLC. J Thorac Oncol Off Publ Int Assoc Study Lung Cancer. 2021 Feb;16(2):250–8. 415. Merker JD, Oxnard GR, Compton C, Diehn M, Hurley P, Lazar AJ, et al. Circulating Tumor DNA Analysis in Patients With Cancer: American Society of Clinical Oncology and College of American Pathologists Joint Review. J Clin Oncol Off J Am Soc Clin Oncol. 2018 Jun;36(16):1631–41. 416. Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland C, Joon OP, et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science (80- ). 2007;316(5827):1039–43. 417. Váraljai R, Elouali S, Lueong SS, Wistuba-Hamprecht K, Seremet T, Siveke JT, et al. The predictive and prognostic significance of cell-free DNA concentration in melanoma. J Eur Acad Dermatol Venereol. 2021 Feb;35(2):387–95. 418. Papadimitriou M-A, Levis P, Kotronopoulos G, Stravodimos K, Avgeris M, Scorilas A. Preoperative Cell-Free DNA (cfDNA) in Muscle-Invasive Bladder Cancer Treatment Outcome. Clin Chem. 2023 Feb; 419. Valpione S, Gremel G, Mundra P, Middlehurst P, Galvani E, Girotti MR, et al. Plasma total cell-free DNA (cfDNA) is a surrogate biomarker for tumour burden References 359 and a prognostic biomarker for survival in metastatic melanoma patients. Eur J Cancer. 2018 Jan;88:1–9. 420. Hamfjord J, Guren TK, Dajani O, Johansen JS, Glimelius B, Sorbye H, et al. Total circulating cell-free DNA as a prognostic biomarker in metastatic colorectal cancer before first-line oxaliplatin-based chemotherapy. Ann Oncol Off J Eur Soc Med Oncol. 2019 Jul;30(7):1088–95. 421. Fettke H, Kwan EM, Bukczynska P, Ng N, Nguyen-Dumont T, Southey MC, et al. Prognostic Impact of Total Plasma Cell-free DNA Concentration in Androgen Receptor Pathway Inhibitor-treated Metastatic Castration-resistant Prostate Cancer. Eur Urol Focus. 2021 Nov;7(6):1287–91. 422. Gristina V, Barraco N, La Mantia M, Castellana L, Insalaco L, Bono M, et al. Clinical Potential of Circulating Cell-Free DNA (cfDNA) for Longitudinally Monitoring Clinical Outcomes in the First-Line Setting of Non-Small-Cell Lung Cancer (NSCLC): A Real-World Prospective Study. Cancers (Basel). 2022 Dec;14(23). 423. Keller L, Pantel K. Unravelling tumour heterogeneity by single-cell profiling of circulating tumour cells. Nat Rev Cancer. 2019 Oct;19(10):553–67. 424. Krebs MG, Sloane R, Priest L, Lancashire L, Hou J-M, Greystoke A, et al. Evaluation and prognostic significance of circulating tumor cells in patients with non-small-cell lung cancer. J Clin Oncol Off J Am Soc Clin Oncol. 2011 Apr;29(12):1556–63. 425. Alix-Panabières C, Schwarzenbach H, Pantel K. Circulating tumor cells and circulating tumor DNA. Annu Rev Med. 2012;63:199–215. 426. Mego M, De Giorgi U, Dawood S, Wang X, Valero V, Andreopoulou E, et al. Characterization of metastatic breast cancer patients with nondetectable circulating tumor cells. Int J cancer. 2011 Jul;129(2):417–23. 427. Shi Z-M, Wang L, Shen H, Jiang C-F, Ge X, Li D-M, et al. Downregulation of miR-218 contributes to epithelial-mesenchymal transition and tumor metastasis in lung cancer by targeting Slug/ZEB2 signaling. Oncogene. 2017 May;36(18):2577–88. 428. Pantazaka E, Vardas V, Roumeliotou A, Kakavogiannis S, Kallergi G. Clinical Relevance of Mesenchymal- and Stem-Associated Phenotypes in Circulating Tumor Cells Isolated from Lung Cancer Patients. Cancers (Basel). 2021 Apr;13(9). 429. Bortolini Silveira A, Bidard F-C, Tanguy M-L, Girard E, Trédan O, Dubot C, et al. Multimodal liquid biopsy for early monitoring and outcome prediction of chemotherapy in metastatic breast cancer. NPJ breast cancer. 2021 Sep;7(1):115. PATRICIA MONDELO MACÍA 360 430. Ruiz-Bañobre J, Molina-Díaz A, Fernández-Calvo O, Fernández-Núñez N, Medina-Colmenero A, Santomé L, et al. Rethinking prognostic factors in locally advanced or metastatic urothelial carcinoma in the immune checkpoint blockade era: a multicenter retrospective study. ESMO Open. 2021;6(2):100090. 