Desarrollo y optimización de nuevos métodos de análisis de péptidos inmunogénicos del gluten: aplicaciones en seguridad alimentaria y en el seguimiento de la enfermedad celíaca

DESARROLLO Y OPTIMIZACIÓN DE NUEVOS MÉTODOS DE ANÁLISIS DE PÉPTIDOS INMUNOGÉNICOS DEL GLUTEN: APLICACIONES EN SEGURIDAD ALIMENTARIA Y EN EL SEGUIMIENTO DE LA ENFERMEDAD CELÍACA Verónica Segura Montero Sevilla 2024 DEPARTAMENTO DE MICROBIOLOGÍA Y PARASITOLOGÍA FACULTAD DE FARMACIA UNIVERSIDAD DE SEVILLA DESARROLLO Y OPTIMIZACIÓN DE NUEVOS MÉTODOS DE ANÁLISIS DE PÉPTIDOS INMUNOGÉNICOS DEL GLUTEN: APLICACIONES EN SEGURIDAD ALIMENTARIA Y EN EL SEGUIMIENTO DE LA ENFERMEDAD CELÍACA VERÓNICA SEGURA MONTERO SEVILLA, 2024 La mejor forma de predecir el futuro es crearlo (Peter Drucker) RESUMEN RESUMEN RESUMEN El gluten es una mezcla compleja de proteínas de almacenamiento llamadas prolaminas presentes en ciertos cereales como el trigo, la cebada, el centeno, la avena y sus derivados. Estas proteínas desempeñan un papel fundamental al conferir elasticidad y viscosidad a la masa de harina durante la elaboración de productos horneados, lo que influye directamente en su textura y estructura final. El gluten tiene una gran importancia en la industria alimentaria y agrícola, y un gran interés médico debido a las reacciones adversas que puede provocar en la salud. Está presente en más del 80% de los productos alimenticios manufacturados, de hecho, después del azúcar, es el ingrediente alimentario más usado en la civilización Occidental. Las enfermedades relacionadas con la ingesta de gluten engloban diferentes entidades con características epidemiológicas, clínicas y fisiopatológicas propias tales como, alergia al trigo, ataxia por gluten, dermatitis herpetiforme, sensibilidad al gluten no celíaca o enfermedad celíaca (EC). Entre estas patologías, la más conocida es esta última, que se define como un proceso sistémico que produce fundamentalmente una enteropatía crónica del intestino delgado debido a una respuesta inadecuada al gluten en pacientes predispuestos genéticamente. Actualmente, el único tratamiento disponible para los pacientes con EC es la DSG durante toda la vida. De hecho, se ha demostrado que una estricta adherencia a la DSG es fundamental para eliminar los síntomas de la enfermedad, evitar deficiencias nutricionales y mejorar la calidad de vida de estos pacientes, evitando complicaciones a corto-medio plazo y reduciendo el coste socio-sanitario. No obstante, numerosos estudios han sugerido que las transgresiones de la dieta tanto voluntarias como involuntarias son relativamente frecuentes. Entre las principales causas de la no adherencia a la DSG se encuentran los condicionamientos de la vida social, el elevado coste de los productos dietéticos especiales con buenas características nutricionales o las dificultades tecnológicas para garantizar la ausencia de gluten en los alimentos manufacturados. La falta de alternativas terapéuticas para la EC subraya la necesidad de métodos analíticos sensibles y de bajo coste para medir con precisión el contenido de gluten en todo tipo de muestras, tales como, alimentos, incluidos los hidrolizados, bebidas, cosméticos y medicamentos. En Europa, el etiquetado de los productos sin gluten está regulado por el Codex Alimentarius que establece el límite ≤ 20 ppm para los alimentos etiquetados como sin gluten. En contraste, otras asociaciones como Gluten-Free Certification Organization (GFCO/ Norteamérica, USA) y Celiac Support Association®/United States of America, Inc. (CSA/USA, Inc.) ABREBIATURAS Y SÍMBOLOS FODMAPs Fermentables, oligosacáridos, disacáridos, monosacáridos y polioles g gramos GFCO Gluten-Free Certification Organization GIP Péptidos inmunogénicos del gluten h Horas HLA Antígenos leucocitarios humanos HMW Móleculas de alto peso molecular IC50 Concentración de la línea que reduce la absorbancia pico en un 50% en el ensayo Ig Inmunoglobulina IL Interleuquina Kg Kilogramos LFIA Inmunoensayos de flujo lateral LIEs Linfocitos intraepiteliales LMW Móleculas de peso molecular LOD Límite de detección LOQ Límite de cuantificación M Molar mM Milimolar mg Miligramos min Minutos MMW Móleculas de medio peso molecular N.A. No aplicable P Prolina ABREBIATURAS Y SÍMBOLOS PCR Reacción en cadena de la polimerasa ppm Partes por millón PVP Polivinilpirrolidona PWG Prolamin Working Group Q Glutamina r2 Coeficiente de correlación SDS-PAGE Electroforesis en gel de poliacrilamida con dodecilsulfato de sodio SEEC Sociedad Española de Enfermedad Celíaca SII Síndrome del intestino irritable SGNC Sensibilidad al gluten no celíaca tTG Transglutaminasa tisular TCR Receptores de células T UGES Universal Gluten Extraction Solution VPN Valor predictivo negativo VPP Valor predictivo positivo ÍNDICE ÍNDICE CAPÍTULO 1. INTRODUCCIÓN ....................................................................................................... 3 1. EL GLUTEN ..................................................................................................................... 4 1.1. Patologías relacionadas con el gluten ..................................................................... 6 2. LA ENFERMEDAD CELÍACA ...................................................................................................... 9 2.1. Patogenia .............................................................................................................. 9 2.1.1. Factores genéticos ............................................................................................. 10 2.1.2. Factores ambientales ......................................................................................... 11 2.1.3. Respuesta inmunitaria ....................................................................................... 12 2.2. Epidemiología ...................................................................................................... 14 2.3. Manifestaciones clínicas ...................................................................................... 15 2.4. Diagnóstico ......................................................................................................... 18 2.4.1. Pruebas genéticas .............................................................................................. 18 2.4.2. Pruebas serólogicas ........................................................................................... 18 2.4.3. Pruebas histopatológicas ................................................................................... 19 3. TRATAMIENTO DE LA ENFERMEDAD CELÍACA ..................................................................... 20 3.1. Terapia dietética: dieta libre de gluten ................................................................. 20 3.1.1. Dieta sin gluten .................................................................................................. 20 3.1.1.1. Control de productos sin gluten. Métodos de análisis ............................... 21 3.1.2. Métodos de monitorización de la dieta sin gluten ............................................ 25 3.1.2.1. Sintomatología .......................................................................................... 25 3.1.2.2. Test serológicos ......................................................................................... 25 3.1.2.3. Cuestionarios y entrevistas dietéticas ....................................................... 26 3.1.2.4. Biopsia intestinal ........................................................................................ 26 3.1.2.5. Detección de péptidos inmunogénicos de gluten en muestras humanas .. 27 4. BILBLIOGRAFÍA ...................................................................................................................... 29 ÍNDICE CAPÍTULO 2. ANTECEDENTES DEL TEMA Y OBJETIVOS ............................................................. 44 CAPÍTULO 3 ................................................................................................................................. 56 Prospective longitudinal study: use of faecal gluten immunogenic peptides to monitor children diagnosed with coeliac disease during transition to a gluten-free diet. ..................................... 56 CAPÍTULO 4 ................................................................................................................................. 68 Evaluation of the usefulness of an automatable immunoassay for monitoring celiac disease by quantification of immunogenic gluten peptides in urine. ........................................................... 68 CAPÍTULO 5 ................................................................................................................................. 84 Rapid, effective, and versatile extraction of gluten in food with application on different immunological methods. ............................................................................................................. 84 CAPÍTULO 6 ............................................................................................................................... 104 A highly sensitive method for the detection of hydrolyzed gluten in beer samples using LFIA . 104 CAPÍTULO 7. RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN ..................................... 118 CAPÍTULO 8. CONCLUSIONES ................................................................................................... 164 ANEXO ....................................................................................................................................... 168 Dictamen PEIBA ....................................................................................................................... 170 CAPÍTULO 1. INTRODUCCIÓN 3 INTRODUCCIÓN 4 CAPÍTULO 1 1. EL GLUTEN El gluten es un conjunto de proteínas compuesto principalmente por gliadinas y gluteninas, presentes en los endospermos de diversos cereales como el trigo (Triticum aestivum) (Figura 1), la cebada (Hordeum vulgare), el centeno (Secale cereale), la avena (Avena sativa) y sus derivados híbridos (Comino et al., 2016; Biesiekierski, 2017; Wieser et al., 2020; Aljada et al., 2021). Estas proteínas desempeñan un papel fundamental al conferir elasticidad y viscosidad a la masa de harina durante la elaboración de productos horneados, lo que influye directamente en su textura y estructura final (Shewry, 2019). Las propiedades únicas del gluten se deben a su estructura característica, que permite la formación de una "red proteica" durante el amasado de la harina con agua (Shewry, 2019). Esta red proteica proporciona a la masa de harina su elasticidad, facilitando la fermentación y dando como resultado la textura elástica y esponjosa de los panes y productos horneados. Además, el gluten posee otras características importantes, como su notable capacidad de absorción de agua, cohesividad, viscosidad y elasticidad, lo que lo distinguen de otras proteínas vegetales (Sharma et al., 2024). Estas propiedades lo convierten en un aglutinante efectivo y en un agente gelificante y emulsionante que contribuye a la estructura y consistencia de una amplia variedad de alimentos procesados, incluyendo panes, productos horneados y pasta (Shewry et al., 2002). Figura 1. Identificación del gluten dentro de un grano de trigo. Originalmente, las proteínas de los cereales fueron clasificadas mediante el método de extracción secuencial conocido como fraccionamiento de Osborne (Osborne, 1924). Este método permitía distinguir cuatro fracciones: albuminas (solubles en agua), globulinas (solubles en dilución salina), gliadinas (solubles en 60-70% de alcohol) y glutelinas (solubles en condiciones alcalinas extremas), siendo estas dos últimas consideradas como las proteínas del INTRODUCCIÓN 11 2.1.2. Factores ambientales En los últimos años, se ha planteado la posibilidad de la implicación de otros factores ambientales en el desarrollo de la EC. Entre ellos se encuentran las diferencias entre la cantidad y calidad del gluten consumido, los patrones de alimentación en la infancia, las enfermedades infecciosas en la infancia o la composición de la microbiota intestinal, entre otros (Størdal et al., 2021; Iversen y Sollid, 2023). No obstante, aún sigue existiendo controversia al respecto, y se requiere de más estudios para establecer unas conclusiones firmes. La lactancia materna y el momento de la introducción del gluten se han considerado factores particularmente influyentes (Villamil et al., 2021). Sin embargo, la evidencia disponible sobre la relación entre la EC y la leche materna sigue sin ser concluyente. La lactancia materna prolongada y, en particular, la lactancia materna en el momento de la introducción del gluten, parecen disminuir el riesgo de desarrollar EC o al menos posponer su aparición (Aronsson y Agardh, 2023). Según Logan et al. (2020), la introducción del gluten a partir de los 4 meses de edad se ha asociado con una prevalencia reducida de EC. Las directrices actuales de la Sociedad Europea de Gastroenterología, Hepatología y Nutrición Pediátrica (ESPGHAN) recomiendan introducir los cereales con gluten entre los 4 y 12 meses, evitando grandes cantidades de gluten en las primeras semanas (Szajewska et al., 2016). Por otra parte, varios estudios han sugerido una relación entre las infecciones intestinales y respiratorias y el riesgo de EC. Recientemente, se ha investigado la posible relación entre las infecciones virales y la pérdida de tolerancia oral, lo que abre nuevas perspectivas en la comprensión de la EC y podría llevar al desarrollo de estrategias preventivas en el futuro. Entre ellas, se han estudiado especialmente las infecciones frecuentes por rotavirus, astrovirus y adenovirus y otras como la producida por Campylobacter (Cohen et al., 2023). Algunos de estos virus pasan desapercibidos en el intestino, siendo rápidamente eliminados por el sistema inmunológico, mientras que, otros pueden causar disrupciones en la mucosa intestinal y generar cambios en el sistema inmunológico. Posteriormente, la exposición a proteínas dietéticas como las del gluten podrían desencadenar una reacción indeseada, lo que los convierte en posibles factores de riesgo para la EC en individuos genéticamente predispuestos (Kahrs et al., 2019; Gómez-Rial et al., 2020). De manera similar, parece que la exposición a antibióticos en estas edades también podría contribuir, induciendo cambios en la microbiota, alterando la permeabilidad intestinal y activando los mecanismos de inflamación (Protocolo para el diagnóstico precoz de la enfermedad celíaca, 2018; Dydensborg et al., 2019). Finalmente, se ha observado una asociación entre la EC y la disbiosis intestinal, lo que sugiere un posible papel de la microbiota en la patogénesis de la enfermedad (Yemula, 2024). Se ha especulado sobre cómo la microbiota intestinal puede influir en la permeabilidad INTRODUCCIÓN 12 intestinal, la maduración intestinal y la respuesta inmune (Gilbert et al., 2018; D’Avino et al., 2021). Sin embargo, los mecanismos exactos involucrados aún no se comprenden completamente. 2.1.3. Respuesta inmunitaria El modelo más aceptado para explicar la inmunopatogenia de la EC establece que el gluten tiene un efecto dual mediado por la inmunidad innata (efecto directo sobre el epitelio intestinal) y la inmunidad especifica adaptativa, a través de la activación de los linfocitos T CD4+ de la lámina propria (Lebwohl et al., 2018; Tye-Din et al., 2018; Lindfors et al., 2019; D’Avino et al., 2021). La respuesta innata parece ser una forma de defensa rápida por parte de las células epiteliales intestinales (CEI) en una situación de estrés. El estímulo que desencadena esta respuesta aún no está confirmado, pero se cree que pueden ser péptidos del gluten, otros componentes del trigo o alguna interacción microorganismo-huésped (Tye-Din et al., 2018; Lindfors et al., 2019). Todo ello provoca un aumento de la expresión de las moléculas MIC y HLA- E en las células epiteliales estresadas, así como la liberación de citoquinas inflamatorias como la interleuquina-15 (IL-15). Ésta juega un papel relevante, ya que, por un lado, suprime la acción de las células T reguladoras, lo que conduce a la pérdida de tolerancia al gluten y a la alteración en la regulación de la respuesta inmunitaria. Por otro lado, activa la citotoxicidad de los LIEs. Estos últimos, en su mayoría, son células T CD8+ citotóxicas que contienen gránulos con perforina y granzimas, cuya función principal es reconocer y eliminar células epiteliales dañadas. La activación de los LIEs se produce a través de una mayor expresión de los receptores NKG2D y NKG2C que se unen a las moléculas MIC y HLA-E presentes en las células epiteliales, lo que resulta en la apoptosis de los enterocitos (Tye-Din et al., 2018; Lindfors et al., 2019; D’Avino et al., 2021). La respuesta adaptativa se inicia cuando los péptidos del gluten llegan a la lámina propria, y son reconocidos por las células presentadoras de antígenos (CPA) mediante las moléculas HLA DQ2 y/o DQ8 localizadas en la superficie celular. Estas CPA a su vez presentan los péptidos a las células T CD4+ colaboradoras (Th, de helper en inglés). Para que esto ocurra de forma más eficiente, los péptidos del gluten deben ser primero desaminados por la TG2, lo que les confiere aminoácidos con carga negativa. Por lo tanto, la TG2 juega un rol muy importante en el desarrollo de la EC (D’Avino et al., 2021). Recientemente, se ha demostrado que esta enzima también puede realizar la desaminación de los péptidos del gluten en el lumen intestinal (Iversen et al., 2020; Iversen y Sollid, 2023) (Figura 4). INTRODUCCIÓN 13 Una vez que las CPA presentan los péptidos a las células T CD4+ y éstas reconocen los epítopos de los antígenos presentados, las células T pasan a diferenciarse en distintos subtipos que dan lugar a la producción de citoquinas proinflamatorias, así como a la liberación de células T CD8+ que causan inflamación en distintos sistemas. Se han identificado tres subtipos principales de células T CD4+ en la EC: Th1 (productoras de IFN-γ), Th17 (productoras de IL-17) y células T reguladoras (productoras de IL-10) (D’Avino et al., 2021). Las citoquinas liberadas tienen un efecto paracrino en las células colindantes, por ejemplo, inducen cambios en los receptores de las CPA y su producción de IL-15, así como la actividad de los neutrófilos y la producción del factor de crecimiento tumoral alfa (TNF-α), todo ello contribuye al estado inflamatorio crónico característico de la EC (Tye-Din et al., 2018; Lindfors et al., 2019; D’Avino et al., 2021). Además, se produce un aumento de células plasmáticas e inmunoglobulinas de tipo A (IgA), M (IgM) y G (IgG) específicas. Tras la presentación de los péptidos del gluten a las células T CD4+, estas envían señales a células B específicas del gluten. Estas se activan ante estas señales y comienzan su diferenciación a células plasmáticas capaces de secretar anticuerpos contra los péptidos de gliadina desaminada (DGP) y contra la TG2 (Stamnaes y Sollid, 2015; Tye-Din et al., 2018; Lindfors et al., 2019). Generalmente estas poblaciones de anticuerpos se encuentran en altos niveles en el torrente sanguíneo, y en la mucosa intestinal de pacientes con EC que están consumiendo gluten. La función de estos anticuerpos en la patogenia de la EC podría estar relacionada con el aumento de la permeabilidad intestinal, permitiendo el acceso de más péptidos a la lámina propria. Finalmente, todo esto inicia una cascada de reacciones que degeneran en hiperplasia de las criptas y aplanamiento de las vellosidades intestinales (López- Casado et al., 2018; Lindfors et al., 2019; McAllister et al., 2019) (Figura 4). INTRODUCCIÓN 14 Figura 4. Respuesta inmunitaria de la EC. Las proteínas del gluten son parcialmente digeridas por proteasas gastrointestinales para producir péptidos del gluten (azul). Los residuos de glutamina de estos péptidos son desamidados por la tTG. Los péptidos desamidados (rojo) o complejos tTG-péptido de gluten se someten a endocitosis por parte de las células presentadoras de antígenos (CPA) y los péptidos se presentan posteriormente en los receptores HLA-DQ asociados a la enfermedad. Tras el reconocimiento por los receptores de células T (TCR) en las células T CD4 + específicas de la EC, estos complejos MHC-péptido desencadenan la respuesta inflamatoria característica. EC, enfermedad celíaca; HLA, antígeno leucocitario humano; LIEs, linfocitos intraepiteliales; tTG, transglutaminasa 2 (modificado de Besser y Khosla, 2023). 2.2. Epidemiología En términos globales, la seroprevalencia de la EC se sitúa en torno al 1,4% (Mendía et al., 2023), con una cifra del 0,7% cuando se considera el diagnóstico por biopsia (Singh et al., 2018). Sin embargo, la prevalencia real puede variar entre estas dos cifras debido a las limitaciones de las herramientas de diagnóstico, lo que puede llevar a la sobre- o subestimación de casos (Lebwohl y Rubio-Tapia, 2021). En España, la prevalencia en la población infantil oscila entre el INTRODUCCIÓN 15 1% y el 3%, mientras que en adultos es del 0,26% al 0,35% (Singh et al., 2018; Fueyo-Díaz et al., 2019). El metaanálisis llevado a cabo por King et al., 2020, que incluyó estudios desde 1950 a 2017, demostró que la incidencia de la EC aumentó un 7,5% (IC95%: 5,8-9,3%) de forma anual en este periodo. Aunque los motivos exactos de este incremento no están completamente claros, no cabe duda de que una mayor concienciación de la población y de los profesionales sanitarios sobre la EC, junto con la mejora en las herramientas de diagnóstico, han sido factores determinantes (Al-Toma et al., 2019; Bai y Ciacci, 2017). Inicialmente se consideraba una enfermedad pediátrica europea, pero se ha demostrado que la EC puede desarrollarse en cualquier etapa de la vida edad y en diversas etnias (Ludvigsson et al., 2014; Singh et al., 2018). Estudios de cohortes recientes han mostrado que la mayoría de los casos de EC se desarrollan antes de los 10 años, aunque también puede aparecer en la etapa adulta e incluso en la etapa geriátrica (Collin et al., 2018; Aronsson et al., 2019; ).El riesgo de desarrollar EC varía entre regiones geográficas, razas y etnias, lo que sugiere una influencia de factores genéticos, ambientales y dietéticos en su aparición (Choung et al., 2015; Al-Toma et al., 2019). Por ejemplo, en el estudio National Health and Nutrition Examination Survey (NHANES) realizado en Estados Unidos se observó una seroprevalencia significativamente mayor en la población caucásica en comparación con otras etnias (Choung et al., 2015). En Europa, se ha observado una mayor prevalencia en Finlandia en comparación con otros países como Alemania e Italia (Mustalahti et al., 2010). Además, diferencias en el conocimiento del personal sanitario, los protocolos de diagnóstico y la disponibilidad de recursos también pueden influir en las diferencias observadas entre regiones y etnias (King et al., 2020; Lebwohl y Rubio-Tapia, 2021; Elli et al., 2024). 