431. Papachristos A, Sivolapenko GB. Pharmacogenomics, Pharmacokinetics and Circulating Proteins as Biomarkers for Bevacizumab Treatment Optimization in Patients with Cancer: A Review. J Pers Med. 2020 Aug 4;10(3). 432. Veyssière H, Bidet Y, Penault-Llorca F, Radosevic-Robin N, Durando X. Circulating proteins as predictive and prognostic biomarkers in breast cancer. Clin Proteomics. 2022 Dec 11;19(1):25. 433. Christensen TD, Maag E, Larsen O, Feltoft CL, Nielsen KR, Jensen LH, et al. Development and validation of circulating protein signatures as diagnostic biomarkers for biliary tract cancer. JHEP Reports. 2023 Mar;5(3):100648. Annex 1 367 Figure 1.10. (A) MET CN analysis. Plasma MET CN detected in healthy controls (n= 49), non-metastatic patients (n=34) and metastatic patients (n= 140). Differences were analysed using the Mann-Whitney-Wilcoxon U-Test. (B) Correlation between cfDNA levels and plasma MET CN in metastatic cancer patients (n= 140) using Pearson’s correlation. Abbreviations: CN, copy number. Figure 1.11. Plasma MET CN values in the different metastatic cancers included in the study. Abbreviations: CN, copy number. Figure 1.12. Comparison of MET CN status in tissue and cfDNA. (A) Distribution of MET CN measured by ddPCR and FISH (the point larger indicates the discordant value, whereas the horizontal and vertical dotted lines indicate cut-off points of ddPCR and FISH, respectively). (B) Representative example of a negative case for MET amplification obtained in a NSCLC patient by FISH. (C) Representative example of a positive case for MET amplification obtained in a NSCLC patient by FISH. Abbreviations: CN, copy number. Figure 1.13. Evaluation of the enrichment capacity of CellSearch® and Parsortix systems, using healthy blood spiked with cancer cell lines. (A) Percentage of spiked tumor cancer cells captured using CellSearch® and Parsortix systems. p-value <5x10-3, in all comparisons between CellSearch® and Parsortix System in each cell line using paired student t test. (B) Correlation between EPCaM mRNA expression levels measured by Affymetrix microarrays and RNA-seq. LNCaP, NCI-N87, SNU- 5, and AU565 express EpCAM while Hs746T and C32 express low levels or do not express EpCAM respectively. (C) Immunofluorescence characterization of EpCAM in our cancer cell lines. Scale bar represents 100 µm. Figure 1.14. Detection of c-MET expression using tumour cancer cells with the CellSearch® and Parsortix systems (Figure A and B, respectively). Representative images of c-MET expression scored on score 0 (cell line LNCaP), 1 (cell line AU565), 2 (cell line Hs746T) and 3 (cell line SNU-5). PATRICIA MONDELO MACÍA 368 Figure 1.15. CTCs enumeration and c-MET expression in blood samples evaluated by the CellSearch® (right panel) and Parsortix (left panel) systems. Distribution of c- MET scores in CTCs from patients with NSCLC. Abbreviations: NT, not tested. Figure 1.16. Timeline for the clinical course of patient id60. The blue and yellow bars represent the treatments time frame, and the red drops indicate blood collection time points. Percent mutant allelic frequency (L858R and T790M) and MET CN for patient id60 are shown. Abbreviations: CN, copy number. CHAPTER II Figure 2.1 Study schema and monitoring of the patient cohort, including patient enrolment and sample collection. Figure 2.2. Correlation between cfDNA levels using Qubit method and qPCR assay (Pearson correlation, R2= 0.79). Abbreviations: GE, genome equivalents. Figure 2.3. Kaplan-Meier survival analysis of hTERT cfDNA levels at baseline. Kaplan-Meier plots of progression-free survival (A) and overall survival (B). Figure 2.4. hTERT cfDNA levels at different time points (baseline, 6 and 12 weeks). No significant differences were found among any group (Wilcoxon test). Abbreviations: GE, genomic equivalents. Figure 2.5. hTERT cfDNA levels during pembrolizumab therapy and their association with therapy response. (A) cfDNA levels at baseline according the response to therapy. No significant associations were found (Wilcoxon test). (B) Clinical course for patients during pembrolizumab treatment. Swimmer plots for each patient (n=50) showing the levels of hTERT cfDNA at baseline (yellow colour indicates high hTERT cfDNA levels, and blue colour indicates low hTERT cfDNA levels). The total length of each bar indicates the duration of survival from start of pembrolizumab treatment. Left, squares are coloured according to the response based on RECIST1.1 criteria. Abbreviations: GE, genome equivalents; CR, complete response; PR, partial response; SDi, stable disease; PD, progression disease. Annex 1 369 Figure 2.6. hTERT cfDNA changes from baseline to 12 weeks. (A) hTERT cfDNA concentrations for the two cfDNA patterns (increase/decrease at 12 weeks) and showing the response to therapy. (B) Percentage of patients and median PFS for each cfDNA pattern. (C) Proportion of patients with high and low levels baseline, 6 and 12 weeks. p-value was calculated by Fisher’s exact test. (D) Kaplan-Meier plot of progression-free survival of the favourable/unfavourable changes at 12 weeks*. Abbreviations: CR/PR, complete response/partial response SDi/PD, stable disease/progression disease. Figure 2.7. Immunofluorescence characterization of PD-L1 in cancer cell lines. We used three lung cancer cell lines with different grade of PD-L1 expression (NCI-H460, medium-high expression, NCI-H322, low-medium expression and A549, no expression). Scale bar represents 100 µm. Figure 2.8. (A-B) Detection of PD-L1 expression after spiking cancer cells lines in healthy blood and analyse them with the CellSearch® and Parsortix systems, respectively and representative images of different grades of PD-L1 expression. NCI- H460 stimulated with IFN-γ shown high expression, NCI-H460 medium expression, NCI-H322, low-medium expression and A549, no expression. Figure 2.9. (A) Representative images of CTCs detected with the CellSearch® system in patients with NSCLC. Samples were subjected to immunostaining with DAPI, CD45 (APC), CKs (FLU) and PD-L1 (PE). (B) Representative images of CTCs detected with the Parsortix system in patients with NSCLC. Samples were subjected to immunostaining with DAPI, CD45 (AF647), CKs (AF488) and PD-L1 (PE). Figure 2.10. Correlation of PD-L1 positivity between tumour tissues (by tumour proportion scores) and CTCs with the CellSearch® (A) and Parsortix systems (B). Figure 2.11. Correlation of PD-L1 positivity between tumour tissues (by tumour proportion scores) and CTCs with the CellSearch® (A) and Parsortix systems (B). PATRICIA MONDELO MACÍA 370 Figure 2.12. Kaplan–Meier survival analysis of CTCs at baseline. Kaplan–Meier plots of PFS (A) and OS (B). (C) Comparison of the response to pembrolizumab based on CTC detection using CellSearch®. p-value was calculated by Fisher’s exact test. Figure 2.13. CTCs and hTERT cfDNA levels correlate with the prognosis of patients with NSCLC treated with pembrolizumab and the response of therapy. (A) Kaplan- Meier survival plot of progression-free survival based on the combination of cfDNA and CTC levels at baseline. (B) Kaplan-Meier survival plot of overall survival based on the combination of cfDNA and CTC levels at baseline. (C) Objective response rate in patients with low cfDNA levels and undetectable CTCs (n=12) versus patients with high cfDNA levels and undetectable CTCs or low cfDNA levels and detectable CTCs or high cfDNA levels and detectable CTCs (n=18). CHAPTER III Figure 3.1. Study sampling points