2.3. Manifestaciones clínicas La EC presenta un amplio abanico de formas clínicas de expresión con signos y síntomas muy difusos que son los responsables del retraso en el diagnóstico de muchos pacientes (Durazzo et al., 2022; Mehta et al., 2024). Aunque la forma clásica de la EC se caracteriza por enteropatía, ha habido un aumento en los casos de EC no clásica, donde los síntomas sistémicos pueden ser confundidos con otras enfermedades (Lindfors et al., 2019; Rej et al., 2020; Lebwohl y Rubio-Tapia, 2021). Las manifestaciones clínicas pueden variar en intensidad y presentación según la edad y otros factores individuales (Tabla 1). INTRODUCCIÓN 16 Tabla 1. Manifestaciones clínicas de la EC según la edad. EC, enfermedad celíaca; SII, síndrome del intestino irritable. Niños Adolescentes Adultos Síntomas Vómitos Asintomáticos Dispepsia Diarreas fétidas Dolor abdominal Diarrea malabsortiva Nauseas Diarrea malabsortiva Dolor abdominal Anorexia Cefaleas Síndrome del intestino irritable (SII) Astenia Artralgias Vómitos Dolor abdominal Menarquia retrasada Estreñimiento Irritabilidad Irregularidades menstruales Astenia Apatía Estreñimiento Dolores articulares y óseos Tristeza Hábito intestinal irregular Infertilidad Introversión Abortos recurrentes Hipotonía Parestesias Ansiedad Depresión Irritabilidad Epilepsia Ataxia Signos y alteraciones analíticas Malnutrición Aftas orales Aftas orales Distensión abdominal Hipoplasia del esmalte Malnutrición con o sin pérdida de peso Hipotrofia muscular Distensión abdominal Edemas periféricos Retraso en la talla Debilidad muscular Talla baja Ferropenia Talla baja Neuropatía periférica Anemia ferropénica Retraso puberal Miopatía proximal Hipoproteinemia Artritis Ferropenia Hipertransaminasemia Osteopenia Anemia ferropénica Pelo frágil Ferropenia Anemia megaloblástica Leucopenia Anemia ferropénica Hipertransaminasemia Dislexia Hipertransaminasemia Hipoesplenismo Autismo Queratosis folicular Osteopenia Hiperactividad Pérdida de peso Disminución del tiempo de protombina Alteraciones de la piel y dentición Déficit de ácido fólico Dermatitis atópica Déficit de vitamina B12 Hipoalbuminemia Menopausia precoz INTRODUCCIÓN 17 En los niños, los síntomas digestivos suelen ser más predominantes e incluyen diarrea crónica, distensión abdominal, dolor abdominal recurrente, vómitos, estreñimiento y pérdida de peso. Además, el retraso en el crecimiento y en el desarrollo, así como la baja estatura, son manifestaciones comunes en niños con EC no diagnosticada o no tratada. Estos síntomas pueden afectar negativamente a la absorción de nutrientes y el desarrollo adecuado en la infancia (Protocolo para el diagnóstico precoz de la enfermedad celíaca, 2018; Bolia y Thapar, 2023). En contraste, los adolescentes y los adultos con EC pueden presentar una variedad de síntomas extradigestivos, lo que a menudo complica el diagnóstico. Estos síntomas pueden incluir fatiga crónica, anemia inexplicada, irritabilidad, cambios de humor, depresión, ansiedad, dolores de cabeza recurrentes, problemas de piel como dermatitis herpetiforme, osteoporosis, artritis, hormigueo o entumecimiento en manos y pies (neuropatía periférica), problemas menstruales en mujeres, infertilidad, pérdida de cabello, defectos en el esmalte dental, aftas bucales recurrentes y úlceras bucales, entre otros (Lindfors et al., 2019; Domsa et al., 2020; Rej et al., 2020; Inchingolo et al., 2024). Es importante destacar que también se han observado casos de EC asintomática, donde los pacientes no experimentan ningún síntoma evidente, pero aun así tienen daño intestinal característico de la enfermedad. Estos casos a menudo se diagnostican accidentalmente durante exámenes médicos de rutina o pruebas de detección de EC en familiares de pacientes diagnosticados (Singh et al., 2018). La exposición continua al gluten puede conducir a complicaciones graves, como la EC refractaria y las neoplasias intestinales y extraintestinales, aumentando el riesgo de mortalidad en pacientes con EC (Ludvigsson, 2012; Holmes y Muirhead, 2018; Koskinen et al., 2020; Lebwohl et al., 2020). La clasificación de Oslo establece varias formas de presentación de la EC, incluyendo clásica, no clásica, sintomática, subclínica, potencial, seronegativa y refractaria (Ludvigsson et al., 2013; Protocolo para el diagnóstico precoz de la enfermedad celíaca, 2018; Domsa et al., 2020; Therrien et al., 2020; Schiepatti et al., 2021a, 2021b): • Clásica: enteropatía inducida por gluten que se presenta con signos y síntomas de malabsorción, incluyendo diarrea crónica con emaciación, retraso de crecimiento y anemia. Es más frecuente en la edad pediátrica. El término “clásica” no implica que este tipo de presentación sea la más común. De hecho, este patrón de presentación es, hoy en día, inusual en la población adulta. • No clásica: pacientes con síntomas digestivos menores, sin malabsorción y manifestaciones extraintestinales, más frecuente en adultos. INTRODUCCIÓN 18 • Latente: pacientes que tomando gluten en el momento de ser evaluados presentan una mucosa “normal”, pero que en un tiempo anterior o posterior padecieron (o desarrollarán) una enteropatía característica. • Potencial: pacientes con mucosa “normal” pero con un riesgo aumentado de desarrollar la EC por presentar serología positiva y HLA compatible. • Subclínica: aquella que está por debajo del umbral de detección clínica, sin signos ni síntomas suficientes para iniciar una búsqueda intencionada de la enfermedad. También engloba a los pacientes diagnosticados a través de programas de cribado poblacional o en estrategias de detección en pacientes con enfermedades que se asocian a la EC. • Seronegativa: para pacientes con daño intestinal, pero resultados serológicos negativos. • Refractaria: persistencia de atrofia vellositaria y malabsorción clínica a pesar de una DSG mantenida de forma correcta durante al menos 12 meses. 2.4. Diagnóstico 2.4.1. Pruebas genéticas La prevalencia de la EC entre los familiares de primer grado es significativamente mayor que en la población general, oscilando entre el 2,6% y el 12,6%, en comparación con el 1% al 2%, respectivamente. Estas cifras destacan el papel importante del componente genético en el desarrollo de la enfermedad (Iversen y Sollid, 2023). El mayor factor de riesgo genético para la EC es la presencia de heterodímeros HLA-DQ2 y -DQ8, que se identifica en el 90% y 5-7% de los pacientes con EC, respectivamente (Ludvigsson y Murray, 2019; Iversen y Sollid, 2023). Debido a que esos alelos se hallan en el 30-40% de la población general (siendo el más común el HLADQ2) la ausencia de éstos es importante por su valor predictivo negativo (VPN) (Sollid, 2002). Un resultado negativo para HLA-DQ2/HLA-DQ8 significa muy baja probabilidad de desarrollo de la enfermedad. Por lo que esta prueba puede emplearse como apoyo al diagnóstico de EC, ya que tiene un alto VPN permitiendo la exclusión de la enfermedad con un 99% de certeza (Protocolo para el diagnóstico precoz de la enfermedad celíaca, 2018) y puede ser de gran utilidad como método de detección en situaciones específicas en sujetos con sospecha de EC, pero con resultados dudosos en otras pruebas (Protocolo para el diagnóstico precoz de la enfermedad celíaca, 2018; Lebwohl y Rubio-Tapia, 2021). 2.4.2. Pruebas serólogicas La medida de anti-transglutaminasa tisular (anti-tTG) junto con IgA total es la primera opción durante la fase de ingesta de gluten, mostrando una sensibilidad y especificidad del 95% INTRODUCCIÓN 19 (Lebwohl y Rubio-Tapia, 2021). Las concentraciones elevadas de este anticuerpo, sobre todo si exceden 10 veces el valor normal, se asocian a un daño histológico y endoscópico más grave y a la presencia de anemia (Ziv-Baran et al., 2021). Por ello, algunas de las guías más recientes han incorporado un enfoque sin biopsia como primera vía para diagnosticar la EC en la población pediátrica (Al-Toma et al., 2019; Husby et al., 2020). Este enfoque elimina la necesidad de una biopsia al permitir la confirmación de EC en niños cuyos títulos anti-tTG se encuentran por encima de 10 veces el valor de referencia, y sean positivos para los anticuerpos anti-endomisio (anti-EMA) en una segunda muestra de sangre. No obstante, es importante tener en cuenta que los anticuerpos anti-tTG pueden dar resultados negativos en un pequeño porcentaje de pacientes con EC confirmada histológicamente. Otros marcadores como los anticuerpos antiendomisio (anti-EMA) y anti-DGP pueden ser útiles en casos específicos, aunque menos recomendados como primera opción, pueden complementar el diagnóstico en algunos casos (Husby et al., 2012; Downey et al., 2015). Los anticuerpos anti-tTG y EMA son menos sensibles en niños menores de 2 años (Rubio-Tapia et al., 2013). En el caso de existir una deficiencia en los niveles de IgA, se recomienda la determinación de anticuerpos IgG anti-DGP y anti-tTG (Protocolo para el diagnóstico precoz de la enfermedad celíaca, 2018; Lebwohl y Rubio-Tapia, 2021). 2.4.3. Pruebas histopatológicas El diagnóstico de la EC se basa en la evaluación histológica mediante biopsia intestinal, reconocida como el estándar de oro para confirmar la enfermedad. Esta técnica revela cambios característicos como el aumento de LIEs, hiperplasia de las criptas y atrofia vellositaria, clasificados según diferentes sistemas, siendo el de Marsh-Oberhuber el más utilizado (Marsh, 1992; Oberhuber et al., 1999). En esta clasificación encontramos diversos grados de lesión: Marsh 0 (mucosa preinfiltrativa); Marsh 1 (lesión infiltrativa e incremento en el número de LIEs); Marsh 2 (incremento en el número de LIEs e hiperplasia de criptas); Marsh 3 (incremento en el número de LIEs, hiperplasia de criptas y atrofia vellositaria [3a] parcial, [3b] subtotal, [3c] total) (Figura 5). Sin embargo, para mejorar la reproducibilidad diagnóstica, Corazza-Villanacci propusieron una clasificación con tres categorías principales (grado A, infiltrativo, grado B1 aplanamiento de vellosidades, B2 vellosidades no detectables) (Corazza y Villanacci, 2005; Corazza et al., 2007). Si bien no existen sugerencias formales sobre cuál de ellas usar, la de Marsh-Oberhuber es la más utilizada por la mayoría de los patólogos, tanto para el diagnóstico como para asegurar la regresión de la lesión tras una DSG (Moscoso y Quera, 2016; Al-Toma et al., 2019). INTRODUCCIÓN 20 Figura 5. Descripción gráfica de la clasificación de Marsh-Oberhuber (modificado según Stages of Coeliac Disease in Upper Jejunum, 2006). En niños y adolescentes, la ESPGHAN recomienda evitar la biopsia en aquellos con niveles elevados de anti-tTG, especialmente si estos niveles son ≥10 veces al valor de referencia, confirmado además con anticuerpos anti-EMA en una segunda muestra de sangre, sin necesidad de considerar el perfil genético o la sintomatología (Husby et al., 2020). Esta estrategia está siendo evaluada en adultos, considerando la eficacia de niveles elevados de anticuerpos antitTG como criterio diagnóstico (Penny et al., 2021). Algunos países como Finlandia y el Reino Unido han adoptado esta práctica en ciertos casos, mientras otras sociedades de gastroenterología aún recomiendan la biopsia intestinal para evitar errores diagnósticos y asegurar el manejo adecuado de la enfermedad (Reilly et al., 2018; Husby et al., 2019). 3. TRATAMIENTO DE LA ENFERMEDAD CELÍACA 3.1. Terapia dietética: dieta libre de gluten 3.1.1. Dieta sin gluten El único tratamiento eficaz y seguro para los pacientes con EC consiste en la exclusión permanente del gluten de la dieta (Aljada et al., 2021; Kivelä et al., 2021; Mazzola et al., 2024). La DSG estricta permite revertir el daño de la mucosa intestinal, reducir la inflamación y los niveles de anticuerpos, mejorar la sintomatología propia de la enfermedad y prevenir el desarrollo de complicaciones, asociadas a la exposición prolongada al gluten (Caio et al., 2019; Lindfors et al., 2019; Therrien et al., 2020). El período de recuperación varía, siendo aproximadamente 1-2 años en niños y puede ser más prolongado en el caso de pacientes adultos (Rubio-Tapia y Murray, 2010; Ludvigsson et al., 2014). Cuando se inicia una DSG, se recomienda principalmente consumir alimentos naturales libres de gluten, como frutas, verduras, legumbres, cereales y pseudocereales sin gluten (arroz, maíz, mijo, sorgo, trigo sarraceno, amaranto y quinoa), tubérculos, carne, pescado, frutos secos INTRODUCCIÓN 27 endoscopia oral y la toma de biopsias del duodeno. Sin embargo, una vez instaurada la DSG, algunos pacientes pueden normalizar los niveles de anticuerpo y la respuesta clínica mientras mantienen daños histológicos (Ruiz-Canicer et al, 2020; Galli et al., 2021). El proceso completo de recuperación de la mucosa intestinal puede prolongarse durante varios años después de la adopción de la DSG. Se ha observado que este proceso es más corto en niños, con una recuperación completa en la mayoría de los casos en 1-2 años, mientras que en adultos puede requerir aproximadamente tres años en promedio, y en algunos casos, no se logra la remisión completa (Ludvigsson et al., 2014; Fernández-Bañares et al., 2021). Aunque la biopsia intestinal es altamente precisa, su naturaleza invasiva y costosa no la hace ideal para el seguimiento rutinario de pacientes con síntomas remitidos y serología normalizada (Raju et al., 2024). Generalmente, se reserva para aquellos pacientes con respuesta clínica inadecuada o síntomas recurrentes a pesar de una adherencia aparentemente adecuada a la DSG (Rubio-Tapia et al., 2013; Ludvigsson et al., 2014). En la evaluación de la recuperación de la mucosa intestinal, se pueden usar métodos cuantitativos como el puntaje Elli-Ferrero y la ratio VH:CD, junto con el recuento de LIEs por cada 100 células epiteliales. Estos métodos ofrecen una evaluación objetiva y reproducible de los cambios mucosos (Adelman et al., 2018; Lähdeaho et al., 2019). A pesar de su utilidad, su aplicación clínica aún se limita por la falta de conocimiento y el tiempo requerido. Sin embargo, se han desarrollado algoritmos como Quantitative-Mucosal Algorithmic Rule for Scoring Histology (Q-MARSH) para estandarizar la clasificación histológica (Adelman et al., 2018). 3.1.2.5. Detección de péptidos inmunogénicos de gluten en muestras humanas Los GIP son resistentes a la digestión gastrointestinal y responsables de reacciones inmunogénicas en las células T de pacientes con EC (Cebolla et al., 2018). A diferencia de los métodos tradicionales para monitorizar la adherencia a la DSG, que solo evalúan las consecuencias de las transgresiones de la DSG, este método no invasivo permite una evaluación directa y cuantitativa de la exposición al gluten (Comino et al., 2012b, 2016b, Moreno et al., 2017; Syage et al., 2018). Los GIP pueden ser evaluados en heces y orina mediante sistemas comerciales (ELISA y LFIA) basados en los AcMos G12 y A1. Estas metodologías se han propuesto como nuevos biomarcadores no invasivos para detectar la ingesta de gluten y verificar la adherencia a la DSG en pacientes con EC. La Tabla 4 presenta una recopilación exhaustiva de los estudios que respaldan y validan estas metodologías. Estos estudios proporcionan una sólida base de evidencia científica que respalda la eficacia y fiabilidad de los GIP para detectar la ingesta INTRODUCCIÓN 28 de gluten y verificar la adherencia a la DSG en pacientes con EC (Comino et al., 2016b; Soler et al., 2016; Moreno et al., 2017; Cebolla et al., 2018; Gerasimidis et al., 2018) Los análisis de GIP en heces (ELISA y LFIA) fueron los primeros en describirse como métodos novedosos para detectar el consumo de gluten durante el diagnóstico y seguimiento de los pacientes con EC (Comino et al., 2012b). El test LFIA ofrece una mayor simplicidad para su uso en el ámbito clínico, mientras que el ELISA se propone como una opción para el laboratorio cuando el objetivo es alcanzar una mayor sensibilidad o la cuantificación de los GIP (Comino et al., 2012b). Este método fue validado para su uso clínico en un estudio multicéntrico con niños y adultos con EC que llevaban un año a DSG. La detección de GIP en heces demostró que un 30% de los pacientes estaban expuestos al gluten, mientras que la serología y los cuestionarios dietéticos sólo consiguieron detectar un 18%. Sorprendentemente, el 70% de los pacientes que no eran conscientes de haber ingerido gluten según el cuestionario dietético obtuvieron resultados positivos de GIP en heces, lo que confirma las dificultades que tienen los pacientes con EC para adherirse a la DSG correctamente y/o al menos transmitir correctamente la información al dietista, ya que en la mayoría de los casos la exposición al gluten no fue comunicada (Comino et al., 2016b). Por otro lado, Moreno et al. (2017) mostraron la utilidad de la detección de los GIP en muestras de orina mediante LFIA. En este estudio se incluyeron 58 pacientes entre 3 y 64 años y los resultados revelaron un alto porcentaje de transgresiones dietéticas, ya que los GIP fueron detectados en el 48% de adultos y 45% de niños. Además, Soler et al. (2016), desarrollaron un biosensor SPR (Surface Plasmon Resonance) altamente sensible para la detección de GIP en orina. Realizaron un estudio que incluyó muestras de orina de pacientes con diferentes patrones dietéticos: una DSG (n=4), una dieta con bajo consumo de gluten (n=4), y una dieta normal con consumo alto/moderado de gluten (n=4). Los resultados obtenidos demostraron la capacidad del biosensor para detectar la ingesta de gluten en individuos sanos y con EC con diferentes niveles de consumo. Posteriormente, la determinación de los GIP en heces y orina se incluyó en diversos estudios clínicos a nivel nacional e internacional, demostrando su utilidad en la monitorización de la DSG a largo plazo con respecto a la serología y los cuestionarios dietéticos. Así, por ejemplo, en un estudio que incluyeron niños y adolescentes con EC se demostró que la mayoría de ellos tenían buena adherencia a la DSG. Según los cuestionarios dietéticos el grado de adherencia fue superior al 90% de los pacientes, pero la presencia de GIP en heces (8-35%) mostró que una parte de los considerados como adherentes están expuestos al gluten. Los anticuerpos anti-tTG tampoco fueron capaces de detectar la mayoría de esos casos (Gerasimidis et al., 2018). Además, se comprobó que, generalmente, los cuestionarios fallan en detectar la ingesta de INTRODUCCIÓN 29 gluten en comparación con los GIP, ya que la mayoría de los pacientes no adherentes según los cuestionarios tenían presencia de GIP en heces, y al contrario solo sucedió en algunos casos. En pacientes pediátricos, se ha observado que, a pesar de una alta adherencia a la DSG según los cuestionarios dietéticos, la presencia de GIP en heces indica una exposición al gluten en un porcentaje significativo de casos (Gerasimidis et al., 2018). Esto resalta la importancia de implementar métodos de monitorización más precisos y objetivos para evaluar la adherencia a la DSG en esta población. En contraste, en pacientes adultos, se ha observado que las transgresiones dietéticas son más frecuentes, y la correlación entre los cuestionarios dietéticos y la presencia de GIP en heces u orina es más baja (Costa et al., 2019). Estos resultados subrayan la necesidad de estrategias de monitorización más efectivas en esta población, dada la prevalencia de transgresiones dietéticas y la importancia de un manejo adecuado de la enfermedad. En resumen, la detección de GIP en heces y orina mediante inmunoensayos ha emergido como una herramienta valiosa para evaluar la adherencia a la DSG y monitorizar la exposición al gluten en pacientes con EC. Ofrecen una evaluación objetiva de la adherencia a la DSG y permiten una vigilancia continua de la respuesta del paciente, detectando transgresiones dietéticas y facilitando intervenciones tempranas para prevenir complicaciones a largo plazo. 