and different cohorts included in the study. Figure 3.2. Scatter plot representing correlation between cfDNA levels using Qubit method and qPCR assay at different times of therapy (baseline, 3 weeks and progression disease) using Pearson’s correlation method. A good correlation between was found (R2 =0.959). Abbreviations: GE, genomic equivalents. Figure 3.3. CfDNA levels using two different approaches. (A-C) Total cfDNA levels in healthy controls and patients with SCLC. Statistical analysis between both groups was performed by the Mann–Whitney–Wilcoxon U-Test. (B-D) ROC curves for qPCR assay (B) and Qubit method (D) show high sensitivity and specificity. Abbreviations: GE, genomic equivalents; AUC, area under the curve; CI, confidence interval. Figure 3.4. cfDNA levels as a prognostic biomarker at baseline. (A-B) Kaplan-Meier survival analysis of cfDNA levels at baseline for PFS (A) and OS (B). Figure 3.5. Clinical course of all patients included in the study. Swimmers’ plot showing each patient therapy and the different times of sample collection. The total length of each bar indicates the duration of survival from the diagnoses. Annex 1 371 Figure 3.6. CfDNA levels in SCLC patients at different time-points (baseline, 3 weeks after treatment onset and at progression disease). Total cfDNA levels at baseline were significantly higher than at 3 weeks after the therapy onset (Wilcoxon test p= 0.002). Abbreviations: GE, genomic equivalents. Figure 3.7. cfDNA levels as a prognostic biomarker during treatment. Kaplan-Meier survival analysis of cfDNA levels at 3 weeks for PFS (A) and OS (B). Figure 3.8. Kaplan-Meier survival analysis of cfDNA levels at progression disease for OS. Figure 3.9. CTCs detected in SCLC patient using CellSearch® system. (A) Representative images of CTCs detected in our cohort using the CellSearch® system. (B) Number of CTCs detected in each patient. Patients with presence of <10 CTCs are represented in red columns. Figure 3.10. Kaplan-Meier survival analysis of CTCS levels at baseline for PFS. The presence of ≥150 CTCs/7.5mL of blood was significantly associated with shorter PFS rates. Figure 3.11. CTCs and association with clinical characteristics. The number of CTCs detected in our cohort were significantly associated with the stage of patients and a poor performed status. Also, high cfDNA levels and the presence of CTCs at baseline were significantly associated. Abbreviations: CH, chemotherapy; CH + IM, chemotherapy plus immunotherapy. Figure 3.12. Final multivariate Cox regression prognostic model for (A) PFS and (B) OS. Abbreviations: HR, hazard ratio; CI, confidence interval. Figure 3.13. Kaplan-Meier survival analysis according to risk-group PFS (A) and OS (B). Abbreviations: NA, not applicable. CHAPTER IV Figure 4.1. (A) Swimmers’ plot on patients showing the response of therapy. The total length of each bar indicates the duration of survival from the diagnoses. (B-C) Kaplan-Meier plots show highly significant differences between responders and nonresponders to pembrolizumab treatment in PFS (B) OS (C). PATRICIA MONDELO MACÍA 372 Figure 4.2. Proteomic of NSCLC response to pembrolizumab treatment. (A) The proteomic workflow involved plasma extraction from blood samples of 48 NSCLC patients undergoing pembrolizumab Then, 30 µl of plasma samples therapy. were employed to proteins extraction. The proteins were trypsin-digested to obtained peptides (fractionation). A pool of total peptides samples was employed to generate a spectral library using a DDA method. Finally, pooled peptides of each sample were analysed using a SWATH-MS method employing the previous spectral library and the Human Pan-Human Library. (B) Total number of proteins quantified in each group of patients. Figure 4.3. (A) Volcano plot of