4. BILBLIOGRAFÍA Adelman, D. C., Murray, J., Wu, T. T., Mäki, M., Green, P. H., & Kelly, C. P. (2018). Measuring change in small intestinal histology in patients with celiac disease. The American Journal of Gastroenterology, 113(3), 339–347. Aljada, B., Zohni, A., & El-Matary, W. (2021). The gluten-free diet for celiac disease and beyond. Nutrients, 13(11), 3993. Al-Toma, A., Volta, U., Auricchio, R., Castillejo, G., Sanders, D. S., Cellier, C., Mulder, C. J., & Lundin, K. E. A. (2019). European Society for the Study of Coeliac Disease (ESsCD) guideline for coeliac disease and other gluten-related disorders. United European gastroenterology journal, 7(5), 583–613. Aronsson, C., Lee, H. S., Hård Af Segerstad, E. M., Uusitalo, U., Yang, J., Koletzko, S., Liu, E., Kurppa, K., Bingley, P. J., Toppari, J., Ziegler, A. G., She, J. X., Hagopian, W. A., Rewers, M., Akolkar, B., Krischer, J. P., Virtanen, S. M., Norris, J. M., Agardh, D., & TEDDY Study Group. INTRODUCCIÓN 30 (2019). Association of gluten intake during the first 5 years of life with incidence of celiac disease autoimmunity and celiac disease among children at increased risk. JAMA, 322(6), 514–523. Aronsson, C., & Agardh, D. (2023). Intervention strategies in early childhood to prevent celiac disease-a mini-review. Frontiers in immunology, 14, 1106564. Atsawarungruangkit, A., Silvester, J. A., Weiten, D., Green, K. L., Wilkey, K. E., Rigaux, L. N., Bernstein, C. N., Graff, L. A., Walker, J. R., & Duerksen, D. R. (2020). Development of the dietitian integrated evaluation tool for gluten-free diets (DIET-GFD). Nutrition (Burbank, Los Angeles County, Calif.), 78, 110819. Bai, J. C., & Ciacci, C. (2017). World Gastroenterology Organisation Global Guidelines: Celiac Disease February 2017. Journal of clinical gastroenterology, 51(9), 755–768. Bascuñán, K. A., Roncoroni, L., Branchi, F., Doneda, L., Scricciolo, A., Ferretti, F., Aray, M. & Elli, L. (2018). The 5 Ws of a gluten challenge for gluten-related disorders. Nutrition Reviews, 76(2), 79–87. Besser, H. A., & Khosla, C. (2023). Celiac disease: mechanisms and emerging therapeutics. Trends in pharmacological sciences, 44(12), 949–962. Biagi, F., Campanella, J., Martucci, S., Pezzimenti, D., Ciclitira, P. J., Ellis, H. J., & Corazza, G. R. (2004). A milligram of gluten a day keeps the mucosal recovery away: a case report. Nutrition reviews, 62(9), 360–363. Biesiekierski, J. R. (2017). What is gluten? Journal of Gastroenterology and Hepatology, 32(1), 78–81. Bolia, R., & Thapar, N. (2024). Celiac disease in children: A 2023 Update. Indian journal of pediatrics, 91(5), 481–489. Bruins, I. D., Bremer, M. G., van der Fels-Klerx, I., & Hamer, R. J. (2015). Evaluating the performance of gluten ELISA test kits: The numbers do not tell the tale. Cereal Chemistry, 92, 513–521. Cabanillas, B. (2020). Gluten-related disorders: Celiac disease, wheat allergy, and nonceliac gluten sensitivity. Critical Reviews in Food Science and Nutrition, 60(15), 2606–2621. Caio, G., Volta, U., Sapone, A., Leffler, D. A., De Giorgio, R., Catassi, C., & Fasano, A. (2019). Celiac disease: a comprehensive current review. BMC medicine, 17(1), 142. Cárdenas-Torres, F. I., Cabrera-Chávez, F., Figueroa-Salcido, O. G., & Ontiveros, N. (2021). Non- Celiac Gluten Sensitivity: An Update. Medicina (Kaunas), 57(6):526. Cárdenas-Torres, F. I., Cabrera-Chávez, F., Figueroa-Salcido, O. G., & Ontiveros, N. (2021). Non- Celiac Gluten Sensitivity: An Update. Medicina (Kaunas, Lithuania), 57(6), 526. Catassi, C., Fabiani, E., Iacono, G., D'Agate, C., Francavilla, R., Biagi, F., Volta, U., Accomando, S., Picarelli, A., De Vitis, I., Pianelli, G., Gesuita, R., Carle, F., Mandolesi, A., Bearzi, I., & Fasano, INTRODUCCIÓN 31 A. (2007). A prospective, double-blind, placebo-controlled trial to establish a safe gluten threshold for patients with celiac disease. The American Journal of Clinical Nutrition, 85(1), 160– 166. Cebolla, Á., Moreno, M., Coto, L., Sousa, C. (2018). Gluten immunogenic peptides as standard for the evaluation of potential harmful prolamin content in food and human specimen. Nutrients, 10(12), 1927. Cenni, S., Sesenna, V., Boiardi, G., Casertano, M., Russo, G., Reginelli, A., Esposito, S., & Strisciuglio, C. (2023). The role of gluten in gastrointestinal disorders: a review. Nutrients, 15(7), 1615. Choung, R. S., Ditah, I. C., Nadeau, A. M., Rubio-Tapia, A., Marietta, E. V., Brantner, T. L., Camilleri, M. J., Rajkumar, S. V., Landgren, O., Everhart, J. E., & Murray, J. A. (2015) Trends and racial/ethnic disparities in gluten-sensitive problems in the United States: findings from the national health and nutrition examination surveys from 1988 to 2012. American Journal of Gastroenterology, 110(3), 455–461. Cohen, I.S., Day, A.S., & Shaoul, R. (2019) Gluten in celiac disease more or less? Rambam Maimonides Medical Journal, 10(1), 1–6. Cohen, R., Mahlab-Guri, K., Atali, M., & Elbirt, D. (2023). Viruses and celiac disease: what do we know ?. Clinical and experimental medicine, 23(7), 2931–2939. Collin, P., Vilppula, A., Luostarinen, L., Holmes, G. K. T., & Kaukinen, K. (2018). Review article: coeliac disease in later life must not be missed. Alimentary Pharmacology and Therapeutics, 47(5), 563–572. Comino, I., Real, A., de Lorenzo, L., Cornell, H., López-Casado, M. Á., Barro, F., Lorite, P., Torres, M. I., Cebolla, A., & Sousa, C. (2011). Diversity in oat potential immunogenicity: basis for the selection of oat varieties with no toxicity in coeliac disease. Gut, 60(7), 915–922. Comino, I., Real, A., Gil-Humanes, J., Pistón, F., de Lorenzo, L., Moreno, M.deL., López-Casado, M. Á., Lorite, P., Cebolla, A., Torres, M. I., Barro, F., & Sousa, C. (2012a). Significant differences in coeliac immunotoxicity of barley varieties. Molecular nutrition & food research, 56(11), 1697– 1707. Comino, I., Real, A., Vivas, S., Síglez, M. Á., Caminero, A., Nistal, E., Casqueiro, J., Rodríguez- Herrera, A., Cebolla, A., & Sousa, C. (2012b). Monitoring of gluten-free diet compliance in celiac patients by assessment of gliadin 33-mer equivalent epitopes in feces. The American journal of clinical nutrition, 95(3), 670–677. INTRODUCCIÓN 32 Comino, I., Bernardo, D., Bancel, E., de Lourdes Moreno, M., Sánchez, B., Barro, F., Šuligoj, T., Ciclitira, P. J., Cebolla, Á., Knight, S. C., Branlard, G., & Sousa, C. (2016a). Identification and molecular characterization of oat peptides implicated on coeliac immune response. Food & nutrition research, 60, 30324. Comino, I., Fernández-Bañares, F., Esteve, M., Ortigosa, L., Castillejo, G., Fambuena, B., Ribes- Koninckx, C., Sierra, C., Rodríguez-Herrera, A., Salazar, J. C., Caunedo, Á., Marugán-Miguelsanz, J. M., Garrote, J. A., Vivas, S., Lo Iacono, O., Nuñez, A., Vaquero, L., Vegas, A. M., Crespo, L., Fernández-Salazar, L., … Sousa, C. (2016b). Fecal gluten peptides reveal limitations of serological tests and food questionnaires for monitoring gluten-free diet in celiac disease patients. The American journal of gastroenterology, 111(10), 1456–1465. Corazza, G.R., & Villanacci, V. (2005). Coeliac disease. Journal of Clinical Pathology, 58(6), 573– 574. Corazza, G. R., Villanacci, V., Zambelli, C., Milione, M., Luinetti, O., Vindigni, C., Chioda, C., Albarello, L., Bartolini, D., & Donato, F. (2007). Comparison of the interobserver reproducibility with different histologic criteria used in celiac disease. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association, 5(7), 838– 843. Costa, A. F., Sugai, E., Temprano, M. P., Niveloni, S. I., Vázquez, H., Moreno, M. L., Domínguez- Flores, M. R., Muñoz-Suano, A., Smecuol, E., Stefanolo, J. P., González, A. F., Cebolla-Ramirez, A., Mauriño, E., Verdú, E. F., & Bai, J. C. (2019). Gluten immunogenic peptide excretion detects dietary transgressions in treated celiac disease patients. World journal of gastroenterology, 25(11), 1409–1420. D’Avino, P., Serena, G., Kenyon, V., & Fasano, A. (2021). An updated overview on celiac disease: from immuno-pathogenesis and immuno-genetics to therapeutic implications. Expert Review of Clinical Immunology, 17 (3), 269–284. Domsa, E. M., Berindan-Neagoe, I., Para, I., Munteanu, L., Matei, D., & Andreica, V. (2020). Celiac disease: a multi-faceted medical condition. Journal of physiology and pharmacology: an official journal of the Polish Physiological Society, 71 (1), 1–12. Downey, L., Houten, R., Murch, S., Longson, D., & Guideline Development Group. (2015). Recognition, assessment, and management of coeliac disease: summary of updated NICE guidance. BMJ (Clinical research ed.), 351, h4513. Durazzo, M., Ferro, A., Brascugli, I., Mattivi, S., Fagoonee, S., & Pellicano, R. (2022). Extraintestinal manifestations of celiac disease: what should we know in 2022? Journal of clinical medicine, 11(1), 258. INTRODUCCIÓN 33 Dydensborg Sander, S., Nybo Andersen, A.M., Murray, J.A., Karlstad, Ø., Husby, S., & Størdal, K. (2019). Association between antibiotics in the first year of life and celiac disease. Gastroenterology, 156(8), 2217-2229. Elli, L., Leffler, D., Cellier, C., Lebwohl, B., Ciacci, C., Schumann, M., Lundin, K. E. A., Chetcuti Zammit, S., Sidhu, R., Roncoroni, L., Bai, J. C., Lee, A. R., Dennis, M., Robert, M. E., Rostami, K., Khater, S., Comino, I., Cebolla, A., Branchi, F., Verdu, E. F., … Sanders, D. S. (2024). Guidelines for best practices in monitoring established coeliac disease in adult patients. Nature reviews. Gastroenterology & hepatology, 21(3), 198–215. Expósito Miranda, M., García-Valdés, L., Espigares-Rodríguez, E., Leno-Durán, E., & Requena, P. (2023). Non-celiac gluten sensitivity: Clinical presentation, etiology and differential diagnosis. Sensibilidad al gluten no celiaca: etiología, diagnóstico diferencial y presentación clínica. Gastroenterologia y hepatologia, 46(7), 562–571. Fasano, A., Sapone, A., Zevallos, V., & Schuppan, D. (2015). Nonceliac gluten sensitivity. Gastroenterology, 148(6), 1195-204. Fernández-Bañares, F., Beltrán, B., Salas, A., Comino, I., Ballester-Clau, R., Ferrer, C., Molina- Infante, J., Rosinach, M., Modolell, I., Rodríguez-Moranta, F., Arau, B., Segura, V., Fernández- Salazar, L., Santolaria, S., Esteve, M., Sousa, C., & CADER study group (2021). Persistent villous atrophy in de novo adult patients with celiac disease and strict control of gluten-free diet adherence: a multicenter prospective study (CADER Study). The American journal of gastroenterology, 116(5), 1036–1043. Fueyo-Díaz, R., Magallón-Botaya, R., Masluk, B., Palacios-Navarro, G., Asensio-Martínez, A., Gascón-Santos, S., Olivan-Blázquez, B., & Sebastián-Domingo, J. J. (2019) Prevalence of celiac disease in primary care: The need for its own code. BMC Health Services Research. 19(1), 1–9. Galli, G., Carabotti, M., Pilozzi, E., Lahner, E., Annibale, B., & Conti, L. (2021) Relationship between persistent gastrointestinal symptoms and duodenal histological findings after adequate gluten-free diet: A gray area of celiac disease management in adult patients. Nutrients, 13(2), 600. García-Molina, M.D., Giménez, M.J., Sánchez-León, S., & Barro, F. (2019). Gluten free wheat: Are we there? Nutrients, 11(3), 1–13. Gerasimidis, K., Zafeiropoulou, K., Mackinder, M., Ijaz, U. Z., Duncan, H., Buchanan, E., Cardigan, T., Edwards, C. A., McGrogan, P., & Russell, R. K. (2018). Comparison of clinical methods with the faecal gluten immunogenic peptide to assess gluten intake in coeliac disease. Journal of Pediatric Gastroenterology & Nutrition. 67 (3), 356–360. Garzón-Benavides, M., Ruiz-Carnicer, Á., Segura, V., Fombuena, B., García-Fernandez, F., Sobrino-Rodriguez, S., Gómez-Izquierdo, L., Montes-Cano, M. A., Millan-Domínguez, R., Del INTRODUCCIÓN 34 Carmen Rico, M., González-Naranjo, C., Bozada-García, J. M., Coronel-Rodríguez, C., Espin, B., Díaz, J., Comino, I., Argüelles-Arias, F., Cebolla, Á., Romero-Gómez, M., Rodriguez-Herrera, A., … Pizarro-Moreno, Á. (2023). Clinical utility of urinary gluten immunogenic peptides in the follow-up of patients with coeliac disease. Alimentary pharmacology & therapeutics, 57(9), 993– 1003. Gilbert, J. A., Blaser, M. J., Caporaso, J. G., Jansson, J. K., Lynch, S. V., & Knight, R. (2018). Current understanding of the human microbiome. Nature Medicine, 24(4), 392–400. Gómez-Rial, J., Rivero-Calle, I., Salas, A., Martinón-Torres, F. (2020). Rotavirus and autoimmunity. The Journal of infection, 81(2), 183–189. Grupo de trabajo del Protocolo para el diagnóstico precoz de la enfermedad celíaca (2018) Protocolo para el diagnóstico precoz de la enfermedad celíaca. Ministerio de Sanidad, Servicios Sociales e Igualdad. Servicio de Evaluación del Servicio Canario de la Salud (SESCS). Disponible en: https://www.sanidad.gob.es/profesionales/prestacionesSanitarias/publicaciones/Celiaquia/enf ermedadCeliaca.pdf Güven, E., & Azizoglu, R. O. (2023). Enhancing gluten detection assay development through optimization of gliadin extraction conditions. Heliyon, 9(9), e19432. Hadjivassiliou, M., Sanders, D. D., & Aeschlimann, D. P. (2015). Gluten-related disorders: gluten ataxia. Digestive diseases (Basel, Switzerland), 33(2), 264-268. Hadjivassiliou, M., Grünewald, R. A., Sanders, D. S., Shanmugarajah, P., & Hoggard, N. 2017). Effect of gluten-free diet on cerebellar MR spectroscopy in gluten ataxia. Neurology, 89(7), 705- 709. Holmes, G.K.T., & Muirhead, A. (2018). Mortality in coeliac disease: A population-based cohort study from a single centre in Southern Derbyshire, UK. BMJ Open Gastroenterology, 5 (1). Husby, S., Koletzko, S., Korponay-Szabó, I. R., Mearin, M. L., Phillips, A., Shamir, R., Troncone, R., Giersiepen, K., Branski, D., Catassi, C., Lelgeman, M., Mäki, M., Ribes-Koninckx, C., Ventura, A., Zimmer, K. P., ESPGHAN Working Group on Coeliac Disease Diagnosis, ESPGHAN Gastroenterology Committee, & European Society for Pediatric Gastroenterology, Hepatology, and Nutrition. (2012). European society for pediatric gastroenterology, hepatology, and nutrition guidelines for the diagnosis of coeliac disease. Journal of pediatric gastroenterology and nutrition, 54(1), 136–160. Husby, S., Murray, J.A. & Katzka, D.A. (2019). AGA Clinical practice update on diagnosis and monitoring of celiac disease-changing utility of serology and histologic measures: Expert review. Gastroenterology, 156(4), 885–889. Husby, S., Koletzko, S., Korponay-Szabó, I., Kurppa, K., Mearin, M. L., Ribes-Koninckx, C., Shamir, R., Troncone, R., Auricchio, R., Castillejo, G., Christensen, R., Dolinsek, J., Gillett, P., INTRODUCCIÓN 35 Hróbjartsson, A., Koltai, T., Maki, M., Nielsen, S. M., Popp, A., Størdal, K., Werkstetter, K., … Wessels, M. (2020) European society paediatric gastroenterology, hepatology and nutrition guidelines for diagnosing coeliac disease 2020. Journal of Pediatric Gastroenterology and Nutrition, 70 (1), 141–156. Husby, S., & Bai, J. C. (2019). Follow-up of celiac disease. Gastroenterol Clin North Am, 48(1), 127- 136. Inchingolo, A. D., Dipalma, G., Viapiano, F., Netti, A., Ferrara, I., Ciocia, A. M., Mancini, A., Di Venere, D., Palermo, A., Inchingolo, A. M., & Inchingolo, F. (2024). Celiac Disease-Related Enamel Defects: A Systematic Review. Journal of clinical medicine, 13(5), 1382. Iversen, R., Amundsen, S. F., Kleppa, L., du Pré, M. F., Stamnaes, J., & Sollid, L. M. (2020). Evidence that pathogenic transglutaminase 2 in celiac disease derives from enterocytes. Gastroenterology, 159(2), 788–790. Iversen, R., & Sollid, L. M. (2023). The Immunobiology and Pathogenesis of Celiac Disease. Annual review of pathology, 18, 47–70. Jansson-Knodell, C. L., Hujoel, I. A., West, C. P., Taneja, V., Prokop, L. J., Rubio-Tapia, A., & Murray, J. A. (2019). Sex difference in celiac disease in undiagnosed populations: a systematic review and meta-analysis. Clinical gastroenterology and hepatology, 17(10), 1954–1968.e13. Jansson-Knodell, C. L., Celdir, M. G., Hujoel, I. A., Lyu, R., Gardinier, D., Weekley, K., Prokop, L. J., & Rubio-Tapia, A. (2023). Relationship between gluten availability and celiac disease prevalence: A geo-epidemiologic systematic review. Journal of gastroenterology and hepatology, 38(10), 1695–1709. Khan, A., Suarez, M. G., & Murray, J. A. (2020). Nonceliac gluten and wheat sensitivity. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association, 18(9), 1913–1922.e1. King, J. A., Jeong, J., Underwood, F. E., Quan, J., Panaccione, N., Windsor, J. W., Coward, S., deBruyn, J., Ronksley, P. E., Shaheen, A. A., Quan, H., Godley, J., Veldhuyzen van Zanten, S., Lebwohl, B., Ng, S. C., Ludvigsson, J. F., & Kaplan, G. G. (2020). Incidence of celiac disease is increasing over time: a systematic review and meta-analysis. The American journal of gastroenterology, 115(4), 507–525. Kivelä, L., Caminero, A., Leffler, D. A., Pinto-Sanchez, M. I., Tye-Din, J. A., & Lindfors, K. (2021). Current and emerging therapies for coeliac disease. Nature reviews. Gastroenterology & hepatology, 18(3), 181–195. Koskinen, I., Virta, L. J., Huhtala, H., Ilus, T., Kaukinen, K., & Collin, P. (2020). Overall and causespecific mortality in adult celiac disease and dermatitis herpetiformis diagnosed in the 21st Century. The American journal of gastroenterology, 115(7), 1117–1124. INTRODUCCIÓN 36 Lähdeaho, M. L., Scheinin, M., Vuotikka, P., Taavela, J., Popp, A., Laukkarinen, J., Koffert, J., Koivurova, O. P., Pesu, M., Kivelä, L., Lovró, Z., Keisala, J., Isola, J., Parnes, J. R., Leon, F., & Mäki, M. (2019). Safety and efficacy of AMG 714 in adults with coeliac disease exposed to gluten challenge: a phase 2a, randomised, double-blind, placebo-controlled study. The lancet. Gastroenterology & hepatology, 4(12), 948–959. Lebwohl, B., Sanders, D.S., Green, P.H.R. (2018). Seminar coeliac disease. The Lancet, 391 (10115), 70–81. Lebwohl, B., Green, P. H. R., Söderling, J., Roelstraete, B., & Ludvigsson, J. F. (2020). Association between celiac disease and mortality risk in a swedish population. JAMA, 323(13), 1277–1285. Lebwohl, B., Rubio-Tapia, A. (2021). Epidemiology, presentation, and diagnosis of celiac disease. Gastroenterology, 160 (1), 63–75. Leffler, D. A., Dennis, M., Edwards George, J., Jamma, S., Cook, E. F., Schuppan, D., & Kelly, C. P. (2009). A validated disease-specific symptom index for adults with celiac disease. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association, 7(12), 1328–1334.e13343. Lindfors, K., Ciacci, C., Kurppa, K., Lundin, K. E. A., Makharia, G. K., Mearin, M. L., Murray, J. A., Verdu, E. F., & Kaukinen, K. (2019). Coeliac disease. Nature reviews. Disease primers, 5(1), 3. Logan, K., Perkin, M. R., Marrs, T., Radulovic, S., Craven, J., Flohr, C., Bahnson, H. T., & Lack, G. (2020). Early gluten introduction and celiac disease in the eat study: a prespecified analysis of the EAT randomized clinical trial. JAMA pediatrics, 174(11), 1041–1047. López Casado, M. Á., Lorite, P., Ponce de León, C., Palomeque, T., & Torres, M. I. (2018). Celiac disease autoimmunity. Archivum immunologiae et therapiae experimentalis, 66(6), 423–430. Ludvigsson J. F. (2012). Mortality and malignancy in celiac disease. Gastrointestinal endoscopy clinics of North America, 22(4), 705–722. Ludvigsson, J. F., Leffler, D. A., Bai, J. C., Biagi, F., Fasano, A., Green, P. H., Hadjivassiliou, M., Kaukinen, K., Kelly, C. P., Leonard, J. N., Lundin, K. E., Murray, J. A., Sanders, D. S., Walker, M. M., Zingone, F., & Ciacci, C. (2013). The Oslo definitions for coeliac disease and related terms. Gut, 62(1), 43–52. Ludvigsson, J. F., Bai, J. C., Biagi, F., Card, T. R., Ciacci, C., Ciclitira, P. J., Green, P. H., Hadjivassiliou, M., Holdoway, A., van Heel, D. A., Kaukinen, K., Leffler, D. A., Leonard, J. N., Lundin, K. E., McGough, N., Davidson, M., Murray, J. A., Swift, G. L., Walker, M. M., Zingone, F., … British Society of Gastroenterology. (2014). Diagnosis and management of adult coeliac disease: guidelines from the British Society of Gastroenterology. Gut, 63(8), 1210–1228. Ludvigsson, J. F., Ciacci, C., Green, P. H., Kaukinen, K., Korponay-Szabo, I. R., Kurppa, K., Murray, J. A., Lundin, K. E. A., Maki, M. J., Popp, A., Reilly, N. R., Rodriguez-Herrera, A., Sanders, D. S., INTRODUCCIÓN 43 Ziv-Baran, T., Dubov, Y., Weinberger, R., Guz-Mark, A., Shamir, R., & Assa, A. (2021). Anti-tissue transglutaminase titers are associated with endoscopic findings and severity of mucosal damage in children with celiac disease. European journal of pediatrics, 180(1), 263–269. 44 CAPÍTULO 2. ANTECEDENTES DEL TEMA Y OBJETIVOS 45 ANTECEDENTES DEL TEMA Y OBJETIVOS 46 CAPÍTULO 2 El único tratamiento disponible para la EC implica una estricta adherencia a la DSG durante toda la vida, con el fin de reducir y/o eliminar los síntomas y lograr la recuperación de la mucosa (Lindfors et al., 2019; Aljada et al., 2021; Kivelä et al., 2021). Sin embargo, la adherencia a la DSG presenta desafíos significativos debido a la ubicuidad del gluten en los alimentos, la desinformación educativa, los problemas del etiquetado de productos y el riesgo de contaminación cruzada (Caio et al., 2019; Therrien et al., 2020; Vargas et al., 2024). Por ello, es fundamental desarrollar nuevas estrategias para mejorar el control de la DSG. Esto implica desarrollar metodologías innovadoras que garanticen la cuantificación precisa del gluten tóxico en los alimentos destinados a pacientes con EC, asegurando su efectiva extracción y evaluación. Además, se requiere el desarrollo de herramientas destinadas especializadas para monitorizar la adherencia a la DSG de los pacientes con EC. Las proteínas en los alimentos se modifican durante su fabricación mediante diversos procesos para mejorar su funcionalidad y aumentar su aplicación en la industria alimentaria. Estas modificaciones incluyen principalmente la desamidación, la transamidación y la degradación por diferentes tipos de hidrólisis. Además, las modificaciones también pueden ocurrir de forma natural debido a enzimas presentes en las semillas de cereales que se liberan cuando las células se descomponen durante el procesamiento. La hidrólisis puede alterar la toxicidad del gluten, lo cual puede complicar su extracción y análisis en los alimentos. En el caso específico de las cervezas, normalmente elaboradas de cebada y/o trigo malteado se ha demostrado la presencia de péptidos derivados de la hidrolisis de las proteínas del gluten (Comino et al., 2013; Real et al., 2014), la cual ocurre durante el proceso de elaboración de la cerveza debido a la acción de proteasas endógenas activadas durante la germinación del cereal y la fermentación por las levaduras cerveceras (Rostami et al., 2017; Yang et al., 2022). La heterogeneidad de los péptidos derivados del gluten dificultan la determinación precisa de la toxicidad del producto final (Comino et al., 2013; Real et al., 2014). Por lo tanto, se es crucial desarrollar métodos que permitan analizar estos productos de manera sencilla y precisa, con el fin de controlar su potencial riesgo para los pacientes con EC. Asimismo, es esencial asegurar la extracción completa del gluten en las matrices alimentarias. Los métodos convencionales de extracción de gluten no son eficaces para muestras de alimentos procesados debido a la complejidad de su composición y la presencia de ingredientes y/o aditivos que interfieren en el proceso de extracción (Xhaferaj et al., 2020). Por consiguiente, disponer de un ANTECEDENTES DEL TEMA Y OBJETIVOS 47 método eficiente que garantice la total extracción del gluten de las muestras es fundamental para una cuantificación precisa de estas proteínas. Varios estudios han demostrado que entre un 16 y un 70% de los pacientes con EC no siguen adecuadamente el tratamiento, siendo la falta de adherencia rigurosa en la dieta la principal causa (Comino et al., 2012, 2016, 2019; Silvester et al., 2016, 2020a, 2020b; Moreno et al., 2017; Muhammad et al., 2017; Gerasimidis et al., 2018; Fueyo-Diaz et al., 2016; Dana et al., 2020; Ruiz-Carnicer et al., 2020; Segura et al., 2021; Coto et al., 2021). Este comportamiento puede conducir a un deterioro en la calidad de vida y a complicaciones a largo plazo como anemia, osteoporosis e incluso un mayor riesgo de desarrollar linfomas o carcinomas (Husby y Bai, 2019; Lenti et al., 2019). Las herramientas actuales para el control de la DSG se basan en la determinación de la serología, el control de los síntomas y la realización de entrevistas o encuestas dietéticas. Sin embargo, ninguna de estas ofrece una medida precisa sobre las transgresiones dietéticas y la adherencia a la DSG (Silvester et al., 2016, 2020a, 2020b; Mahadev et al., 2017; Caio et al., 2019; Husby y Bai, 2019; Skodje et al., 2022; Russell et al., 2024). La biopsia duodenal, aunque es el estándar de oro para el diagnóstico de la EC, es una técnica invasiva y costosa, por lo que no se recomienda para el seguimiento rutinario de la DSG (Silvester et al., 2017; Husby et al., 2019; Raju et al., 2024). El desarrollo de nuevas técnicas de monitorización de la DSG que sean no invasivas, precisas y fiables podría mejorar la identificación de pacientes con EC que están expuestos regularmente al gluten, ya sea de manera voluntaria o involuntaria. A este respecto, los AcMos G12 y A1 han sido utilizados en el desarrollo de inmunoensayos para la detección de GIP en muestras de heces y orina, permitiendo así determinar el consumo inadvertido de gluten y las transgresiones en la dieta. A diferencia de los métodos tradicionales que sólo evalúan las consecuencias de las transgresiones dietéticas, la determinación de GIP permite una evaluación directa y cuantitativa de la exposición al gluten. Por ello, el desarrollo y validación clínica de nuevas herramientas para la detección de GIP en muestras humanas tienen una importancia transcendental en el seguimiento de la EC (Comino et al., 2012, 2016, 2019; Moreno et al., 2017, 2021; Ruiz-Carnicer et al., 2021; Garzón et al., 2023; Elli et al., 2024). Con base a estos antecedentes, el OBJETIVO GENERAL de esta Tesis Doctoral ha sido el desarrollo y optimización de nuevos sistemas de análisis de gluten más eficientes, con aplicaciones en seguridad alimentaria, y en el seguimiento de la DSG de los pacientes con EC. ANTECEDENTES DEL TEMA Y OBJETIVOS 48 Para llevar a cabo este Objetivo General se han planteado los siguientes objetivos específicos: 1. Realizar un estudio clínico prospectivo longitudinal en pacientes pediátricos con EC para establecer la utilidad clínica de la determinación de GIP en heces como nuevo biomarcador para evaluar la adherencia a la DSG, comparándolo con los métodos tradicionales. 2. Desarrollar un método analítico cuantitativo, automatizable y de alta sensibilidad para la detección de GIP en orina, evaluando su eficacia como biomarcador para el seguimiento de la DSG en pacientes celíacos. 3. Evaluar la eficacia y versatilidad de UGES (Universal Gluten Extraction Solution), una nueva solución de extracción de gluten, bajo diversas condiciones metodológicas y en diferentes matrices alimentarias con grados variables de complejidad. Estudiar la compatibilidad de esta solución con distintas técnicas inmunológicas y compararla con las soluciones tradicionales de extracción de gluten. 4. Optimizar un nuevo sistema de análisis de gluten dirigido a la evaluación de matrices complejas y con proteínas hidrolizadas, como son las cervezas. BIBLIOGRAFÍA Aljada, B., Zohni, A., & El-Matary, W. (2021). The gluten-free diet for celiac disease and beyond. Nutrients, 13(11), 3993. Caio, G., Volta, U., Sapone, A., Leffler, D. A., De Giorgio, R., Catassi, C., & Fasano, A. (2019). Celiac disease: a comprehensive current review. BMC medicine, 17(1), 142. Comino, I., Fernández-Bañares, F., Esteve, M., Ortigosa, L., Castillejo, G., Fambuena, B., Ribes- Koninckx, C., Sierra, C., Rodríguez-Herrera, A., Salazar, J. C., Caunedo, Á., Marugán- Miguelsanz, J. M., Garrote, J. A., Vivas, S., Lo Iacono, O., Nuñez, A., Vaquero, L., Vegas, A. M., Crespo, L., Fernández-Salazar, L., … Sousa, C. (2016). Fecal gluten peptides reveal limitations of serological tests and food questionnaires for monitoring gluten-free diet in celiac disease patients. The American journal of gastroenterology, 111(10), 1456–1465. Comino, I., Real, A., Moreno, M.deL., Montes, R., Cebolla, A., & Sousa, C. (2013). Immunological determination of gliadin 33-mer equivalent peptides in beers as a specific and practical analytical method to assess safety for celiac patients. Journal of the science of food and agriculture, 93(4), 933–943. ANTECEDENTES DEL TEMA Y OBJETIVOS 49 Comino, I., Real, A., Vivas, S., Síglez, M. Á., Caminero, A., Nistal, E., Casqueiro, J., Rodríguez- Herrera, A., Cebolla, A., & Sousa, C. (2012). Monitoring of gluten-free diet compliance in celiac patients by assessment of gliadin 33-mer equivalent epitopes in feces. The American journal of clinical nutrition, 95(3), 670–677. Comino, I., Segura, V., Ortigosa, L., Espín, B., Castillejo, G., Garrote, J. A., Sierra, C., Millán, A., Ribes-Koninckx, C., Román, E., Rodríguez-Herrera, A., Díaz, J., Silvester, J. A., Cebolla, Á., & Sousa, C. (2019). Prospective longitudinal study: use of faecal gluten immunogenic peptides to monitor children diagnosed with coeliac disease during transition to a gluten-free diet. Alimentary Pharmacology & Therapeutics, 49(12), 1484–1492. Coto, L., Sousa, C., & Cebolla, A. (2021). Dynamics and considerations in the determination of the excretion of gluten immunogenic peptides in urine: individual variability at low gluten intake. Nutrients, 13(8), 2624. Dana, Z. Y., Lena, B., Vered, R., Haim, S., & Efrat, B. (2020). Factors associated with non adherence to a gluten free diet in adult with celiac disease: A survey assessed by BIAGI score. Clinics and research in hepatology and gastroenterology, 44(5), 762–767. Elli, L., Leffler, D., Cellier, C., Lebwohl, B., Ciacci, C., Schumann, M., Lundin, K. E. A., Chetcuti Zammit, S., Sidhu, R., Roncoroni, L., Bai, J. C., Lee, A. R., Dennis, M., Robert, M. E., Rostami, K., Khater, S., Comino, I., Cebolla, A., Branchi, F., Verdu, E. F., … Sanders, D. S. (2024). Guidelines for best practices in monitoring established coeliac disease in adult patients. Nature reviews. Gastroenterology & hepatology, 21(3), 198–215. Fueyo-Diaz, R., Gascón-Santos, S., Asensio-Martínez, Á., Sánchez-Calavera, M. A., & Magallón- Botaya, R. (2016). Transcultural adaptation and validation of the Celiac Dietary Adherence Test. A simple questionnaire to measure adherence to a gluten-free diet. Revista española de enfermedades digestivas, 108(3), 138–144. Garzón-Benavides, M., Ruiz-Carnicer, Á., Segura, V., Fombuena, B., García-Fernandez, F., Sobrino-Rodriguez, S., Gómez-Izquierdo, L., Montes-Cano, M. A., Millan-Domínguez, R., Del Carmen Rico, M., González-Naranjo, C., Bozada-García, J. M., Coronel-Rodríguez, C., Espin, B., Díaz, J., Comino, I., Argüelles-Arias, F., Cebolla, Á., Romero-Gómez, M., Rodriguez-Herrera, A., … Pizarro-Moreno, Á. (2023). Clinical utility of urinary gluten immunogenic peptides in the follow-up of patients with coeliac disease. Alimentary pharmacology & therapeutics, 57(9), 993– 1003. Gerasimidis, K., Zafeiropoulou, K., Mackinder, M., Ijaz, U. Z., Duncan, H., Buchanan, E., Cardigan, T., Edwards, C. A., McGrogan, P., & Russell, R. K. (2018). Comparison of clinical methods with the faecal gluten immunogenic peptide to assess gluten intake in coeliac disease. Journal of Pediatric Gastroenterology & Nutrition. 67 (3), 356–360. ANTECEDENTES DEL TEMA Y OBJETIVOS 50 Husby, S., & Bai, J. C. (2019). Follow-up of celiac disease. Gastroenterol Clin North Am, 48(1), 127-136. Kivelä, L., Caminero, A., Leffler, D. A., Pinto-Sanchez, M. I., Tye-Din, J. A., & Lindfors, K. (2021). Current and emerging therapies for coeliac disease. Nature reviews. Gastroenterology & hepatology, 18(3), 181–195. Lenti, M. V., Biagi, F., Lucioni, M., Di Sabatino, A., Paulli, M., & Corazza, G. R. (2019). Two cases of monomorphic epitheliotropic intestinal T-cell lymphoma associated with coeliac disease. Scandinavian journal of gastroenterology, 54(8), 965–968. Lindfors, K., Ciacci, C., Kurppa, K., Lundin, K. E. A., Makharia, G. K., Mearin, M. L., Murray, J. A., Verdu, E. F., & Kaukinen, K. (2019). Coeliac disease. Nature reviews. Disease primers, 5(1), 3. Mahadev, S., Murray, J. A., Green, P. H. R., Adelman, D. A., & Lebwohl, B. (2017). Editorial: risk factors for persistent villus atrophy in coeliac disease - is it time to reconsider definitions for refractory coeliac disease? Authors' reply. Alimentary pharmacology & therapeutics, 45(11), 1478–1479. Moreno, M. L., Cebolla, Á., Muñoz-Suano, A., Carrillo-Carrion, C., Comino, I., Pizarro, Á., León, F., Rodríguez-Herrera, A., & Sousa, C. (2017). Detection of gluten immunogenic peptides in the urine of patients with coeliac disease reveals transgressions in the gluten-free diet and incomplete mucosal healing. Gut, 66(2), 250–257. Moreno, M. L., Sánchez-Muñoz, D., Sanders, D., Rodríguez-Herrera, A., & Sousa, C. (2021). Verifying diagnosis of refractory celiac disease with urine gluten immunogenic peptides as biomarker. Frontiers in medicine, 7, 601854 Muhammad, H., Reeves, S., Ishaq, S., Mayberry, J., & Jeanes, Y. M. (2017). Adherence to a gluten free diet is associated with receiving gluten free foods on prescription and understanding food labelling. Nutrients, 9(7), 705. Raju, S. A., Shiha, M. G., & Penny, H. A. (2024). Monitoring coeliac disease in 2024, time to change practice?. Current opinion in gastroenterology, 40(3), 190–195. Real, A., Comino, I., de Lorenzo, L., Merchán, F., Gil-Humanes, J., Giménez, M. J., López-Casado, M. Á., Torres, M. I., Cebolla, Á., Sousa, C., Barro, F., & Pistón, F. (2012). Molecular and immunological characterization of gluten proteins isolated from oat cultivars that differ in toxicity for celiac disease. PloS one, 7(12), e48365. Real, A., Comino, I., Moreno, M.D.L., López-Casado, M.Á., Lorite, P., Isabel Torres, M., Cebolla, Á., Sousa, C. (2014). Identification and in vitro reactivity of celiac immunoactive peptides in an apparent gluten-free beer. PLoS one, 9, 12–14. Rostami, K., Bold, J., Parr, A., Johnson, M.W. (2017). Gluten-Free diet indications, safety, quality, labels, and challenges. Nutrients, 9, 846. ANTECEDENTES DEL TEMA Y OBJETIVOS 51 Ruiz-Carnicer, Á., Garzón-Benavides, M., Fombuena, B., Segura, V., García-Fernández, F., Sobrino-Rodríguez, S., Gómez-Izquierdo, L., Montes-Cano, M. A., Rodríguez-Herrera, A., Millán, R., Rico, M. C., González-Naranjo, C., Bozada-García, J. M., Díaz, J., Coronel-Rodríguez, C., Espín, B., Romero-Gómez, M., Cebolla, Á., Sousa, C., Comino, I., … Pizarro, Á. (2020). Negative predictive value of the repeated absence of gluten immunogenic peptides in the urine of treated celiac patients in predicting mucosal healing: new proposals for follow-up in celiac disease. The American journal of clinical nutrition, 112(5), 1240–1251. Russell, A. K., Lucas, E. C., Henneken, L. M., Pizzey, C. J., Clarke, D., Myleus, A., & Tye-Din, J. A. (2024). Stool gluten peptide detection is superior to urinary analysis, coeliac serology, dietary adherence scores and symptoms in the detection of intermittent gluten exposure in coeliac disease: a randomised, placebo-controlled, low-dose gluten challenge study. Nutrients, 16(2), 279. Segura, V., Ruiz-Carnicer, Á., Sousa, C., Moreno, M. de L. (2021). New insights into non-dietary treatment in celiac disease: Emerging therapeutic options. Nutrients, 13 (7), 2146. Silvester, J. A., Comino, I., Kelly, C. P., Sousa, C., Duerksen, D. R., & DOGGIE BAG Study Group. (2020). Most patients with celiac disease on gluten-free diets consume measurable amounts of gluten. Gastroenterology, 158(5), 1497–1499.e1. Silvester, J. A., Comino, I., Rigaux, L. N., Segura, V., Green, K. H., Cebolla, A., Weiten, D., Dominguez, R., Leffler, D. A., Leon, F., Bernstein, C. N., Graff, L. A., Kelly, C. P., Sousa, C., & Duerksen, D. R. (2020). Exposure sources, amounts and time course of gluten ingestion and excretion in patients with coeliac disease on a gluten-free diet. Alimentary pharmacology & therapeutics, 52(9), 1469–1479. Skodje, G. I., van Megen, F., Stendahl, M., Henriksen, C., Lundin, K. E. A., & Veierød, M. B. (2022). Detection of gluten immunogenic peptides and the Celiac Disease Adherence Test to monitor gluten-free diet: a pilot study. European journal of clinical nutrition, 76(6), 902–903. Silvester, J. A., Weiten, D., Graff, L. A., Walker, J. R., & Duerksen, D. R. (2016). Living glutenfree: adherence, knowledge, lifestyle adaptations and feelings towards a gluten-free diet. Journal of human nutrition and dietetics: the official journal of the British Dietetic Association, 29(3), 374–382. Therrien, A., Kelly, C. P., Silvester, J. A. (2020). Celiac disease. Journal of Clinical Gastroenterology, 54(1), 8–21. Vargas, F. M., Cardoso, L. T., Didoné, A., Lima, J. P. M., Venzke, J. G., & de Oliveira, V. R. (2024). Celiac disease: risks of cross-contamination and strategies for gluten removal in food environments. International journal of environmental research and public health, 21(2), 124. ANTECEDENTES DEL TEMA Y OBJETIVOS 52 Xhaferaj, M., Alves, T. O., Ferreira, M. S., Scherf, K. A. (2020). Recent progress in analytical method development to ensure the safety of gluten-free foods for celiac disease patients. Journal of Cereal Science, 96. Yang, D., & Gao, X. (2022). Progress of the use of alternatives to malt in the production of glutenfree beer. Critical reviews in food science and nutrition, 62(10), 2820–2835 2 | COMINO et al. 1 | INTRODUCTION Coeliacdiseaseisanimmune‐mediateddisorderofthesmallintestinethataffects~1%ofmostpopulations.1,2Thislifelongcondition isinitiatedbyexposuretodietarygluteningeneticallysusceptible individualsandcanaffectanyorganortissue.Consequently,avarietyofgastrointestinalandextra‐intestinalsymptomsareobserved, suchasabdominalpain,malabsorption,anaemia,failuretothrive, osteoporosisand,occasionally,lymphoma.3 A lifelong gluten‐free diet (GFD) is currently the only availabletreatmentforcoeliacdiseaseandhasbeenshowntoimprove quality of life substantially, both in patients initially presenting with gastrointestinal symptoms and in asymptomatic patients.4 However, adherence to the GFD is not easy because gluten is ubiquitousandoftennotexplicitlylistedasaningredientinfood products.Asaresult,persistentsymptomsandenteropathyare commonamongcoeliacpatientswhoaretryingtofollowaGFD. Specifically,researchshowsthat25%‐40%ofadultswithcoeliac disease have persistent enteropathy after 2 years on a GFD.5‐7 Childrenarethoughttorecovermorequickly,anddatasuggests that 5%‐19% of coeliac children on a GFD may have persistent enteropathydespitetreatmentwithaGFDforatleast1year.8‐13 Moreover,whiletheexpertsagreestrictadherencetotheGFDis crucialforthehealthofcoeliacpatients,therearenoevidencebasedrecommendationsregardingthemostefficientwaytoassessGFDadherence.14Typically,dietaryhistory,symptomsand serumantibodytestsareusedtoassessnon‐adherence.However, adetaileddietaryhistoryistimeconsumingandrequiresthecollaborationofadietician,soinroutineclinicalpractice,physicians oftenrelyuponthepatient'sownsubjectiveassessmentofdietary adherence,orthepresenceofadversesymptomstoguidedieticianreferral.Thedifficultywiththisapproachisthatmanynonadherentpatientsmaybeasymptomatic.Inaddition,ifpatientsdo havesymptoms,thesemaybeasaresultofcoexistingdisorders, suchasirritablebowelsyndrome,microscopiccolitisorpancreaticinsufficiency.15Concerningly,althoughcoeliacantibodytests arewidelyusedforroutinemonitoringofcoeliacpatients,these testswereneverapprovedforthispurposeandhavelowsensitivityindetectionofpersistentmucosallesionincoeliacpatients onaGFD.16Therefore,noneofthesemethodsofferanaccurate measureofdietaryadherence.Whilstduodenalbiopsyisconsideredthe‘goldstandard’forassessmentofceliacdiseaseactivity,17 certainassociations,suchastheEuropeanSocietyforPediatric Gastroenterology,HepatologyandNutrition(ESPGHAN)Coeliac Diseaseworkinggroupstronglyadvisesagainstregularre‐biopsy inchildrenonaGFD,duetounnecessaryrisksforthepatient,negativeimpactsonqualityoflife,andpointlesshighmedicalcosts.18 Monoclonal antibodies capable of sensitively and specifically detecting gluten immunogenic peptides (GIP) have been used to develop assays for detecting inadvertent gluten consumption by measuringGIPinhumansamples(faecesandurine).Theseassays maybeapracticalmethodtoassessdietaryadherenceincoeliac patients.19‐21GIPareresistanttogastrointestinaldigestionandaccountforimmunogenicreactionsinTcellsofpatientswithcoeliac disease.22 Unlike traditional methods to monitor GFD adherence whichonlyevaluatetheconsequencesofGFDtransgressions,this non‐invasivemethodenablesadirectandquantitativeassessment ofglutenexposure.ArecentstudydemonstratedtheclinicalusefulnessofGIPasamarkerofGFDadherenceinpatientsfollowinga GFDforatleast1year.20Althoughitiswellknownthatcompliance with theGFDoverall decreasesonlong term follow‐up, however strictadherencetoaGFDinthefirstyearafterthediagnosiscrucial tofavourmucosalrecovery.Forthisreason,theaimofthepresent studywastoevaluatethemeasurementofGIPinstoolsasamarker ofGFDadherenceinnewlydiagnosedcoeliacchildreninaprospectivemulticentreclinicaltrial.Furthermore,thiswascomparedtotraditionalmethodsusedtoassessGFDadherence:dietitianreviewof four‐dayfoodrecrodandcoeliacserology(tissuetransglutaminase, tTGanddeamidatedgliadinpeptideantibodies,DGP). 2 | MATERIALS AND METHODS 2.1 | Study design AmulticentreprospectiveobservationalstudyofdetectionofGIP instoolinacohortofchildrennewlydiagnosedwithcoeliacdiseasewasperformedatsevenSpanishsecondaryandtertiaryreferralhospitals:HospitalNuestraSeñoradelaCandelaria(Tenerife), Hospital Regional Carlos Haya (Málaga), Hospital Universitario y PolitécnicoLaFe(Valencia),HospitalUniversitarioVirgendelRocío (Sevilla),HospitalUniversitarioVirgendeValme(Sevilla),Instituto HispalensedePediatría(Sevilla),HospitalUniversitarioRíoHortega (Valladolid)andHospitalSantJoandeReus(Tarragona).Therewere four study visits at diagnosis and, 6 (from 4‐9 months), 12 (from 10‐15months)and24(from16‐24months)monthsafterdiagnosis. Thefirstvisit(diagnosis)wasatthetimeofdiagnosticendoscopy when patients were untreated (diet with gluten). All participants wereinstructedtofollowaGFDbyclinicaldieticianswithexpertise incoeliacdisease.Samplesoffaecesandbloodwerecollectedat eachstudyvisitandparticipantscompletedafour‐dayfoodrecord accordingtothedieticianinstructions.Thestudyprotocolwasreviewedbytheethicscommitteeateachparticipatinghospitaland writteninformedconsentwasobtainedfromtheparentsorlegal guardians. 2.2 | Study population Children (less than 18 years of age) with active coeliac disease wererecruitedattimeofdiagnosticendoscopy.Inclusioncriteria were:(a)symptoms—eithergastrointestinal(eg,abdominalpain, diarrhea,constipation,weightloss,flatulence,bloating,vomiting, lackofappetite)oratypical(e.g.,irondeficiencyanaemia,chronic fatigue, behavioural changes, poor growth); (b) elevated serum EMAIgA,tTGIgA/IgGand/orDGPIgA/IgG;(c)HLADQ2and/or | 3 COMINO et al. HLADQ8genotype;and,(d)smallintestinalhistologyconsistent withcoeliacdisease(MarshII‐III).Exclusioncriteriawere:(a)historyofkidney,liverorseverepsychiatricdisease;(b)seizuredisorderand/orcurrentuseofanticonvulsants;(c)useofanyantibiotics withintheyearpriortoenrolment. Allsubjectdatawererecordedinanelectronicdatacapture (EDC)system,including:age,weight,height,intestinalhistology, symptoms, comorbiditions, clinical test results, adverse events, family history of coeliac disease, date of coeliac disease diagnosis,adherencetoGFD,anddetailssamplecollectionandvisit attendance. 2.3 | Faeces and blood collection Subjects were instructed to collect 2‐4 g stool in a sealed containerafterrecordingtheirfoodintakefor4days.Specimenswere dropped‐offwithin24hoursofcollectionandwerekeptat−20°Cat alltimesuntilprocessing. Bloodsampleswerecollectedintwo3mLvacutainertubeswith EDTA‐K3anticoagulantandcentrifugedat2000gwithin30minutesofcollectiontoobtainplasmawhichwasstoredat−80°Cuntil analysis.Investigatorsperformingstoolandserumanalysiswere blindedtoGFDstatusatthetimeofsamplecollection. 2.4 | Quantification of GIP in stool samples Stool GIP concentration was determined by sandwich enzymelinkedimmunosorbentassay(ELISA;iVYDALInVitroDiagnostics iVYLISAGIPStoolkit,BiomedalS.L.,Seville,Spain)accordingto the manufacturer's protocol. Briefly, stool samples were mixed with9mlUniversalGlutenExtractionSolution(UGES;Biomedal S.L.,Seville,Spain)pergramofstoolthenincubatedat50°Cfor 60minuteswithgentleagitationtoreleasetheGIPfromthestool matrix.Afterextraction,sampleswerediluted1:10withdilution solutionandELISAwasperformedusingtheprovidedG12coated microtiterplate,standards(50,25,6.25,3.13,1.56ng/mlGIP)and positive and negativecontrols.Thus,resultswere expressed as μgGIPpergramfaeces.Eachsamplewasruninduplicateandat leasttwodifferentaliquotsofeachsampleweretestedondifferentdays. The validity of this method in detecting GFD transgressions wasdeterminedbytheanalyticalsensitivity(limitsofdetectionand quantification0.06and0.16μgGIPpergramfaeces,respectively) and the diagnostic sensitivity and specificity (98.5% and 100%, respectively).20 2.5 | Estimation of gluten consumption A calibration factor allowed estimation of the ingestion of glutenincoeliacpatientsfromstoolmeasurements.Specifically,the equationforestimatingdailyglutenconsumptioninmilligrams(y variable)baseduponfaecalGIPconcentration(inmicrogramsper gram) (x variable) was determined from measured mean values of6.2and14.9μgGIPpergramfaecesduringcontrolledgluten challenges of 9 and 30 grams per day.19,20,23Fittingtoasecond‐orderpolynomialgoingthroughtheorigingavetherelation y=0.0649x2+1.0461x. 2.6 | Serology ThelevelsoftTGIgAandDGPIgAantibodies(IgGinIgA‐deficient patients)weredeterminedbyELISAusingtheEliATMCelikey® and EliATM Gliadin kits, respectively, according to the manufacturer's protocol (Phadia, Freiburg, Germany). Measurements were performedinduplicateandtheresultsexpressedasU/ml.Themanufacturerrecommendedcut‐offof>10U/mLwasused. 2.7 | Dietary questionnaire To assessglutenexposure,a structuredfoodquestionnaire of27 itemswasadministeredtorecordthefoodsconsumedduringthe 4dayspriortostoolandbloodcollection.Thefooditemswereclassifiedintoeightpredefinedgroups: dairy(milkandcheese); complexcarbohydrates(bread,cereals,pasta,rice,potato,legumes,and nuts);meats (redmeat,fish,coldcuts,andeggs);fruits(wholeor juiced);vegetables;fats(vegetableoils,butter,andcream);sweetenedbeverages(sodas,bottledjuices,andenergydrinks);andother (bakedgoods,candy,snacks,etc.).Imagesofstandardportionsizes were included as a guideline for portion quantification. Subjects wereaskedtorecordtheamountandtypeoffoodconsumed,brand, timeofmeal,andifitwaslabelledasgluten‐free.Theywerealso askedtonoteiftheywereawareofhavingconsumedanyglutencontainingfoods. 2.8 | Statistical analysis Quantitiative variables are expressed as median with interquartile range (IQR), and the categorical variables as percentages. Goodness‐of‐fittonormalwascheckedusingtheShapiro‐Wilktest. Pearson'schi‐squaredtestwasusedforcategoricalvariables,and thechi‐squaredtestforordinalvariables.Statisticalanalysisofthe degreeofconcordanceofthedichotomouslyevaluateddiagnostic techniqueswasperformedusingCohen'skappaindex(κ),following thecriteriaofLandisandKoch24forthequalitativeinterpretationof thestrengthofconcordance.TheMann‐WhitneyUtestwasusedto comparequantitativevariablesinindependentgroups.Forallcases, P<0.05wasconsideredstatisticallysignificant.Dataanalysiswas performedwithSPSS23.0forWindows(SPSSInc). 