differentially expressed proteins at baseline. Upregulated proteins are in blue and downregulated proteins are in yellow. (B) Heatmap of 324 differentially expressed proteins at baseline (p-value <0.05), that discriminate between responder and non-responder patients to pembrolizumab therapy. Figure 4.4. Bar graph for viewing of the top 15 enriched GO process involved the 324 DEPs. A colour scale represents statistical significance. Figure 4.5. Heatmap showing the top enrichment clusters, one row per group, using a colour scale to represent statistical significance. Gray colour indicates a lack of significance. Figure 4.6. Proteomaps that showed the functional differences between responders (left) and non-responders (right) to immunotherapy. Each polygon corresponds to a single KEGG pathway, and its size correlates with the ratio between both groups. Figure 4.7. (A) Volcano plot and heat map of differentially expressed proteins at baseline. (B) PCA analysis showing the separation of samples from responders (blue) and non-responders (yellow) to pembrolizumab therapy according the 66 DEPs found by LC-MS/MS analysis. Figure 4.8. (A) Volcano plot showing the 7 proteins of our predictive model. (B) ROC curves of each protein and the combination in the discovery cohort. (C) ROC curve Annex 1 373 analyses of the PD-L1 expression on tumour tissue. Abbreviations: AUC, area under the curve; FC, foldchange. Figure 4.9. ROC curves of each protein and the combination in the validation cohort. Figure 4.10. Kaplan-Meier survival analysis of ATG9A, SPTN2, HPS5 and DCDC2 proteins levels at baseline. Kaplan-Meier plots of progression-free survival (A-C) and overall survival (D-E) showed significant associations between proteins levels before start therapy and prognosis in metastatic NSCLC under immunotherapy regimens. Figure 4.11. Protein levels of our protein models were measured at different time points during the immunotherapy regimen. Abbreviations: PD, progression of disease. Figure 4.11. ROC analyses with the protein model at different time points during the immunotherapy regimen (at 6 weeks and at 12 weeks after start therapy). Overall discussion Figure 25. Number of publications per year containing the terms “precision medicine” and “liquid biopsy” in PubMed between 2012 and 2022. Figure 26. Sankey diagram showing the total lung cancer patients included in the present thesis, subdivided by lung cancer type, stage and therapy. Figure 27. Total cfDNA levels in lung cancer patients and healthy controls. (A) Boxplot of cfDNA levels in NSCLC, SCLC and healthy controls (Mann-Whitney test). (B) ROC showing the discriminatory value of cfDNA levels to discriminate against lung cancer patients versus healthy controls (AUC=0.8885). Figure 28. CTCs detected in 36 NSCLC patients employing two different isolation methods. Figure 29. CTCs detected in NSCLC and SCLC cohorts using the CellSearch® system. SCLC patients presented statistically higher number of CTCs than NSCLC patients (p-value <0.001; Mann-Whitney test). PATRICIA MONDELO MACÍA 374 2. TABLES INDEX Introduction Table 1. Eight Edition of lung cancer stage grouping according the AJCC-TNM staging system (21) and the correspondence VALG staging for SCLC (23). Table 2. Summary of the most common strategies for ctDNA analysis. CHAPTER I.A Table 1.1. Sensitivity of Idylla™ system using commercial cfDNA (160 ng). Table 1.2. Demographics and clinical characteristics of the patients. Table 1.3. Sample determinations included in the study. CHAPTER I.B Table 1.4. Demographics and clinical characteristics of the patients at baseline. Table 1.5. Baseline demographics and clinical characteristics of the metastatic cancer patient population analysed for MET amplification. Table 1.6. CTC detection and c-MET expression on CTCs using CellSearch® and Parsortix systems in patients with NSCLC. CHAPTER II Table 2.1. Demographics and clinical characteristics of the patients at baseline. Table 2.2. ROC analysis to determine the value of hTERT cfDNA levels to discriminate progression or death. Table 2.3. Univariate and multivariate Cox regression analyses of cfDNA levels, CTC counts and clinical parameters. Table 2.4. Circulating tumour cells enumeration and PD-L1 analysed using CellSearch® and Parsortix systems. Table 2.5. Comparison of the CTCs levels according to the response to therapy. Table 2.6. Univariate and multivariate Cox regression analyses of combined changes in CTCs and cfDNA levels. Annex 1 375 Table 2.7. Univariate and multivariate Cox regression analyses of combined changes in CTCs and cfDNA levels and clinical parameters. CHAPTER III Table 3.1. Patients’ demographics and clinical characteristics at baseline with cfDNA levels. Table 3.23. Statistics of cfDNA levels at baseline, at 3 weeks after the onset of therapy and at disease progression. Table 3.3. Receiver operating characteristic curves analysis to determine the value of cfDNA levels to discriminate progression or death. Table 3.4. Analysis to determine the prognostic value of CTCs to discriminate progression or death. Table 3.5. Univariate and multivariate Cox regression analyses of cfDNA levels, CTC counts and clinical parameters. Table 3.6. PFS and OS probabilities estimated according two risk groups. CHAPTER IV Table 4.1. Patients’ demographics and clinical characteristics at baseline. Table 4.2. Characteristics of the 7 proteins that allow us to predict immunotherapy response in advanced NSCLC patients. Annex 2 383 Contribution to this work: I, Patricia Mondelo, has been involved in the Conceptualization, Methodology, Investigation, Writing – original draft, Writing – review & editing and Visualization. Particularly, I have performed IdyllaTM system set up, blood samples processing, IdyllaTM assays, statistical analyses, and manuscript writing. PATRICIA MONDELO MACÍA 384 b. Detection of MET Alterations Using Cell Free DNA and Circulating Tumor Cells from Cancer Patients - Article Research. Patricia Mondelo-Macía#, Carmela Rodríguez-López#, Laura Valiña, Santiago Aguín, Luis León-Mateos, Jorge García-González, Alicia Abalo, Óscar Rapado- González, Mercedes Suárez-Cunqueiro, Angel Díaz-Lagares, Teresa Curiel, Silvia Calabuig-Fariñas, Aitor Azkárate, Antònia Obrador-Hevia, Ihab Abdulkader, Laura Muinelo-Romay*, Roberto Diaz-Peña* and Rafael López-López. * Corresponding. / # These authors have contributed equally to this work Cells. 2020; 9(2):522 ISSN 2073-4409 MDPI, 21/02/2020 Full-text available: https://doi.org/10.3390/cells9020522 Permissions: This is an open access article distributed under the terms of the Creative Commons CC-BY license which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Journal Citation Reports (JCR) category rank (2020): Q2 (53/195): Cell biology Journal Impact factor (2020) – 6.600 Contribution to this work: I, Patricia Mondelo, has been involved in the Methodology, Validation, Formal Analyses, Investigation, Writing – original draft and Visualization. Particularly, I have performed cell culture, spiked experiments, immunofluorescence, blood samples processing, nuclei acid extraction, ddPCR assays, Parsortix and CellSearch® assays, statistical analyses, and manuscript writing. Annex 2 385 c. Clinical Potential of Circulating Free DNA and Circulating Tumour Cells in Patients with Metastatic Non-Small-Cell Lung Cancer treated with Pembrolizumab - Article Research. Patricia Mondelo-Macía, Jorge García-González, Luis León-Mateos, Urbano Anido, Santiago Aguín, Ihab Abdulkader, María Sánchez-Ares, Alicia Abalo, Aitor Rodríguez-Casanova, Ángel Díaz-Lagares, Ramón Manuel Lago-Lestón, Laura Muinelo-Romay, Rafael López-López and Roberto Díaz-Peña*. * Corresponding. Molecular Oncology. 