3 | RESULTS The study population consisted of 64 children (21 males and 43 females) withamedianageof 4years(IQR1.5‐9years;Table1). Participantretentionwas94%at6months,78%at12monthsand 55%at24months(themostcommonreasonforlosstofollow‐up 4 | COMINO et al. were:movingoutofthestudyarea,notattendingfollow‐upvisits andforgettingtocollectsamples). 3.1 | Detection of GIP in stool Attheinitialvisit,beforestartingtheGFD,duringtheexpectedglutenchallengefordiagnosis,62(97%)patientshaddetectableGIP intheprovidedstoolsample.Afterdiagnosisandtreatmentwitha GFD,13(23%)ofpatientshaddetectablelevelsofGIPinatleastone visit,whereas11(20%)werenoncompliantaccordingtothequestionnaire.Overall,11%ofallcoeliacpatientswerenoncompliantby bothmethods,48%wereglutenfreebybothmethods(κ=0.5)and 39%werediscordant.Ingeneral,GFDadherenceratesdeclinedas thestudyprogressed(Figure1A,B).TherateofGIPpositivestools was13%at6months,18%at12monthsand25%at24months. Notably,46%oftransgressorshaddetectableGIPattwoormore follow‐upvisits(Figure1C). TherateofGIPpositivestoolsduringfollow‐upincreasedwith age(P = 0.041).Specifically,GIPweredetectedinstoolsfrom6%of childrenwithcoeliacdiseasebefore2yearsofage,andin24%and 35%inthosediagnosedat2‐6yearsandatanolderage,7‐18years, respectively(Figure2). Elevenparticipants(17%)reportednoclassicsymptomsprior todiagnosiswithcoeliacdisease and insteadwere biopsiedfor atypical symptoms or laboratory abnormalities. The rate of dietarytransgressionsinthisgroupwas33%comparedto19%for the group diagnosed due to classical coeliac disease symptoms (P = 0.57). 3.2 | Estimated gluten ingestion Anestimateofglutenconsumptionbycoeliacpatientsfollowinga GFDwasdeterminedbymeasuringGIPinstool.Boththemeanand medianvaluesforeachvisitarereportedduetothenon‐normaldistributionofgluteningestion(Table2).Atdiagnosis,estimateddailyglutenconsumptionwas5543mg(mean)witha95%CI[4345‐6741mg] and3882mg(median).DuringtransitiontoaGFD,wefound342mg gluten(mean),95%CI[67‐616mg]and104mg(median)insamples tested.Therewasatrendtowardincreasedglutenconsumptionduringthefollow‐upperiod.Afterdiagnosis,estimatedglutenexposure was (mean/median) 144/99 mg/d at 6 months, 452/105 mg/d at 12months,and606/117mg/dat24months. 3.3 | Correlation between faecal GIP and serum antibodies Serum tTG and DGP antibodies levels declined during follow‐up (Figure3A).FortTGantibody,48%werepositiveatthe6months visit,34%at12monthsand20%at24months.ForDGPantibody, positiveresultswereobtainedin11%at6months,5%at12months and0%at24months.Incontrast,ratesofGIPpositivestoolsshowed anupwardtrendduringthefollow‐upperiodandwerehighestatthe lastvisit(Figure3B). ConsideringtheabsolutevaluesforbothtTGandDGPantibodiesateachvisit,wedidnotfindconcordanceforGIP,withverylow kappavalues,farfromstatisticalsignificance(P > 0.1).Norwasthere statisticalconcordancewhenusingdichotomoustermsthatis,positivewhenGIPweredetectedatleastonefollow‐upvisit(6,12,or 24months),andnegativewhenGIPwerenotdetectedduringfollow‐up(Table3).Thekappavalueswereverylow,farfromstatistical significance(P > 0.1)inallcases. Figure3displaystheabsolutereductioninthetTGantibody titerafter6months(Figure3A)and12months(Figure3B)ofGFD therapy.Wealsofoundasubstantial,significant,reductioninthe tTGantibodywithrespecttothebasal(diagnosis)levels;thistrend was stronger in the group of patients adhering to the GFD, as demonstratedbythenegativeGIP,withreductionsof103U/mL (IQR50‐122)vs28U/mL(IQR12‐63)inthepositiveGIPpatients after6monthsofGFD(P = 0.028)(Figure4A).Thisdifferential kineticbehaviourofthetTGantibodyinthetwogroupswasalso observedat12monthsoftreatment:116U/mL(IQR61‐127)in adherentsvs48U/mL(IQR31‐101)(P = 0.038)innon‐adherents (Figure4B). 4 | DISCUSSION Inthisstudyofchildrenwithcoeliacdisease,therateofGIPdetectioninstooldroppeddramaticallyfrom97%atdiagnosisto13%after 6monthsofGFD.ThishighlevelofadherenceafteraGFDwasnot maintainedonfollow‐up.SomechildrenmayrelaxtheGFDasevidencedbyanincreasingrateofGIPdetectionwithtime.Wefound that46%ofnon‐adherentparticipantshadtwoormoreGIPpositivestoolsonaGFD.Thissuggestsabehaviouralpatterninvolving repeatedorchronicglutenexposureratherthaninfrequentepisodic exposures.Asexpected,thestoolGIPconcentrationinmanynon‐ad‐ herentpatientswaslow,closetothelimitofquantification(0.16μg GIPpergramfaeces).Nevertheless,thislevelofexposureislikely significantaschildrenwhohaddetectableGIPintheirstoolatany timeafterdiagnosishadamoreprolongedelevationoftheirserum tTGantibodythanthosewithallstoolsbeingnegativewhiletrying TABLE 1 Characteristicsof64patientswithceliacdisease enrolledinthestudy Characteristic Patients, n (%) Sex Male 21(33%) Female 43(67%) Age(y) <2 16(25%) 2‐6 21(33%) 7‐18 27(42%) Medianage 4 Interquartilerange(P25‐P75)1.5‐9 | 5 COMINO et al. tofollowaGFD.Furthermore,somepatientshadfaecalGIPlevels similartothoseofhealthycontrolsonaglutencontainingdiet.19,20 Investigation of factors underlying gluten exposure was beyond the scope of the current study; however, it must not be assumedthatgluteningestionwasintentional.AlthoughallfamiliesreceivedGFDeducationfromadietician,itcannotbeexpected thattheyareimplementingaGFDcorrectly.Similartoourpreviousstudies,20childrenyoungerthantwoyearsoldhadthehighest andmostsustainedadherencetoGFDusingobjectivemeasures. Thisgroup,besidehavingoverallamorecontrolleddiet,ishighly dependentontheirparentorguardiantobefed;therefore,strict controloverthedietmaybeeasiertoachievecomparedwitholder children(7‐18yearsold).Thusthesedifferencesmayreflectthe increasingautonomy ofthepatientregarding dietary decisions. Adolescence is a developmental phase characterised by rebellion and others have found that teenagers may be particularly susceptibletotheburdenofaGFDrelatedtostigmatizationand aremorelikelyintentionallynon‐adherentin suchsettings.25‐27 Alternatively,olderchildrenandadolescentsarealsomorelikely toeatoutsidethehome(eg,atschool,atafriends’home)thus havingfoodpreparedbypersonsnotalwayswell‐informedabout theGFDrequirements.TheincreasedrateofGIPdetectionwith agemaysimplyreflectthatthesebehavioursareinherentlymore “risky”withregardstoglutenexposure.Amongadults,symptomatic suspected gluten reactions are most commonly associated witheatingatrestaurants.28Alternatively,itmaybethatthehabit ofGFDadherence,whenestablishedinearlychildhood,ismaintained through adolescence and adulthood. This would suggest thatdiagnosingpatientswithcoeliacdiseaseasearlyinlifeaspossibleshouldbeaprioritytoavoidsocialproblemsandphysicaland psychologicaldeterioration.29 FIGURE 1 Concentrationofglutenimmunogenicpeptides(GIP)instoolsofpatientswithnewlydiagnosedcoeliacdiseaseduring monitoringofthegluten‐freediet.(A)LevelsoffaecalGIPatthebasalandfollow‐upvisits(basal,6,12and24mo).(B)LevelsoffaecalGIP at6,12and24mo.(C)LevelsofGIPintransgressingpatientsinthedifferentfollow‐upvisits(logscale).GIP,glutenimmunogenicpeptides; LOQ,limitofquantification µg GIP per g faeces µg GIP per g faeces (B) (A) 5 µg GIP per g faeces 1.5 1.0 LOQ = 0.16 µg GIP per g faeces 0.5 LOQ = 0.16 µg GIP per g faeces 6 months 12 months 24 months (C) 5 µg GIP per g faeces 6 months 12 months 24 months 6 months Basal 12 months 24 months 4 2 1 0.1 µg GIP per g faeces FIGURE 2 GlutenimmunogenicpeptidesdetectiononaGFD accordingtopatientage.Percentagedistributionofstoolscollected duringaGFDwithdetectableGIPaccordingtoage.GIP,gluten immunogenicpeptides.GFD,gluten‐freediet GIP positive 7-18 years2-6 years < 2 years 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 6 94 24 76 35 65 Patients with detectable GIP on GFD GIP negative 6 | COMINO et al. Theinitialsymptomatic responsetoaGFD tends toberelativelyrapid,occurringwithindaystoweeks,especiallyinpatients presenting with classical symptoms. Conversely, gluten exposureonaGFDmaynotevokesymptoms.28,30Wefoundtherate ofGIPpositivestoolsincreasedwithtimesincediagnosis,which mayreflectatendencytorelaxGFDadherenceovertimeaspatientsidentifythattheymaytoleratesomeglutenwithoutsymptomsandadjusttheirdietaccordingly.Theestimatedglutenintake ofpatientsonaGFD(median0.104gramsperday)wasbemore than30‐foldlessthanthetypicalamountusedinglutenchallenge (3‐10g/d).30Perhaps,ingestionof0.1‐0.5gglutendoesnotgeneratesymptomsinasignificantproportionofnon‐adherentpatients. Asafethresholdforglutenconsumptionbypersonswithcoeliac diseasehasnotbeenestablished;however,itisacceptedthatglutentolerancevarieswidely.31,32Consistentwiththisnotion,those whoweresymptomaticatdiagnosiswerelesslikelythanthosewho wereasymptomatictohavedetectableGIPinthestool.Otherinvestigatorshavealsoidentifiedadesiretoavoidsymptomsasa motivationforadheringstrictlytoaGFD.33‐35 Ourfindingthatmostchildrenwereabletosubstantiallyreduce theirgluteningestionwithin6monthshasimplicationsforthedefinitionof“non‐responsivecoeliacdisease”.Acategorizationrather thanadiagnosis,non‐responsivecoeliacdiseasehasbeendefined as “persistent symptoms, signs, or laboratory abnormalities typicalofcoeliacdiseasedespite6‐12monthsofdietaryglutenavoidance”.36Thedurationofglutenavoidancerequiredtobeconsidered TABLE 2 EstimatedglutenconsumptionbaseduponGIPmeasuredinstoola GFD duration (mo) n GIP concentration (μg/g stool) Estimated gluten consumption (mg/d) Mean Median IQR Mean Median IQR 0(diagnosis) 64 3.82 3.11 0.60‐7.60 5543 3882 691‐11699 654 0.14 0.09 0.05‐0.14 144 99 49‐149 12 39 0.34 0.10 0.05‐0.15 452 105 52‐160 24 24 0.43 0.11 0.06‐0.18 606 117 58‐190 GFD,gluten‐freediet;GIP,glutenimmunogenicpeptides. aConversionfactorx=GIP(μg/gstool)toy=glutendailyconsumption(mg)isy=0.0649x2+1.0461x. FIGURE 3 EvolutionofGIP,tTGandDGPantibodiesinpatientswithnewlydiagnosedcoeliacdisease.(A)tTGandDGPantibodylevels vstime.(B)PercentageofdietarytransgressionsaccordingGIP,tTGandDGPantibodiesduringthestudyperiod.DGP,deamidatedgliadin peptideantibody;tTG,tissuetransglutaminaseantibody;GFD,gluten‐freediet;GIP,glutenimmunogenicpeptides.Thecut‐off>10U/mL Antibody title (U/ml) Percentage of positive (A) 100 (B) 100 90 80 70 DGP IgA tTG IgA 60 50 40 10 30 20 10 Diagnosis 6 months 12 months 24 months GIP tTG IgA DGP IgA 1 Basal 6 months 12 months 24 months GIP 97% (62/64) 13% (7/54) 18% (7/39) 25% (6/24) tTG IgA 97% (59/61) 48% (29/60) 34% (17/50) 20% (7/35) DGP IgA 75% (45/60) 11% (7/63) 5% (2/44) 0% (0/27) TABLE 3 EvolutionoftheGFDaccordingtothecomparativeGIPandcoeliacdiseaseserologies Time of GFD Comparison GIP vs λc95% CI λcCriterios azzimonti Kappa (P) Evolutionary tTGIgA 57.4 42.3‐68.4 No 0.107(0.342) DGPIgA 65.6 52.9‐78.0 No 0.103(0.423) CI,confidenceinterval;DGP,deamidatedgliadinpeptideantibody;GFD,gluten‐freediet;GIP,glutenimmunogenicpeptides;tTG,tissuetransglutaminaseantibody. | 7 COMINO et al. non‐responsivehasbeencontroversial,withsomegroupsrequiring atleast12monthsofglutenrestrictionbeforeconsideringpatients non‐reponsive37,38andotherssuggestingthatinadequatesymptomaticresponseafter6monthsonaGFDshouldpromptinvestigation for etiologies other than coeliac disease.39 Our data suggest that 6monthsmaybeareasonabledefinitionasmostchildrensubstantiallyreducedtheirglutenintakewithinthistimeframe.Regardless ofthedefinition,gluteningestion(eitherintentionalorinadvertent) isconsistentlyfoundtobetheleadingcauseofnon‐responsivecoeliacdisease.38,39TestingforGIPinstool maybe ausefultool in theevaluationofnon‐responsivecoeliacdisease.Identificationof ongoingglutenexposuremayguidetreatmentandobviatetheneed forexpensiveandinvasiveinvestigations. Our findings also help to clarify the interpretation of coeliac serology tests after a diagnosis of coeliac disease is established. Although widely used to monitor patients on a GFD and highly specificforpersistentvillousatrophyonaGFD,thelowsensitivity ofserumtTG,EMAandDGPantibodiestestsrendersanegative test substantially less informative.16,40Wefoundthatthosewith detectablelevelsofGIP in stoolhad amoreprolongedelevation anda moregradualfallintTGantibodythanthosewhosestools testednegative.However,inabsoluteandevolutive(dichotomous) terms,noconcordanceisobservedbetweenthetTGantibodyand GIPlevels. Strengthsofourstudyincludetheuseofarobustandobjectivemeasureofglutenexposureinapopulationofchildrenwhoall received formal GFD education and who were followed prospectively from diagnosis. Contemporaneous food record completion andstoolandserumcollectionallowedforaccuratecorrelationof serologywithfaecalGIPdetection.Losstofollow‐upisalimitation ofourstudythatdoescomplicateinterpretationofincreasedgluten exposure. In conclusion, serial analysis of faecal GIP allowed not only the firstdirectconfirmationofglutenintakedaysbeforecoeliacdisease diagnosis,butalsodocumentationofasubstantialdecreaseingluten consumptionafterdiagnosisofcoeliacdiseaseandinstructioninfollowingaGFD.TheintroductionofGIPasanassessmenttoolofGFD adherencemayhelptoascertaindietarycomplianceandreducethe needforadditionalinvasiveinvestigationsonfollow‐up. ACKNOWLEDGEMENTS Declaration of personal interests: A.C. own stock in Biomedal S.L. Othersauthorsdeclarenoconflictofinterest. AUTHORSHIP Guarantor of the article:CarolinaSousa. Author contributions:IC,VS,ACandCSperformedtheresearch; IC,VS,LO,BE,GC,JAG,CSi,AM,CRK,ER,ARH,JD,JAS,ACandCS collectedandanalysedthedata;IC,VS,JAS,AC,CSdesignedthe researchstudyandwrotethepaper;andIC,ACandCScontributed tothedesignofthestudy.Allauthorsapprovedthefinalversionof themanuscript. ORCID Isabel Comino https://orcid.org/0000‐0002‐4982‐273X FIGURE 4 AbsolutereductionoftTGantibodylevelat(A)6moand(B)12moinpatientswithdetectableGIPandnonGIPdetected.GIP, glutenimmunogenicpeptides;tTG,tissuetransglutaminase 160 P75 = 122 Mann-Whitney Test p = 0.0277 P75 = 63 P50 = 103 P50 = 28 P25 = 50 P25 = 12 140 120 100 80 Absolute reduction of tTG IgA from baseline Absolute reduction of tTG IgA from baseline 60 40 20 0 GIP negative GIP positive 160 P75 = 127 Mann-Whitney Test p = 0.038 P75 = 101 P50 = 116 P50 = 48 P25 = 61 P25 = 31 140 120 100 80 60 40 20 0 GIP negative GIP positive (A) (B) 8 | COMINO et al. REFERENCES  1. Rubio‐TapiaA,MurrayJA.Celiacdisease.Curr Opin Gastroenterol. 2010;26:116‐122.  2. SinghP,AroraA,StrandTA,etal.Globalprevalenceofceliacdisease: systematic review and meta‐analysis. Clin Gastroenterol Hepatol.2018;16:823‐836.e2.  3. HusbyS,KoletzkoS,Korponay‐SzabóIR,etal.;ESPGHANWorking GrouponCoeliacDiseaseDiagnosis;ESPGHANGastroenterology Committee; European Society for Pediatric Gastroenterology, Hepatology, and Nutrition. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr. 2012;54:136‐160.  4. PietzakMM.Follow‐upofpatientswithceliacdisease:achieving compliancewithtreatment.Gastroenterology.2005;128:S135‐S141.  5. CiacciC,CirilloM,CavallaroR,MazzaccaG.Long‐termfollow‐upof celiacadultsongluten‐freediet:prevalenceandcorrelatesofintestinaldamage.Digestion.2002;66:178‐185.  6. Rubio‐Tapia A, Rahim MW, See JA, Lahr BD, Wu TT, Murray JA. Mucosal recovery and mortality in adults with celiac disease after treatment with a gluten‐free diet. Am J Gastroenterol. 2010;105:1412‐1420.  7. MahadevS,MurrayJA,WuTT,etal.Factorsassociatedwithvillus atrophyinsymptomaticcoeliacdiseasepatientsonagluten‐free diet. Aliment Pharmacol Ther.2017;45:1084‐1093.  8. BannisterEG,CameronDJ,NgJ,etal.Canceliacserologyalone be used as a marker of duodenal mucosal recovery in children with celiac disease on a gluten‐free diet? Am J Gastroenterol. 2014;109:1478‐1483.  9. GhazzawiY,Rubio‐TapiaA,MurrayJA,AbsahI.Mucosalhealing inchildrenwithtreatedceliacdisease.J Pediatr Gastroenterol Nutr. 2014;59:229‐231. 10. VécseiE,SteinwendnerS,KoglerH,etal.Follow‐upofpediatric celiacdisease:valueofantibodiesinpredictingmucosalhealing,a 289prospectivecohortstudy.BMC Gastroenterol. 2014;14:28. 11. LeonardMM,CuretonP,FasanoA.Indicationsanduseofthegluten contaminationeliminationdietforpatientswithnon‐responsiveceliacdisease.Nutrients.2017;9:pii:E1129. 12. LeonardMM,SaponeA,CatassiC,FasanoA.Celiacdiseaseand nonceliacglutensensitivity:areview.JAMA.2017;318:647‐656. 13. LeonardMM,WeirDC,DeGrooteM,etal.ValueofIgAtTGinpredictingmucosalrecoveryinchildrenwithceliacdisease277ona gluten‐freediet.J Pediatr Gastroenterol Nutr.2017;64:286‐291. 14. LudvigssonJF,BaiJC,BiagiF,etal.BSGCoeliacDiseaseGuidelines Development Group; British Society of Gastroenterology. Diagnosisandmanagementofadultcoeliacdisease:guidelinesfrom theBritishSocietyofGastroenterology.Gut.2014;63:1210‐1228. 15. Dipper CR, Maitra S, Thomas R, et al. Anti‐tissue transglutaminaseantibodiesinthefollow‐upofadultcoeliacdisease.Aliment Pharmacol Ther.2009;30:236‐244. 16. SilvesterJA,KuradaS,SzwajcerA,KellyCP,LefflerDA,Duerksen DR.Testsforserumtransglutaminaseandendomysialantibodies donotdetectmostpatientswithceliacdiseaseandpersistentvillousatrophyongluten‐freediets:ameta‐analysis.Gastroenterology. 2017;153:689‐701. 17. LeonardMM,FasanoA.Zero,one,ortwoendoscopiestodiagnose andmonitorpediatricceliacdisease?thejuryisstillout.J Pediatr Gastroenterol Nutr.2017;65:270‐271. 18. Koletzko S, Auricchio R, Dolinsek J, et al. No need for routine endoscopyinchildrenwithceliacdiseaseonagluten‐freediet.J Pediatr Gastroenterol Nutr.2017;65:267‐269. 19. CominoI,RealA,VivasS,etal.Monitoringofgluten‐freedietcomplianceinceliacpatientsbyassessmentofgliadin33‐merequivalentepitopesinfeces.Am J Clin Nutr.2012;95:670‐677. 20. CominoI,Fernández‐BañaresF,EsteveM,etal.Fecalglutenpeptidesreveallimitationsofserologicaltestsandfoodquestionnaires for monitoring gluten‐free diet in celiac disease patients. Am J Gastroenterol.2016;111:1456‐1465. 21. MorenoML,Rodríguez‐HerreraA,SousaC,CominoI.Biomarkers tomonitorgluten‐freedietcomplianceinceliacpatients.Nutrients. 2017;9:46. 22. Soler M, Estevez MC, Moreno Mde L, Cebolla A, Lechuga LM. Label‐freeSPRdetectionofglutenpeptidesinurinefornon‐inva‐ siveceliacdiseasefollow‐up.Biosens Bioelectron.2016;79:158‐164. 23. SyageJA,Kelly CP,DickasonMA,etal.Determinationof gluten consumptioninceliacdiseasepatientsonagluten‐freediet.Am J Clin Nutr.2018;107:201‐207. 24. LandisJR,KochGG.Themeasurementofobserveragreementfor categoricaldata.Biometrics.1977;33:159‐174. 25. WhiteLE,BannermanE,GillettPM.Coeliacdiseaseandtheglutenfreediet:areviewoftheburdens;factorsassociatedwithadherenceandimpactonhealth‐relatedqualityoflife,withspecificfocus onadolescence.J Hum Nutr Diet.2016;29:593‐606. 26. SwiftC,WoodwardJM.Editorial:knowledge‐aprescriptionforcoeliacdisease.Aliment Pharmacol Ther.2018;48:478‐479. 27. HalmosEP,DengM,KnowlesSR,SainsburyK,MullanB,Tye‐Din JA.Foodknowledgeandpsychologicalstatepredictadherencetoa gluten‐freedietinasurveyof5310AustraliansandNewZealanders withcoeliacdisease.Aliment Pharmacol Ther.2018;48:78‐86. 28. Silvester JA, Graff LA, Rigaux L, Walker JR, Duerksen DR. Symptomatic suspected gluten exposure is common among patientswithcoeliacdiseaseonagluten‐freediet.Aliment Pharmacol Ther.2016;44:612‐619. 29. WagnerG,BergerG,SinnreichU,etal.Qualityoflifeinadolescents withtreatedcoeliacdisease:influenceofcomplianceandageatdiagnosis.J Pediatr Gastroenterol.2008;47:555‐561. 30. BruinsMJ.Theclinicalresponsetoglutenchallenge:areviewofthe literature. Nutrients.2013;5:4614‐4641. 31. CatassiC,FabianiE,IaconoGD,etal.Aprospective,double‐blind, placebo‐controlledtrialtoestablishasafeglutenthresholdforpatientswithceliacdisease.Am J Clin Nutr.2007;85:160‐166. 32. HallNJ,RubinGP,CharnockA.Intentionalandinadvertentnon‐ad‐ herenceinadultcoeliacdisease.Across‐sectionalsurvey.Appetite. 2013;68:56‐162. 33. LefflerDA,Edwards‐GeorgeJ,DennisM,etal.Factorsthatinfluenceadherencetoagluten‐freedietinadultswithceliacdisease. Dig Dis Sci.2008;53:1573‐1581. 34. Olsson C, Lyon P, Hörnell A, Ivarsson A, Sydner YM. Food that makesyoudifferent:thestigmaexperiencedbyadolescentswith celiacdisease.Qual Health Res.2009;19:976‐984. 35. Sansotta N, Amirikian K, Guandalini S, Jericho H. Celiac disease symptomresolution:effectivenessofthegluten‐freediet.J Pediatr Gastroenterol Nutr.2018;66:48‐52. 36. Rubio‐TapiaA,HillID,KellyCP,CalderwoodAH,MurrayJA.ACG clinicalguidelines:diagnosisandmanagementofceliacdisease.Am J Gastroenterol.2013;108:656‐676. 37. AbdulkarimAS,BurgartLJ,SeeJ,MurrayJA.Etiology ofnonresponsive celiac disease: results of a systematic approach. Am J Gastroenterol.2002;97:2016‐2021. 38. Dewar DH, Donnelly SC, McLaughlin SD, Johnson MW, Ellis HJ, Ciclitira PJ. Celiac disease: management of persistent symptoms in patients on a gluten‐free diet. World J Gastroenterol. 2012;18:1348‐1356. 39. Leffler DA, Dennis M, Hyett B, Kelly E, Schuppan D, Kelly CP. Etiologiesandpredictorsofdiagnosisinnonresponsiveceliacdisease.Clin Gastroenterol Hepatol.2007;5:445‐450. 40. Fang H, King KS, Larson JJ, et al. Undetectable negative tissue transglutaminaseIgAantibodiespredictmucosalhealingintreated coeliacdiseasepatients.Aliment Pharmacol Ther.2017;46:681‐687. | 9 COMINO et al. How to cite this article:CominoI,SeguraV,OrtigosaL,etal. Prospectivelongitudinalstudy:useoffaecalgluten immunogenicpeptidestomonitorchildrendiagnosedwith coeliacdiseaseduringtransitiontoagluten‐freediet.Aliment Pharmacol Ther. 2019;00:1–9. https://doi.org/10.1111/ apt.15277 APPENDIX 1 The authors’ complete affiliations IsabelComino,VerónicaSegura,CarolinaSousa,Departamentode MicrobiologíayParasitología,FacultaddeFarmacia,Universidadde Sevilla,Sevilla,Spain;LuisOrtigosa,HospitalUniversitarioNuestra Señora de La Candelaria, Tenerife, Spain; Beatríz Espín, Hospital Universitario Infantil Virgen del Rocío, Sevilla, Spain; Gemma Castillejo,UnitatPediàtricadeTrastornsRelacionatsambelGluten del Camp de Tarragona. IISPV, URV, Reus, Tarragona, Spain; José Antonio Garrote, Hospital Universitario Río Hortega, Valladolid, Spain; Carlos Sierra, Hospital Materno‐Infantil—Hospital Regional Universitario de Málaga, Málaga, Spain; Antonio Millán, Hospital UniversitarioVirgendeValme,Sevilla,Spain;CarmenRibes‐Koninckx, HospitalUniversitarioyPolitécnicoLaFe,Valencia,Spain;Enriqueta Román, Hospital Universitario Puerta del Hierro, Majadahonda, Madrid, Spain; Alfonso Rodríguez‐Herrera, Instituto Hispalense dePediatría,Sevilla,Spain;JacoboDíaz,HospitalUniversitariode Ceuta,INGESA,Ceuta,Spain;JocelynAnneSilvester,Universityof Manitoba,Winnipeg,Manitoba,Canada;BostonChildren'sHospital andBethIsraelDeaconessMedicalCenter,Boston,Massachusetts; ÁngelCebolla,BiomedalS.L.,Sevilla,Spain. 65 66 Segura, V., Ruiz-Carnicer, Á., Mendía, I., Garzón-Benavides, M., Pizarro, Á. E., Comino, I., & Sousa, C. (2023). Evaluation of the usefulness of an automatable immunoassay for monitoring celiac disease by quantification of immunogenic gluten peptides in urine. Nutrients, 15(7), 1730. https://doi.org/10.3390/nu15071730. CAPÍTULO 4 Nutrients 2023,15, 1730 4 of 12 in the result zone of the membrane, providing a signal. The absence suggested a negative result. The blue control line was always used as a test control [5,12,17]. 2.6. Statistical Analysis Statistical analyses were performed using Microsoft ® Office Excel (2016), SPSS 25.0 for Windows (SPSS Inc., Chicago, IL, USA) and the Sigma Plot software package (version 12.0; Systat Software, Inc., San Jose, CA, USA). Relative affinity curves were obtained by plotting the maximum absorbance percentage against the reference standard concentration (ng/mL), and EC 50 was calculated. EC 50 was defined as the concentration of the line that reduced the maximum absorbance by 50% in the assay. The cross-reactivity (CR) was determined by calculating: (EC 50 of the standard with the highest antibody affinity/EC 50 of each tested standard) ×100. The linearity of the method was established using mobile slope calculations between different points of the line and coefficient correlations (r2). The working range was established between the highest and lowest concentration values with satisfactory accuracy and precision. The acceptance criteria were a coefficient of variation of less than 20% [CV (%) = standard deviation (SD)/mean × 100%]. The slope ( ∆ y/ ∆ x) between points was calculated using the following equation: ∆ y/ ∆ x = (Y1 − Y0)/(X1 − X0), Y = concentration (ng/mL 33-mer peptide), and X = absorbance. The limit of quantification (LOQ) was defined as the smallest standard concentration with an intra- and inter-day imprecision lower than 20%. The limit of detection (LOD) of the assay was calculated as follows: mean samples replicates + 3 × SD. A paired Student’s t-test was used to analyze the quantitative variables. A p-value < 0.05 was considered statistically significant [28]. 3. Results 3.1. Optimization and Validation of the ELISA Working Conditions The assay was based on a previously developed sandwich ELISA for the analysis of GIP in stools using G12 moAb [ 21 , 29 ]. To obtain a higher sensitivity, new working conditions and applications were determined. The 33-mer peptide was used as a reference standard, which is one of main contributors to the immunogenicity of gluten [ 30 ] and is recognized by the gluten-specific celiac T cells. The influence of different parameters, such as the dilution solution assay, the coating moAb, the curve standard, and the capture moAb concentration, were studied to improve the ELISA conditions for GIP detection in the urine matrix. The optimal concentration of the capture moAb (G12-HRP) was determined using dilutions of 1:100,000 and 1:200,000, with increments of 20,000. For this, six curves with known concentrations of the 33-mer peptide (1–1000 ng/mL 33-mer) were generated (Figure 1A). The 1:100,000 dilution had the lowest EC 50 of those studied, suggesting that it was more specific for the 33-mer peptide (Figure 1B). Therefore, it was used as the optimal moAb titration in the ELISA. 3.1.1. Linearity and Working Range To define the working range, the 33-mer peptide standard was serially diluted with the dilution solution to known concentrations (12.5, 6.25, 3.12, 1.56 ng/mL) (Figure 2B). Each concentration was tested five times in duplicate. Furthermore, it was performed on different plates, on different days, and by different analysts to demonstrate, in turn, the robustness of the assay. The coefficient of variation (CV%) for the same concentrations was below 20% in the tested standards, ensuring the good precision and robustness of the method. Similarly, the CV between consecutive standards had a CV% between 20–80%, showing the correct discrimination among the consecutive standards, until the 12.5 ng/mL concentration. In this type of assay, the approximation of the different standard values is carried out in a polynomial manner (curve), and the slope ( ∆ y/ ∆ x) between the two concentrations of two contiguous points was determined (12.5–6.25, 6.25–3.12, and 3.12–1.56 ng/mL) . The slope ( ∆ y/ ∆ x) between points was calculated using the follow- Nutrients 2023,15, 1730 5 of 12 ing equation: ∆y/∆x = (Y1 −Y0)/(X1 −X0) , Y = concentration (ng/mL 33-mer peptide), and X = absorbance (Figure 2A). Subsequently, the CV% of the different slopes ( ∆ y/ ∆ x) between the different concentrations was calculated and found to be constant, not exceeding 20% [28].                                                                                                                                      ff    ff     ff             ffi                                                          ff               Δ Δ                   Δ Δ         Δ Δ   − −                   ff   Δ Δ    ff              Figure 1. G12-HRP moAb titration for sandwich ELISA optimization. ( A ). Relative affinity of G12- HRP moAb for 33-mer peptide. (B). Standard reference curves. EC50 and cross-reactivity (CR) were obtained by G12 ELISA. EC 50 is the line concentration that reduces the maximum absorbance by 50% in the assay. CR was calculated as follows: (EC 50 of the antigen for which the moAb was raised/EC 50 of each antigen assayed) × 100. These assays were performed in duplicate moAb, monoclonal antibody. In addition, the regression coefficient was calculated from the polynomial approximation of the analyzed standards, and it was verified that there were no statistically significant differences between the curves (correlation coefficient (r2) > 0.99) (Figure 2C). 3.1.2. Matrix Study A common challenge in immunoassays is matrix interference. These interferences can be reduced by dilution or by using a matrix-matched calibration curve. Therefore, the behavior of the 33-mer peptide standard was evaluated in the dilution solution and the urine matrix. Urine samples from a patient with CD on a GFD strictly controlled by a dietary questionnaire and previously analyzed by LFIA with negative results, were used. Twelve curves with known concentrations of the 33-mer peptide (12.5, 6.25, 3.12, and 1.56 ng/mL) were performed. The CV% for the same concentration in each matrix showed <20%. A comparison of the spiked curves (in dilution solution or urine) did not show a statistically significant difference (p> 0.1). Thus, the ELISA detected only the 33-mer peptide without any interference from the dilution solution or urine matrix (Figure 3). Nutrients 2023,15, 1730 6 of 12 Figure 2. ( A ) Calculation of slopes ( ∆ y/ ∆ x) between the different concentrations of the curve and the correlation coefficient (r2). The slope ( ∆ y/ ∆ x) between concentrations was calculated using the equation; ∆ y/ ∆ x = (Y1 − Y0)/(X1 − X0); Y = concentration (ng/mL 33-mer), and X = absorbance. ( B ). Polynomial representation of 33-mer concentrations (12.5 and 1.56 ng/mL). ( C ). Correlation coefficient (r2) calculated from the polynomial approximation of the standards analyzed. Abs, absorbance; CV%, coefficient of variation; SD, standard deviation.                                                                                       ff      ffi                                                                                                                                          ff                                                                                                  Figure 3. Standard curve for the quantification of the 33-mer used as reference material. The linear portion of the curve is between 12.5 and 1.56 ng/mL. The solid black line represents the mean values and standard deviations (SD) of 12 determinations per concentration in the urine matrix and the dashed line in the dilution solution. Nutrients 2023,15, 1730 7 of 12 3.1.3. Accuracy and Precision The most useful method for calculating the accuracy was the recovery test using three different concentrations. The Association of Official Agricultural Chemists (AOAC) has established an optimal recovery percentage for spiked samples of 80% to 120% [ 31 ]. According to this criterion, the recovery (%R) of all the spiked samples was satisfactory using this method. All of the non-spiked samples were below the LOQ. According to the Eurachem guide [ 28 ], the results must be obtained from an average of 6–15 replicates of each material, with the same equipment, analyst, and laboratory, and in a short period [ 28 ]. In this study, the accuracy was calculated by taking nine measurements per concentration (20, 10, 5, 2.5, 1,25 ng/mL of 33-mer peptide), and these measurements were made on the same day. The results indicated an accuracy of 91% on day one and 89% on day two. In addition, to calculate the precision, nine replicates were made by concentrations on two different days, obtaining a precision of 90% (Table 1). Table 1. Analysis of the spiked urine samples by sandwich ELISA. Results are expressed as ng/mL of 33-mer (mean ± SD) and percentage of recovery (R). N = number of analyses; <LOQ, less than the limit of quantification; SD, standard deviation. Spiked Sample (ng/mL) N ng/mL ±SD % R Accuracy (%) Precision (%) Day 1 20.0 9 20.3 ±3.3 101 91% 90% 10.0 9 8.1 ±0.5 81 5.0 9 4.6 ±0.4 91 2.5 9 <LOQ 1.25 9 <LOQ 0 9 <LOQ Day 2 20.0 9 17.8 ±2.8 89 89% 10.0 9 8.2 ±0.6 82 5.0 9 4.8 ±0.3 96 2.5 9 <LOQ 1.25 9 <LOQ 0 9 <LOQ 3.2. Effect of the Urine Samples on the Assay Performance To evaluate the usefulness of the developed method, seven adult volunteers without suspicion of CD and seven patients with CD on a GFD strictly controlled by a dietary questionnaire were recruited into a pilot study. These samples were previously analyzed by LFIA and showed positive results in the volunteers without suspected CD and negative results in the patients with CD. The urine samples were subjected to extraction and concentration. The urine samples were mixed with surfactant agents and incubated in a thermostatic bath. The sample was passed through a 3 kDa filter and centrifuged. Once the samples were obtained, they were diluted in the dilution solution, and a G12-G12 sandwich ELISA was performed. The results showed that 100% (7/7) of the subjects without CD were GIP positive, with values between 0.40 and 1.01 ng GIP/mL of urine. However, the urine samples from the patients with CD on a GFD were negative for GIP (<LOQ). Considering these results, the LOQ was established as 0.312 ng GIP/mL of urine with a concentration factor of 10, and a dilution factor of 2. The LOQ was determined to be reliable because it was also found to be above the LOD of this procedure. The LOD, calculated as the mean of seven GIP negative sample replicates + 3 SD, was 0.075 ng GIP/mL of urine. Nutrients 2023,15, 1730 8 of 12 3.3. Stability of GIP Conditions that allow the storage/accumulation of urine samples over time, without loss of the GIP signal, were studied to allow for the bulk analysis of urine samples. The storage and transport temperature conditions of the samples should be adequate to guarantee the stability of GIP in the urine. A stability study was carried out with three volunteers: two healthy adult volunteers without suspicion of CD (volunteers 1 and 2) and a patient with CD on a strict GFD (volunteer 3). Four aliquots of each urine sample were stored at room temperature, 4 ◦C and −20 ◦C. The GIP in the urine was evaluated on days 0, 4 and 16 after collection. The urines were tested twice in each condition. To check the stability of the GIP in the urine, each condition was compared with the time point t0, by executing at least two determinations per condition. Our results showed that urinary GIP is stable for up to 16 days when the samples are stored at different temperature conditions (room temperature, 4 ◦ C, and − 20 ◦ C) according to the sandwich ELISA (deviation below 20% of the expected value) (Figure 4).                              −                                                                                                                                               tt     tt                                                                 Figure 4. Stability of urinary GIP over a 16-day period. ( A ) room temperature. ( B ) 4 ◦ C. ( C ) − 20 ◦ C. Volunteer one is represented in blue, two in green, and three in yellow. Each point represents the mean of duplicate measurements plus the SD if the GIP was measurable. GIP, gluten immunogenic peptides; LOQ, limit of quantification; ND, non-determined; n/a, non-applied. 4. Clinical Study 4.1. Study Design and Population The study population consisted of 12 (67%) females and six (33%) males, with a median age of 42 years. Table 2presents the descriptive data of the patients; 72% of the patients started the study because of the presence of symptoms, 94% were seropositive (CD antibodies) at diagnosis, and the most common histological lesion was Marsh II-III (89%). Participant retention was 72% at three months, 72% at six months, and 61% at 12 months (the most common reason for being lost to follow-up was moving out of the study area, not attending follow-up visits, and forgetting to collect samples). Nutrients 2023,15, 1730 9 of 12 Table 2. Characteristics of the patients enrolled in the study. CD, celiac disease. Characteristics Patients, n % Sex Female 12 67 Male 6 33 Age Median age (42) Duodenal histology Marsh 0-I 2 11 Marsh II-III 16 89 Symptoms Asymptomatic 5 28 Symptomatic 13 72 CD antibodies CD antibodies positive 17 94 CD antibodies negative 1 6 4.2. Analysis of Urine GIP At the initial visit, before starting the GFD, 100% (18/18) of the patients had detectable GIP in the provided urine sample, with a range of 0.39 and 2.14 ng GIP/mL of urine. After diagnosis and treatment with a GFD, the rate of GIP positive urine was 38% at three months, 38% at six months, and 27% at 12 months, and in general, the GIP concentration in those urine samples also decreased. In particular, the GFD compliance rates increased as the study progressed (Figure 5A,B). Therefore, significant differences in the GIP excretion levels were observed in the population before and after the GFD initiation, as shown in Figure 5. These results were comparable to those obtained in a pediatric population at the follow-up of two years reported by Comino et al. [ 29 ], in which fecal GIP was evaluated by ELISA G12. Therefore, a study is currently underway in a pediatric population to corroborate these results.                                                                                                                                                                                                                                                                                                                                         ff                                                                                                                                                                                                                 Figure 5. Concentration of urine GIP from patients with CD at the baseline visit and over 12 months after the initiation of a GFD. ( A ) Percentage distribution of the GFD-treated patients with CD according to GIP concentration. ( B ) GIP levels in urine at the basal and follow-up visits (basal, three, six, and 12 months). CD, celiac disease; GFD, gluten-free diet; GIP, gluten immunogenic peptides; LOQ, limit of quantification. 5. Conclusions We propose the first automated and highly sensitive method for the quantitative detection of GIP in urine for the monitoring of CD. In this study, several parameters were optimized to obtain a reproducible, selective, and sensitive method. Additionally, this method demonstrated the feasibility of clearly identifying gluten consumption by measuring multiple urine samples from healthy adult volunteers with habitual consumption of gluten, and from patients with CD on a strict GFD. Furthermore, a small prospective Nutrients 2023,15, 1730 10 of 12 clinical study was carried out and the results showed statistically significant differences in the determination of GIP in urine between individuals at CD diagnosis and follow-up. As the presence of u-GIP is direct evidence that gluten intake has occurred, this method could either be used to evaluate the adherence to a GFD or for the confirmation of gluten intake in cases where a gluten challenge is necessary, such as for confirmation of the disease or in clinical trials where CD drugs are being tested. However, further studies with larger numbers of pediatric and adult patients are needed to support the study findings for the implementation of this new method in the clinical laboratories of specialized health centers. In addition, interlaboratory trial studies would be required to establish the efficacy and comparability of the new method, as well as to validate the uncertainty estimates indicated. Author Contributions: Conceptualization, V.S., Á.R.-C., I.M., M.G.-B., Á.E.P., I.C. and C.S.; Data curation, V.S., Á.R.-C., I.M., I.C. and C.S.; Formal analysis, V.S., Á.R.-C., I.M., I.C. and C.S.; Funding acquisition, I.C. and C.S.; Investigation, V.S., Á.R.-C., I.M., M.G.-B., Á.E.P., I.C. and C.S.; Methodology, V.S., Á.R.-C. and I.M.; Project administration, I.C. and C.S.; Resources, I.C. and C.S.; Software, V.S. and Á.R.-C.; Supervision, I.C. and C.S.; Validation, I.C. and C.S.; Visualization, I.C. and C.S.; Writing—original draft, V.S., Á.R.-C., I.C. and C.S.; Writing—review and editing, V.S., Á.R.-C., I.M., M.G.-B., Á.E.P., I.C. and C.S. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Grant DI-17-09627 and Grant RTC2019-006806-1 funded by MCIN/AEI/10.13039/501100011033, by Grant P18-RT-3004 funded by PAIDI I+D+i Junta de Andalucía (Consejería de Economía y Conocimiento), and by Grant AT17_5489_USE PAIDI Actividades de Transferencia 2017 Junta de Andalucía (Consejería de Economía, Conocimiento, Empresas y Universidad). Institutional Review Board Statement: The study protocol was reviewed by the ethics committee of each institution. CEI de los Hospitales Universitarios Virgen Macarena-Virgen del Rocío (Seville, Spain). Informed Consent Statement: Informed consent was obtained from all of the subjects involved in the study. Data Availability Statement: Data sharing is not applicable to this article. Conflicts of Interest: Author Irati Mendíaz is employed by the company Biomedal S.L. The contribution of the author in the paper is participation in the experimental design. However, the Biomedal S.L. company did not contribute financially. Therefore, there is no conflict of interest in relation to the Biomedal S.L., company. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. References 1. Al-Toma, A.; Volta, U.; Auricchio, R.; Castillejo, G.; Sanders, D.S.; Cellier, C.; Mulder, C.J.; Lundin, K.E.A. European Society for the Study of Coeliac Disease (ESsCD) guideline for coeliac disease and other gluten-related disorders. United Eur. Gastroenterol. J. 2019,7, 583–613. [CrossRef] 2. Husby, S.; Koletzko, S.; Korponay-Szabó, I.; Kurppa, K.; Mearin, M.L.; Ribes-Koninckx, C.; Shamir, R.; Troncone, R.; Auricchio, R.; Castillejo, G.; et al. European Society Paediatric Gastroenterology, Hepatology and Nutrition Guidelines for Diagnosing Coeliac Disease 2020. J. Pediatr. Gastroenterol. Nutr. 2020,70, 141–156. [CrossRef] 3. Lungaro, L.; Manza, F.; Costanzini, A.; Barbalinardo, M.; Gentili, D.; Caputo, F.; Guarino, M.; Zoli, G.; Volta, U.; De Giorgio, R.; et al. Osteoporosis and celiac disease: Updates and hidden pitfalls. Nutrients 2023,15, 1089. [CrossRef] [PubMed] 4. Ludvigsson, J.F.; Bai, J.C.; Biagi, F.; Card, T.R.; Ciacci, C.; Ciclitira, P.J.; Green, P.H.; Hadjivassiliou, M.; Holdoway, A.; van Heel, D.A.; et al. BSG Coeliac Disease Guidelines Development Group; British Society of Gastroenterology. Diagnosis and management of adult coeliac disease: Guidelines from the British Society of Gastroenterology. Gut 2014,63, 1210–1228. [CrossRef] [PubMed] 5. Caio, G.; Ciccocioppo, R.; Zoli, G.; De Giorgio, R.; Volta, U. Therapeutic options for coeliac disease: What else beyond gluten-free diet? Dig. Liver Dis. 2020,52, 130–137. [CrossRef] [PubMed] 6. Caio, G.; Volta, U.; Sapone, A.; Leffler, D.A.; De Giorgio, R.; Catassi, C.; Fasano, A. Celiac disease: A comprehensive current review. BMC Med. 2019,17, 142. [CrossRef] 7. Muhammad, H.; Reeves, S.; Jeanes, Y.M. Identifying and improving adherence to the gluten-free diet in people with coeliac disease. Proc. Nutr. Soc. 2019,78, 418–425. [CrossRef] [PubMed] Nutrients 2023,15, 1730 11 of 12 8. Ruiz-Carnicer, Á.; Garzón-Benavides, M.; Fombuena, B.; Segura, V.; García-Fernández, F.; Sobrino-Rodríguez, S.; Gómez- Izquierdo, L.; Montes-Cano, M.A.; Rodríguez-Herrera, A.; Millán, R.; et al. Negative predictive value of the repeated absence of gluten immunogenic peptides in the urine of treated celiac patients in predicting mucosal healing: New proposals for follow-up in celiac disease. Am. J. Clin. Nutr. 2020,112, 1240–1251. [CrossRef] 9. Coto, L.; Mendia, I.; Sousa, C.; Bai, J.C.; Cebolla, A. Determination of gluten immunogenic peptides for the management of the treatment adherence of celiac disease: A systematic review. World J. Gastroenterol. 2021,27, 6306–6321. [CrossRef] 10. Stefanolo, J.P.; Tálamo, M.; Dodds, S.; de la Paz Temprano, M.; Costa, A.F.; Moreno, M.L.; Pinto-Sánchez, M.I.; Smecuol, E.; Vázquez, H.; Gonzalez, A.; et al. Real-World gluten exposure in patients with celiac disease on gluten-free diets, determined from gliadin immunogenic peptides in urine and fecal samples. Clin. Gastroenterol. Hepatol. 2021,19, 484–491.e1. [CrossRef] 11. Segura, V.; Ruiz-Carnicer, Á.; Sousa, C.; Moreno, M.L. New insights into non-dietary treatment in celiac disease: Emerging therapeutic options. Nutrients 2021,13, 2146. [CrossRef] [PubMed] 12. Wieser, H.; Ruiz-Carnicer, Á.; Segura, V.; Comino, I.; Sousa, C. Challenges of monitoring the gluten-free diet adherence in the management and follow-up of patients with celiac disease. Nutrients 2021,13, 2274. [CrossRef] 13. Sandström, O.; Norström, F.; Carlsson, A.; Högberg, L.; van der Palz, M.; Stenhammar, L.; Webb, C.; Ivarsson, A.; Myléus, A. Five-year follow-up of new cases after a coeliac disease mass screening. Arch. Dis. Child. 2022 ,107, 596–600. [CrossRef] [PubMed] 14. Machado, M.V. New Developments in celiac disease treatment. Int. J. Mol. Sci. 2023,24, 945. [CrossRef] 15. Garzón-Benavides, M.; Ruiz-Carnicer, Á.; Segura, V.; Fombuena, B.; García-Fernandez, F.; Sobrino-Rodriguez, S.; Gómez- Izquierdo, L.; Montes-Cano, M.A.; Millan-Domínguez, R.; María del Carmen Rico, M.; et al. Regular monitoring of patients with coeliac disease by determinations of gluten immunogenic peptides in urine improved gluten-free diet adherence and the healing of duodenal mucosa: Proposal on a new algorithm for the follow-up of coeliac disease. Aliment. Pharmacol. Ther. 2023, in press. 