2021; 15 (11) p. 2923–2940. ISSN 1878-0261 FebsPress, 31/08/2021 Full-text available: https://doi.org/10.3390/cells9020522 Permissions: This is an open access article distributed under the terms of the Creative Commons Attribution license which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Journal Citation Reports (JCR) category rank (2021): Q1 (51/245): Oncology Journal Impact factor (2021) – 7.449 Contribution to this work: I, Patricia Mondelo, has been involved in the Conceptualization, Methodology, Investigation, Data curation, Software, Writing – original draft and Analysis of data. Particularly, I have performed cell culture, spiked experiments, immunofluorescence, blood samples processing, nuclei acid extraction, qPCR assays, Parsortix and CellSearch® assays, statistical analyses and manuscript writing. PATRICIA MONDELO MACÍA 386 d. Plasma cell-free DNA and CTCs as prognostic biomarkers in small cell lung cancer patients - Article Research. Patricia Mondelo-Macía, Jorge García-González, Alicia Abalo, Manuel Mosquera- Presedo, Santiago Aguín, María Mateos, Rafael López-López, Luis León-Mateos*, Laura Muinelo-Romay*, and Roberto Díaz-Peña*. * Corresponding. Translational Lung Cancer Research. 2022; (10):1995-2009. ISSN 2218-6751; EISSN 2226-4477 AME Publishing Company, 31/08/2022 Full-text available: https://doi.org/10.21037/tlcr-22-273 Permissions: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license) Journal Citation Reports (JCR) category rank (2021): Q2 (105/245): Oncology Journal Impact factor (2021) – 4.726 Contribution to this work: I, Patricia Mondelo, has been involved in the Conceptualization and design, Methodology, Investigation, Data analysis and Interpretations, Writing – original draft. Particularly, I have performed data curation, blood samples processing, nuclei acid extraction, qPCR assays, CellSearch® assays, statistical analyses and manuscript writing. 387 1.3 Patents derived from this thesis Patent: MAPRI: In vitro method for predicting cancer patient response to PD-1 and/or PD-L1 inhibitors. Inventors: Ana Belén Dávila Ibáñez, Rafael López López, Roberto Díaz Peña, Susana Bravo López, Patricia Mondelo-Macía, Laura Muinelo Romay, Luis Ángel León Mateos y Jorge José García González. EP No. 23 382 377.2. ANNEX 3 ETHICAL CONSIDERATIONS and ATTACHED PERMISSIONS 391 ANNEX 3: Ethical considerations and attached permissions 1. IMAGES USE Unless clarified, all the images presented in this thesis have been produced by the author. In the case of images reused or adapted from other manuscripts, permission has been granted by the publishers, and the legal use has been clarified at the bottom of the corresponding figures. In addition, the free resource Flaticon.com and Servier Medical Art (smart.servier.com) by Servier (licensed under a Creative Commons Attribution 3.0 Unported License) were employed. 2. HUMAN CELL CULTURE All cancer lines used in this work were acquired from commercially available resources (American Tissue Culture Collection, ATCC) and cultured in the conditions recommended by the manufacturers and only used for the research purposes specifically described in this thesis. 3. PATIENT’S SAMPLES Blood samples were collected in accordance with the guidelines and protocols approved by the Institutional Ethical Committees, Galician Clinical Research Ethics Committee, SERGAS: code approval: 2017/538. All individuals signed informed consent forms and could withdraw their consent at any time. The study was performed in accordance with the Declaration of Helsinki. PATRICIA MONDELO MACÍA 392 4. ATTACHED PERMISIONS a. Favourable report (Clinical research ethics committee of Galicia, CEIC).