16. Rodrigo, L.; Beteta-Gorriti, V.; Alvarez, N.; Gómez de Castro, C.; de Dios, A.; Palacios, L.; Santos-Juanes, J. Cutaneous and mucosal manifestations associated with celiac disease. Nutrients 2018,10, 800. [CrossRef] 17. Protocolo Para el Diagnóstico Precoz de la Enfermedad Celíaca. Available online: https://www3.gobiernodecanarias.org/ sanidad/scs/contenidoGenerico.jsp?idDocument=ed8e95ff-578d-11e8-8844-65f683d38a9e&idCarpeta=0428f5bb-8968-11dd-b7 e9-158e12a49309 (accessed on 10 December 2022). 18. Comino, I.; Real, A.; Vivas, S.; Síglez, M.Á.; Caminero, A.; Nistal, E.; Casqueiro, J.; Rodríguez-Herrera, A.; Cebolla, A.; Sousa, C. Monitoring of gluten-free diet compliance in celiac patients by assessment of gliadin 33-mer equivalent epitopes in feces. Am. J. Clin. Nutr. 2012,95, 670–677. [CrossRef] [PubMed] 19. Soler, M.; Estevez, M.C.; del Moreno, M.; Cebolla, A.; Lechuga, L.M. Label-free SPR detection of gluten peptides in urine for non-invasive celiac disease follow-up. Biosens. Bioelectron. 2016,79, 158–164. [CrossRef] 20. Moreno, M.D.L.; Cebolla, Á.; Munõz-Suano, A.; Carrillo-Carrion, C.; Comino, I.; Pizarro, Á.; León, F.; Rodríguez-Herrera, A.; Sousa, C. Detection of gluten immunogenic peptides in the urine of patients with coeliac disease reveals transgressions in the gluten-free diet and incomplete mucosal healing. Gut 2017,66, 250–257. [CrossRef] 21. Comino, I.; Fernández-Bañares, F.; Esteve, M.; Ortigosa, L.; Castillejo, G.; Fambuena, B.; Ribes-Koninckx, C.; Sierra, C.; Rodríguez- Herrera, A.; Salazar, J.C.; et al. Fecal gluten peptides reveal limitations of serological tests and food questionnaires for monitoring gluten-free diet in celiac disease patients. Am. J. Gastroenterol. 2016,111, 1456–1465. [CrossRef] 22. Gerasimidis, K.; Zafeiropoulou, K.; Mackinder, M.; Ijaz, U.Z.; Duncan, H.; Buchanan, E.; Cardigan, T.; Edwards, C.A.; McGrogan, P.; Russell, R.K. Comparison of clinical methods with the faecal gluten immunogenic peptide to assess gluten intake in coeliac disease. J. Pediatr. Gastroenterol. Nutr. 2018,67, 356–360. [CrossRef] [PubMed] 23. Peláez, E.C.; Estevez, M.C.; Domínguez, R.; Sousa, C.; Cebolla, A.; Lechuga, L.M. A compact SPR biosensor device for the rapid and efficient monitoring of gluten-free diet directly in human urine. Anal. Bioanal. Chem. 2020 ,412, 6407–6417. [CrossRef] [PubMed] 24. Porcelli, B.; Ferretti, F.; Biviano, I.; Santini, A.; Cinci, F.; Vascotto, M.; Grande, E.; Quagliarella, F.; Terzuoli, L.; Bizzaro, N.; et al. Testing for fecal gluten immunogenic peptides: A useful tool to evaluate compliance with gluten-free diet by celiacs. Ann. Gastroenterol. 2020,33, 631–637. [CrossRef] [PubMed] 25. Porcelli, B.; Ferretti, F.; Cinci, F.; Biviano, I.; Santini, A.; Grande, E.; Quagliarella, F.; Terzuoli, L.; Bacarelli, M.R.; Bizzaro, N.; et al. Fecal gluten immunogenic peptides as indicators of dietary compliance in celiac patients. Minerva Gastroenterol. Dietol. 2020 , 66, 201–207. [CrossRef] 26. Fernández-Bañares, F.; Beltrán, B.; Salas, A.; Comino, I.; Ballester-Clau, R.; Ferrer, C.; Molina-Infante, J.; Rosinach, M.; Modolell, I.; Rodríguez-Moranta, F.; et al. Persistent villous atrophy in de novo adult patients with celiac disease and strict control of gluten-free diet adherence: A multicenter prospective study (CADER Study). Am. J. Gastroenterol. 2021 ,116, 1036–1043. [CrossRef] 27. Adachi, J.; Kumar, C.; Zhang, Y.; Olsen, J.V.; Mann, M. The human urinary proteome contains more than 1500 proteins, including a large proportion of membrane proteins. Genome Biol. 2006,7, R80. [CrossRef] 28. Magnusson, B.; Örnemark, U. (Eds.) Eurachem Guide: The Fitness for Purpose of Analytical Methods—A Laboratory Guide to Method Validation and Related Topics, 2nd ed.; Eurachem: Gembloux, Belgium, 2014; ISBN 978-91-87461-59-0. Available online: www.eurachem.org (accessed on 10 December 2022). 29. Comino, I.; Segura, V.; Ortigosa, L.; Espín, B.; Castillejo, G.; Garrote, J.A.; Sierra, C.; Millán, A.; Ribes-Koninckx, C.; Román, E.; et al. Prospective longitudinal study: Use of faecal gluten immunogenic peptides to monitor children diagnosed with coeliac disease during transition to a gluten-free diet. Aliment. Pharmacol. Ther. 2019,49, 1484–1492. [CrossRef] Nutrients 2023,15, 1730 12 of 12 30. Shan, L.; Molberg, Ø.; Parrot, I.; Hausch, F.; Filiz, F.; Gray, G.M.; Sollid, L.M.; Khosla, C. Structural basis for gluten intolerance in celiac sprue. Science 2002,297, 2275–2279. [CrossRef] 31. Abbott, M.; Hayward, S.; Ross, W.; Godefroy, S.B.; Ulberth, F.; Van Hengel, A.J.; Roberts, J.; Akiyama, H.; Popping, B.; Yeung, J.M.; et al. Validation procedures for quantitative food allergen ELISA methods: Community Guidance and Best Practices. J. AOAC Int. 2010,93, 442–445. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. 82 Segura, V., Díaz, J., Ruiz-Carnicer, Á., Muñoz-Suano, A., Carrillo-Carrión, C., Sousa, C., Cebolla, Á., & Comino, I. (2021). Rapid, effective, and versatile extraction of gluten in food with application on different immunological methods. Foods (Basel, Switzerland), 10(3), 652. https://doi.org/10.3390/foods10030652. CAPÍTULO 5 Foods 2021,10, 652 4 of 16 2.4. Techniques Employed 2.4.1. Enzyme-Linked Immunosorbent Assays (ELISAs) Gluten in foods was determined using the RIDASCREEN ® FAST Gliadin (R-Biopharm) and RIDASCREEN ® Gliadin competitive (R-Biopharm) ELISA kits. Both assays utilize the R5 monoclonal antibody (moAb) [ 38 , 39 ]. All samples were analyzed according to the manufacturer’s instructions. 2.4.2. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE) and G12/A1 Western Blotting SDS-PAGE and immunoblotting were performed under standard conditions. Samples were diluted in SDS-PAGE denaturing buffer (62.5 mM Tris–HCl pH 6.8, 5% 2- mercaptoethanol, 2% SDS, 0.001% bromophenol blue, and 10% glycerol) [ 40 – 42 ]. Proteins resolved by 12.5% SDS-PAGE were transferred to a polyvinylidene difluoride (PVDF) membrane. The membranes were incubated with G12 or A1 moAbs o/n at 4 ◦ C. After washing, anti-mouse immunoglobulin G (IgG) alkaline phosphatase conjugated antibody (Sigma Aldrich) was added and incubated for 1 h at room temperature. 2.4.3. Lateral Flow Immunochromatographic Assays (LFIA) The LFIA GlutenTox Stick was used to develop the new analytical method (Hygiena, Seville, Spain). These sticks are based on the G12 and A1 moAbs. The sticks were calibrated against PWG gliadin. The food samples were extracted with 10 mL of UGES per g of food. After extraction, the samples were diluted 1:10 in dilution solution. The test sticks were dipped into the dilution solution (100 µ L) in a well of a microtiter plate. The results were determined quantitatively after 30 min. The GlutenTox ® Reader (Hygiena, Seville, Spain) was used for quantification by continuous scanning of the sticks using an optical detector. 2.5. Statistical Analysis Data analyses was performed using SPSS 25.0 for Windows (IBM SPSS Statistics, Armonk, NY: IBM Corp). Scatter diagrams were used as graphic indicators of the commutability between the values generated by each pair of extraction solutions. The equation of the line was calculated using the nonparametric Passing–Bablok regression [ 43 – 45 ]. This method is a statistical procedure that allows valuable estimation of analytical method agreement and possible systematic bias between them. The material was deemed commutable if the ratio obtained for a reference material using the two methods was consistent with the prediction interval (ISO, 2009). The Lin concordance correlation coefficient is the concordance (rLin) between a new test or measurement and a gold standard test or measurement. This coefficient combines a precision measure, represented by the Pearson’s correlation coefficient (rPerason), with an accuracy measure, represented by the bias correction coefficient (Cb). The McBrige scale was used to more stringently qualify the strength of the agreement. For continuous variables, the scale was almost perfect for values >0.99, substantial for values of 0.95 to 0.99, moderate for values of 0.90 to 0.95, and poor for values <0.90. Differences between groups were statistically compared by the median using the Wilcoxon test. 3. Results and Discussion 3.1. Characterization of UGES Protocol During the processing of some foods, the heat treatment of cooked and baked products leads to the formation of protein aggregates and gluten-derived peptides in insoluble matrices. This results in a heterogeneous mixture with a nonuniform distribution of gluten, which makes analyses even more difficult. It is necessary to optimize the extraction systems to produce the complete release and recovery of both prolamins and glutelins [ 46 , 47 ]. The time and temperature of incubation during the extraction process are two critical factors to consider [48]. Foods 2021,10, 652 5 of 16 In this work, the effect of temperature on UGES was evaluated (Figure 1). Several samples were extracted at two conditions: incubation in wheel agitation at room temperature and incubation in a water bath at 50 ◦ C. In both cases, the time was fixed at 1 h, and the amount of extracted gluten was quantified by sandwich ELISA (Figure 1A). Unheated processed food or simple samples (maize flour, oat flakes) did not need heat to achieve the complete extraction of gluten. In contrast, heat treatment was needed for complete extraction in samples with a complex composition (mixture food additives) and for heatprocessed samples (corn bread). Figure 1. Optimization of time and temperature in the Universal Gluten Extraction Solution (UGES) protocol. ( A ) Effect of temperature in different types of samples. ( B ) Effect of time in simple and complex solid samples. ( C ) Effect of extraction time in liquid samples without suspended solids. The experiments were performed in duplicate, and the mean ±standard deviation (SD) is shown. The influence of different incubation periods (2 min to 1 h) was studied. Maize flour (b1) and a precooked meat food (b2) were selected to optimize the extraction time. The b1 Foods 2021,10, 652 6 of 16 example was extracted by rotation at room temperature. The b2 example was heat-extracted. A minimum of 30 to 40 min was necessary to achieve the complete gluten extraction for both samples (Figure 1B). According to these results, the extraction time was fixed at 40 min for solids samples or containing polyphenol, tannins, or antioxidants, regardless of the use of heating in the extraction. Several additional assays were carried out with liquid samples without solids in suspension, such as milk, juices, organic drinks, beers, and soft drinks, to evaluate if the extraction time for these samples could be reduced. The results obtained by competitive ELISA indicated a reduction in the extraction time of these samples because they were satisfactory with only 2 min of agitation compared to the extraction carried out in solid samples (40 min) (Figure 1C). These findings indicated that the optimum extraction procedure depends largely on the type of sample. However, the advantage of UGES is its adaptability to different types of samples with slight modifications to the protocol, as described in Table 1. Table 1. Scheme of the steps in the extraction procedure with UGES depending on the type of sample. RT, room temperature. Extraction Protocol Type of Sample Amount of Sample Additives UGES Volume Procedure Liquid Without solids in suspension 1 mL 9 mL Shake 1–2 min Solid Non-heat-processed and simple composition 1 g 10 mL Incubate 40 min at RT (wheel agitator) Centrifuge 10 min at 2500×g Solid Heat-processed or complex composition 1 g 10 mL Incubate 40 min at 50 ◦ C (water bath) Centrifuge 10 min at 2500×g Solid or liquid Containing polyphenols, tannins, or antioxidants 1 g 1.0 g gelatin 0.4 g PVP 10 mL Incubate 40 min at 50 ◦ C (water bath) Centrifuge 10 min at 2500×g PVP, polyvinylpyrrolidone. 3.2. UGES Extraction Effectiveness Assessed by Sandwich ELISA 3.2.1. Determination of Gluten in Commercial Food Samples Having characterized UGES using different matrices, we next evaluated the application of UGES to the analysis of different commercial foods, which were labeled and not labeled as being gluten-free, by comparison to standard extraction procedure using 60% ethanol or Cocktail solution. UGES was highly efficient for simple and complex matrices, even with thermal processing. The gluten amount extracted with UGES was always higher (between 1.1- and 5.3-fold increase) than that extracted with 60% ethanol (Table 2). In addition, we found five food samples with a gluten content <20 ppm using 60% ethanol solution but >20 ppm gluten using UGES. The greater increases with UGES compared with 60% ethanol corresponded to heat-processed samples, such as corn bread, linseed toast snack food, and chips. Gluten proteins contain intermolecular bonds, which allow them to form a network when cooked or baked [ 2 ]. The results obtained demonstrated that the use of reducing and disaggregating agents is, thus, necessary to break disulfide bonds during the extraction procedure. Ethanol–water solutions used as extractants solubilize monomeric prolamins in non-processed food like flour. However, their efficiency is low when processed food is analyzed because protein aggregation occurs due to disulfide bond formation [34,49,50]. Foods 2021,10, 652 7 of 16 Table 2. Gluten content in ppm (mg/kg) of different samples by sandwich ELISA with 60% ethanol and Universal Gluten Extraction Solution (UGES). Gluten (ppm) Food Sample 60% Ethanol Solution UGES * Non-heat- processed samples Maize flour 23 26.0 (×1) Corn semolina 11.7 15.7 (×1.3) Oat flakes 60.3 66.8 (×1.1) Glucose syrup 29.3 33.7 (×1.2) Worcester sauce 4 5.2 (×1.3) Yogurt cereal bar 11 14 (×1.3) Garlic soup 30.2 33 (×1.1) Heat-processed or complex composition samples Rice cereals 7 11 (×1.6) Gluten-free-bread <3 4.6 (×1.5) Corn bread 3.2 15.7 (×4.9) Gluten-free breadcrumbs 4.3 8.2 (×1.9) Mixture food additives <3 5.2 (×1.7) Spelt cracker 77.2 115.2 (×1.5) Dog dry food 13.3 28.2 (×2.1) Linseed toast snack 14.9 78.8 (×5.3) Honey corn pops 5.5 24.9 (×4.5) Chips a 5.3 21.1 (×4.0) Dough flour 7.9 40.2 (×5.1) Popcorn 6 16.1 (×2.7) * Values in brackets are the yield factor compared to that with 60% ethanol solution and UGES. Additionally, 93 commercial food samples, labeled and not labeled as being glutenfree, were analyzed using UGES and Cocktail solution (Table 3). Gluten was undetectable, less than the limit of quantification (<LOQ) in 51 of the samples by both extraction methods (data no shown). No significant differences in the gluten content of the food were evident using UGES and Cocktail solution; the yield factors found were 0.5 to 1.7 times. However, the “chips a” sample contained <20 ppm gluten using Cocktail solution and >20 ppm using UGES. This result is very important because individuals with CD can only ingest products with a gluten content <20 ppm. A statistical evaluation of the extraction process was performed by comparing UGES versus ethanol and UGES versus Cocktail solution. Passing–Bablok linear regression analysis with a prediction interval of 95% was calculated to describe the commutability of the results. Lin’s concordance correlation coefficient (r Lin ) was determined using a precision measure (r Pearson ) and an accuracy measure (C b ). Lastly, the differences between groups were statistically compared using the Wilcoxon test. The statistical analysis of the gluten concentrations obtained by UGES and 60% ethanol showed that the data were not commutable (Figure 2B(b1)), since the precision and accuracy were low (r Pearson = 0.8252, C b = 0.8144). This indicated a poor agreement strength, according to the McBrige scale (<0.90), with r Lin = 0.6720 (Figure 2A). In addition, these results showed that UGES had a better extraction capacity than 60% ethanol solution, since there was a significant difference in the results according to the Wilcoxon test (p< 0.0001) (Figure 2A). However, UGES and Cocktail solution were commutable according to Passing–Bablok regression (Figure 2B(b2)) with high precision and accuracy coefficients (r Pearson = 0.9694, C b = 0.9984) (Figure 2A). Lin’s correlation concordance coefficient (r Lin ) was 0.9679, indicating substantial agreement strength according to the McBrige scale (>0.95). Furthermore, no significant difference was found (p= 0.2381) according to the Wilcoxon test, indicating that Cocktail solution could be replaced by UGES. Foods 2021,10, 652 8 of 16 Table 3. Gluten content in ppm (mg/kg) of different samples by sandwich ELISA with Cocktail solution and Universal Gluten Extraction Solution (UGES). Gluten (ppm) Food Sample Cocktail Solution UGES * Non-heat- processed samples Maize flour 27 26.0 (×1.0) Corn semolina 14.8 15.7 (×1.1) Oat flakes 64 66.8 (×1.0) Glucose syrup 36.3 33.7 (×0.9) Worcester sauce 5.3 5.2 (×1.0) Yogurt cereal bar 13 14 (×1.1) Garlic soup 30.8 33.0 (×1.1) Heat-processed or complex composition samples Rice cereals 14 11 (×0.8) Gluten-free-bread 3.9 4.6 (×1.2) Corn bread 16.7 15.7 (×0.9) Gluten-free breadcrumbs 5.6 8.2 (×1.5) Mixture food additives 4.9 5.2 (×1.1) Spelt cracker 111 115.2 (×1.0) Dog dry food 35.4 28.2 (×0.8) Linseed toast snack 89.4 78.8 (×0.9) Honey corn pops 38.3 24.9 (×0.7) Chips a 17.8 21.1 (×1.2) Dough flour 46.5 40.2 (×0.9) Popcorn 17.4 16.1 (×0.9) Appetizers a 74.8 65.2 (×0.9) Appetizers b 80.0 70.1 (×0.9) Chips b 28.5 23.5 (×0.8) Cookies a 7.9 7.7 (×1.0) Cookies b 9.0 9.2 (×1.0) Corn arepa 7.8 4.0 (×0.5) Fishmeal 19.6 19.5 (×1.0) Infusions a 58.4 80.0 (×1.4) Infusions b 80.0 80.0 (×1.0) Lentil grains 77.7 40.6 (×0.7) Meat preparation 7.5 12.5 (×1.7) Pea flour 7.6 8.3 (×1.1) Plate of lentils 15.7 14.1 (x0.9) Quinoa flour 80.0 80.0 (×1.0) Rice paper 8.5 10.2 (×1.2) Rice preparation 80.0 80.0 (×1.0) Spice a 80.0 80.0 (×1.0) Spice b 78.9 69.8 (×0.5) Spice c 39.3 47.1 (×1.2) Sweet a 80.0 80.0 (×1.0) Sweet b 80.0 80.0 (×1.0) Sweet c 80.0 80.0 (×1.0) Sweet d 80.0 80.0 (×1.0) * Values in brackets are the yield factor compared to that with Cocktail solutions and UGES. Foods 2021,10, 652 9 of 16 Figure 2. Statistical evaluation of the extraction procedure. ( A ) Comparison of agreement between pairs of assays for results of the food samples by sandwich ELISA using different extraction solutions different. Statistical analysis was performed by the Wilcoxon test. The p-values are given in the figure. ( B ) Scatter diagram with regression line and concordance bands for regression line; ( b1 ) representation of the 60% ethanol solution versus Universal Gluten Extraction Solution (UGES); ( b2 ) representation of the Cocktail solution versus UGES. The solid lines represent the regression line drawn for the food samples, and the dashed lines represent the 95% prediction interval. 3.2.2. Recovery of Gliadins in Spiked Food Samples An important approach in the analysis of gluten is to use an extraction system that leads to a recovery close to 100% to ensure that the products suitable for consumption by people with gluten-related disorders are truly free of gluten. To check this, commercial samples labeled as gluten-free (bread, soup, chocolate, and spices) were spiked with a PWG gliadin extract to produce gluten concentrations of 10, 20, or 40 ppm gluten (Figure 3A). Each spiked experiment was performed four times. These samples were extracted with UGES and Cocktail solution, and the extracted amount of gluten was quantified by sandwich ELISA. Foods 2021,10, 652 10 of 16 Figure 3. Analysis of spiked samples by sandwich ELISA after extraction with Universal Gluten Extraction Solution (UGES) and Cocktail solutions. ( A ) Recovery of the different levels of gluten provided in the food samples. Results are expressed as ppm of gluten (mean ± SD) and percentage of gliadin recovery (R). N= number of analyses. ( B ) Comparison of agreement between pairs of assays for results of the food samples by sandwich ELISA with different extraction solutions. ( C ) Scatter diagram with regression line and concordance bands for regression line, indicative of commutability. <LOQ, less than the limit of quantification. The Association of Official Agricultural Chemists (AOAC) has established the optimal recovery percentage for a spiked sample as 80% to 120% [ 51 ]. According to this criterion, the recovery of all samples was satisfactory with both methods. All non-spiked samples were below the LOQ. For the samples spiked with 10, 20, and 40 ppm gluten, the recoveries obtained with UGES were 99% to 120%, 89% to 120%, and 86% to 119%, respectively (Figure 3A). Sandwich ELISA combined with UGES extraction reduced the assay time; hence, the extraction of samples was reduced to 40 min compared with Cocktail solution. Passing–Bablok regression showed a low dispersion in the gluten concentrations obtained by both methods, indicating that the results were commutable (Figure 3C). A high precision and accuracy were evident (r Pearson = 0.9810, C b = 0.9848). Lin’s concordance correlation coefficient (r Lin = 0.9661) was indicative of a substantial agreement strength according to the McBrige scale (>0.95). (Figure 3B). These results indicated that Cocktail solution could be replaced by UGES without any significant difference (p= 0.08330) in the amount of gluten extracted (Figure 3B). Foods 2021,10, 652 11 of 16 3.3. UGES Extraction Effectiveness Assessed by Competitive ELISA During the manufacture of many foods, some processes include thermal and enzymatic reactions that can lead to gluten protein hydrolysis [ 52 – 54 ]. As it is not always possible to know with certainty, prior to the analysis, which food may contain gluten in a hydrolyzed form, it is very important to use a technique that guarantees a correct analysis. Sandwich ELISA uses two antibodies that bind to different sites on the antigen. Thus, gluten peptides with few amino acids and a single antibody binding site cannot be detected with this detection system, and the gluten content would probably be less than the actual value. However, competitive ELISA accurately quantifies of both intact and fragmented gluten because it uses only one antibody and requires only one epitope. To evaluate the versatility of UGES in other types of ELISA formats, we analyzed different hydrolyzed commercial gluten-free foods by spiked them with different known amounts of gluten, followed by quantification using competitive ELISA. These samples were spiked both with PWG gliadin and digested PWG gliadin in different levels (10, 20, and 40 ppm gluten) to simulate the presence of both intact gluten proteins and small peptides with a low number of epitopes. Extraction was performed with UGES, 60% ethanol, and Cocktail solutions. Before the spiking, gluten was undetectable in all samples. The recovery with Cocktail solution was between 178% and 530% in all samples; therefore, there was an overestimation of the actual gluten content (data not shown). As other authors have described [ 54 ], Cocktail solution is not compatible in assays in which the antibody comes into direct contact with the sample extract, because of the interfering effect of reducing agents (2-mercaptoethanol) in the immunoassays [ 34 ]. This component can denature the antibody [ 49 , 55 ]. When we used a 60% ethanol solution, we found interference with the analysis, possibly due to the effect of food matrices in the competitive ELISA. An overestimated recovery of 128% to 150% was obtained in some beers that contained a low concentration of gluten (10 ppm). In contrast, an underestimated recovery of 51% to 67% was obtained in some beers containing a high concentration of gluten (40 ppm). The recovery with 60% ethanol was 0% in the “beer H” sample spiked with 10 ppm. However, when the samples were extracted with UGES, a recovery of 80% to 120% was obtained in all the spiked samples (Figure 4A). The statistical results showed poor agreement strength when comparing the 60% ethanol solution and UGES (r Lin = 0.7655, r Pearson = 0.8371, C b = 0.9145). However, no significant differences were found by Wilcoxon test (p= 0.6355) (Figure 4B). This indicated commutability at low gluten concentrations but non-commutability at higher gluten concentrations (Figure 4C). 3.4. UGES Extraction Using other Immunological Methods 3.4.1. Western Blot Western blot provides a qualitative analysis of proteins and, therefore, is useful for the confirmation of gluten content in foods in a manner that avoids false-positive or false-negative results [ 50 ]. Proteins separated in one-dimensional SDS-PAGE are electrotransferred onto a PVDF membrane where proteins are adsorbed. Specific antibody is added, such as G12 or A1 moAbs. These moAbs are able to recognize with great sensitivity peptides (other than the 33-mer peptide) immunotoxic for patients with CD. The sensitivity and epitope preferences of these antibodies were found to be useful for detecting the potential toxicity of food for patients with gluten-related pathologies [38,39]. Foods 2021,10, 652 12 of 16 α β Ω α Figure 4. Analysis of spiked samples by competitive ELISA after extraction with Universal Gluten Extraction Solution (UGES) and 60% ethanol solutions. ( A ) Recovery of the different levels of gluten provided in the food samples. Results are expressed as ppm of gluten (mean ± SD) and percentage of gliadin recovery (R). N= number of analyses. ( B ) Statistical evaluation of the extraction procedure. ( C ) Scatter diagram with regression line and concordance bands for regression line. <LOQ, less than the limit of quantification; PWG, Prolamin Working Group. To prove the efficiency of the Western blot combined with UGES extraction system, we used PWG gliadin as the reference material. After extraction with 60% ethanol, most of the bands were obtained between 25 and 50 kDa with G12 (Figure 5A(a1)) and A1 (Figure 5A(a2)) moAbs. In contrast, extraction with Cocktail and UGES led to an increase in the number and intensity of bands in comparison with 60% ethanol extraction. The bands between 25 and 50 kDa (corresponding to α -, β -, and Ω -gliadins) and bands at 50–75 kDa (corresponding to α -gliadin) were significantly more intense regardless of the antibody used [ 56 – 58 ]. However, Western blot is not a quantitative method, and transfer to the membrane may vary from lane to lane. Foods 2021,10, 652 13 of 16 Figure 5. Western blot analysis of gluten using different extraction procedures. ( A ) The efficiency in the extractability of Prolamin Working Group (PWG) gliadins with ( a1 ) G12 monoclonal antibody (moAb) and ( a2 ) A1 moAb. ( B ) The efficiency in the extractability of gluten from two commercial samples: (b1) biscuits and (b2) wheat flour (Triticum durum). MW, molecular weight marker. The efficiency in the extractability of gluten in two commercial samples, one of which was heat-processed (biscuits) and the other not (wheat flour), with the different extraction procedures was examined (Figure 5B). The same amount of protein was loaded into well for the different extractions. The biggest differences were for durum wheat flour. The extraction with UGES provided bands of higher molecular weight (50–100 kDa) and lower molecular weight (37–25 kDa) in comparison with 60% ethanol extraction and an increase in the intensity of these bands with respect to the samples extracted with Cocktail solution (Figure 5B). Similar results were obtained for the A1 moAb Western blot. 3.4.2. Lateral Flow Immunochromatographic Assays LFIA is a screening method for the detection of gluten in raw or cooked foods. The assay is rapid, highly sensitive, and feasible in a routine laboratory [ 25 ]. Therefore, we tested the efficiency of UGES combined with LFIA in gluten-free samples spiked with 20 ppm gluten (Table 4). The gluten recovery was close to 100% (92–114%) in all the matrices assayed, irrespective of the extraction procedure carried out according to the type of sample. The data proved that combining the UGES extraction with LFIA is an optimum method for confirming gluten content in different samples. Citation: Segura, V.; Siglez, M.Á.; Ruiz-Carnicer, Á.; Martín-Cabrejas, I.; van der Hofstadt, M.; Mellado, E.; Comino, I.; Sousa, C. A Highly Sensitive Method for the Detection of Hydrolyzed Gluten in Beer Samples Using LFIA. Foods 2023,12, 160. https://doi.org/10.3390/ foods12010160 Academic Editors: Manuel Gomez and Eliana Pereira Received: 18 November 2022 Revised: 22 December 2022 Accepted: 23 December 2022 Published: 28 December 2022 Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). foods Article A Highly Sensitive Method for the Detection of Hydrolyzed Gluten in Beer Samples Using LFIA Verónica Segura 1, Miguel Ángel Siglez 2,Ángela Ruiz-Carnicer 1, Izaskun Martín-Cabrejas 3, María van der Hofstadt 4, Encarnación Mellado 1, Isabel Comino 1and Carolina Sousa 1,* 1 Department of Microbiology and Parasitology, Faculty of Pharmacy, University of Seville, 41012 Seville, Spain 2Biomedal S.L., 41900 Camas, Spain 3Food Technology Department, Veterinary Faculty, Complutense University of Madrid, 28040 Madrid, Spain 4ALINUA, Food and Nutrition Cabinet Health Science Faculty, University of Alicante, UA, 03690 Alicante, Spain *Correspondence: [email protected] Abstract: Most gluten analysis methods have been developed to detect intact gluten, but they have shown limitations in certain foods and beverages in which gluten proteins are hydrolyzed. Methods based on G12/A1 moAbs detect the sequences of gluten immunogenic peptides (GIP), which are the main contributors to the immune response of celiac disease (CD). Immunogenic sequences with tandem epitopes for G12/A1 have been found in beers with <20 mg/kg gluten, which could be consumed by CD patients according to the Codex Alimentarius. Therefore, an accurate method for the estimation of the immunogenicity of a beer is to use two moAbs that can recognize celiac T cell epitopes comprising most of the immunogenic response. Here, a specific and sensitive method based on G12/A1 LFIA was developed to detect GIP in beers labeled gluten-free or with low gluten content, with an LOD of 0.5 mg/kg. A total of 107 beers were analyzed, of those 6.5% showed levels higher than 20 mg/kg gluten and 29% showed levels above the LOD. In addition, G12/A1 LFIA detected gluten in 15 more beer samples than competitive ELISA with another antibody. Despite their labeling, these beers contained GIP which may cause symptoms and/or intestinal damage in CD patients. Keywords: celiac disease; hydrolyzed peptides; gluten immunogenic peptides 1. Introduction Gluten refers to a complex set of storage proteins present in the endosperm of cereals, such as wheat, rye, barley, oat, and their derivatives. It comprises two protein fractions: prolamins (monomeric and alcohol-soluble proteins) and glutelins (polymeric and alcohol-insoluble proteins) [1,2]. Prolamins and glutelins are responsible for the functional properties of cereal dough and are characterized by a high content of proline and which makes them resistant to gastrointestinal digestion and encouraging their deamidation by tissue transglutaminase. Consequently, the consumption of these proteins causes adverse reactions in people suffering from gluten-related disorders (GRDs), such as wheat allergy, non-celiac gluten sensitivity (NCGS), dermatitis herpetiformis, gluten ataxia, and celiac disease (CD) [3–6]. Among GRDs, CD has been extensively studied, and the role of gluten in its pathogenesis has been clearly identified. It is a systemic process that produces chronic enteropathy of the small intestine, and it develops through an inadequate immune response to gluten in genetically predisposed individuals [ 7 – 10 ]. Currently, the only effective treatment is a gluten-free diet (GFD) that relies on the consumption of naturally gluten-free foods, such as animal-based products, fruits, vegetables, legumes, and nuts, as well as gluten-free dietary products that do not contain more than 20 mg/kg of gluten according to the Codex Alimentarius [11], including beer and certain other beverages [12–14]. Foods 2023,12, 160. https://doi.org/10.3390/foods12010160 https://www.mdpi.com/journal/foods Foods 2023,12, 160 2 of 14 Beer is a popular alcoholic beverage that has been consumed worldwide for millennia [ 15 ]. It is traditionally composed of four ingredients: malted barley, water, hops, and yeast. Currently, to meet consumer tastes and needs, new types of beers have become prominent, including nonalcoholic, fruit-flavored, craft, and gluten-free [ 16 ]. However, for patients with CD, the safety of beers derived from gluten-containing grains, particularly wheat and barley, remains a considerable concern [ 17 ]. The amount of gluten produced depends on the various stages of the manufacturing process. During mashing and fermentation, most of the protein fraction is removed [ 18 – 21 ], resulting in short peptide fragments and free amino acids remaining in the beer [ 22 , 23 ]. However, precipitation, lautering, and enzymatic hydrolysis do not abolish the gluten epitopes known to trigger CD; consequently, several studies have reported adverse responses of patients with CD to commercial beers. Indeed, the potential diversity in the generation of sequences, relative abundance, and the extent of the resulting gluten peptides are almost unlimited [24,25]. Enzyme-linked immunosorbent assay (ELISA) is a standard method for the detection and quantification of gluten in food. However, the ability of various ELISA kits to determine the true gluten content is questionable [ 26 , 27 ], as currently available methods have been developed to detect intact gluten but show limitations when quantifying hydrolyzed gluten [ 14 ]. Poor recognition of some of major immunodominant gluten peptides can result in an underestimation of the immunogenicity of some foods, especially hydrolysates, in which these peptides are more readily bioavailable to trigger the disease to a much greater extent than undigested gluten. This uncertainty arises when interpreting ELISA results for the quantification of hydrolyzed gluten in terms of equivalent amounts of intact gluten [28–30]. Therefore, the underestimation or even lack of detection of true immunogenicity involves a significant risk for patients with CD. Immunological methods (ELISAs and lateral flow immunochromatographic assays (LFIAs)) have been developed based on G12 and A1 monoclonal antibodies (moAbs). These moAbs can detect gluten immunogenic peptides (GIP), which are gluten fragments resistant to gastrointestinal digestion and the main factors responsible for the immune response of patients with CD [31]. Based on these moAbs, immunogenic peptides have been found in beers with a gluten content of < 20 mg/kg, which, according to the Codex Alimentarius regulations, could be consumed by patients with CD [ 23 , 28 , 29 , 32 – 37 ]. Further analysis of these beers revealed that the reactive fractions of G12/A1 moAbs displayed GIP sequences using matrix-assisted laser desorption/ionization time-of-flight (MALDI TOF/TOF), and most of the immunogenic sequences identified showed tandem epitopes for G12 and A1 moAbs [ 32 ]. Therefore, the best way to measure the level of immunogenicity of a beer is to use two moAbs that can detect gluten peptides involved in most of the immunogenic response to these proteins. These moAbs must recognize sequences that do not overlap each other. Recent advances in food allergen analysis have emphasized the need for rapid immunoanalytical methods such as LFIAs. These tests are currently the most desired methods, as they offer a low-cost alternative to conventional laboratory tests [ 38 ]. This type of test is a suitable, fast, and easy-to-use option for the analysis of mycotoxins and allergens. Currently, the main application of LFIAs in allergen management is the detection of trace amounts of allergens on surfaces. However, they are also suitable for the detection of trace amounts of allergens in food, as they are faster and easier methods to perform than ELISAs. Additionally, some LFIAs can detect proteins not only in raw products, but also in processed food products. They are commercially available for various analytes (gluten, peanut, hazelnut, soy, milk, fish, mustard, crustacea, peanut, etc.), and have been validated for numerous matrices [38–42]. This study aimed to develop a specific and sensitive method (developed following the Quantifying Uncertainty in Analytical Measurement issued by Eurachem in 1995 and its revised version issued jointly with CITAC in 2020 [ 43 ]) based on G12/A1 LFIA, capable of rapidly and reliably detecting gluten present in beer samples labeled as gluten-free without the need for an extraction process, and with a limit of detection (LOD) of 0.5 mg/kg. Foods 2023,12, 160 3 of 14 2. Materials and Methods 2.1. Reference Standards 33-mer peptide (LQLQPFPQPQLPYPQPQLPYPQPQLPYPQPQPF), Prolamin Working Group (PWG)-gliadin and hydrolyzed wheat prolamin of the wheat variety CUBUS (PTW- Cubus; aromaLAB, Planegg, Germany) were used as reference standards. 2.2. Beer Samples A total of 107 commercial beers labeled gluten-free or low-gluten were obtained, which included some craft beers from different countries of Europe (Belgium, Czech Republic, France, Finland, Germany, Italy, Luxembourg, Poland, Portugal, Spain, and Switzerland) as well as other parts of the world. Descriptive factors such as labeling, yeast style, and wheat inclusion as raw materials were analyzed. Additionally, two high-gluten wheat commercial beers (Germany) were also acquired for use as a positive gluten control. 2.3. Spiked Samples Different beer samples with gluten concentrations below the LOD were spiked with PTW-Cubus (0.5 mg/kg PTW-Cubus) to investigate a possible matrix effect and evaluate the recovery. 2.4. Techniques Employed 2.4.1. Development of a New Method Lateral Flow Immunochromatographic Assays (LFIAs) A G12/A1 moAbs strip was used to develop an analytical method based on the reaction of gluten peptides present in beer samples with colored G12 and A1 moAbs already fixed in the absorption zone. Gluten peptides present in beer samples were bound to the antibody labeled with the detection signal during membrane migration to form a labeled antigen–antibody complex. This complex flowed along the membrane to form a sandwich-like immune complex with the captured moAb immobilized on the test line via the free epitope of the analyte. The result was positive, if a red/pink line appeared in the result zone of the membrane, providing a signal. The absence of this line and therefore a signal below the established limit suggested a negative result. The blue control line was always used as a test control. Analysis of Samples and Data The beer samples and reagents were kept at room temperature (15–25 ◦ C). The samples were then diluted at 1:50, using 20 µ L of the beer sample and 980 µ L of the dilution solution. Subsequently, 100 µ L of the diluted sample were added to the shells of the strips and incubated for 30 min. Finally, the strips were ready for visual analysis. The peak area of the immunochromatographic strip was determined using a GlutenTox ® Reader (Hygiena, Seville, Spain) (Figure 1). GlutenTox ® Reader (Hygiena, Seville, Spain) integrates electronics and LFIA Studio software (Hygiena, Seville, Spain) for data processing [ 44 ]. The peak area values for each sample and the control were obtained using this software. Spreadsheets were used to compile raw data, and the results were considered positive when the reading of the test line was above the negative sample value plus three times the standard deviation. If this criterion was not met, the test would yield a negative result. The values obtained using the G12/A1 LFIA performed in duplicate for each sample. The results were semiquantitative and expressed in the unit of mg/kg according to the minimum quantification dilution detected. Foods 2023,12, 160 4 of 14 Figure 1. Working procedure for gluten analysis in beer samples. GIP, gluten immunogenic peptides; LFIA, lateral flow immunochromatographic assays. 2.4.2. Enzyme-Linked Immunosorbent Assay (ELISA) Gluten content in beer was determined using the ELISA RIDASCREEN ® Gliadin competitive R5 moAb (R7021, R-Biopharm, Darmstadt, Germany). All the samples were analyzed in duplicates according to the manufacturer’s instructions. The obtained data were expressed as the mean and standard deviation. 2.5. Statistical Anlysis Relative affinity curves were obtained by plotting the percent maximum absorbance against the reference standard concentration (ng/mL). The Sigma Plot, version 12.0, Systat Software, Inc., software package (San José, CA, USA) was used to calculate the IC 50 and cross-reactivity (CR) for each standard. The IC 50 was defined as the concentration of the line that reduced the peak absorbance by 50% in the assay. Cross-reactivity (CR) was calculated as (IC 50 of the standard with the highest affinity for the antibody/IC 50 of each tested standard) ×100. 3. Results and Discussion 3.1. Validation of the New Analytical Method The proposed method was evaluated according to the Eurachem guidelines based on studies of sensitivity, specificity, suitability, precision, and robustness [ 43 ]. Similar tests have already been validated for the detection of gluten in food and beverages (with nonhydrolyzed gluten) with different compositions and levels of processing, from raw materials to processed food [39]. In this study, the method was validated for beer hydrolysates. 3.1.1. Immunochromatographic Strip Sensitivity and the LOD The sensitivity was determined based on the characteristics of the immunochromatographic strips and the batch-to-batch effect. It was calculated using spikes from 0 to 25 ng/mL PTW-Cubus in a dilution solution with visual measurements after 30 min. These assays were performed in triplicate (analysis I–II–III). All three assays with spikes of 5–25 ng/mL of PTW-Cubus were visually positive; however, only one of the three assays showed positive results with spikes of 2.5 ng/mL PTW-Cubus, and all the assays showed negative results when containing <2.5 ng/mL PTW-Cubus (Table S1). Based on the sensitivity of PTW-Cubus, the LOD of the method was determined as gluten (mg/kg) = [S × FE × DF × 2]/1000, where S is the calculated sensitivity; DF is the dilution factor; FE is the extraction factor equaling to 1; gliadin-to-gluten ratio is 2. A sensitivity of 5 ng/mL PTW-Cubus was established for reading at 30 min and test dilution of 1:50, resulting in an LOD of 0.5 mg/kg for gluten. 3.1.2. Strip Reader Detection Capacity To determine the detection capacity of the reader used in the test, the measurements obtained in the analysis of the beer samples (<LOD) were compared with the measurements of the same beer samples spiked with 0.5 mg/kg PTW-Cubus. Foods 2023,12, 160 5 of 14 First, the average peak area value offered by the reader was established, when five beer samples with the gluten content lower than the LOD were analyzed five times each. In most of the validation protocols reported in the literature, the verification of the method [ 45 ] was calculated by executing at least five determinations per concentration and a deviation below 15% or 20% of the expected value [ 46 ] as the acceptance criterion (excluding negative samples) [ 47 ]. The results showed a negative visual interpretation in all strips. However, the reader detected a signal in 8 of the 25 samples analyzed, with a mean peak area of 35.1 ±9.7 (Table S2). Based on these results, a minimum peak area of approximately 70 was established to consider a sample as positive. This was calculated by adding three times the deviation to the value of the mean peak area (35.1) [43,48]. Furthermore, the validity of this gluten-detection method was determined using 0.5 mg/kg of PTW-Cubus spiked in four samples (<LOD) five times each. All the strips were positive on visual interpretation, and the results provided by the strip reader established a mean peak area of 256.8 ±26.5 in these samples (Table 1). Table 1. Peak area results of beer samples (gluten < LOD) spiked with 0.5 mg/kg PTW-Cubus, obtained by the strip reader. Beer Samples Spiked Sample Peak Area Beer I 1 282.0 2 278.5 3 270.2 4 227.0 5 274.6 Beer II 6 250.6 7 251.7 8 263.6 9 273.9 10 309.0 Beer III 11 288.0 12 232.3 13 279.6 14 214.3 15 228.1 Beer IV 16 233.3 17 279.2 18 244.8 19 234.8 20 220.8 Once the average results of the peak area and standard deviations were obtained using the strip reader, the range of reading variation was calculated as a function of the coefficient of variation ( σ ). Furthermore, confidence intervals around the mean were determined, with confidence levels of 99.7% (peak area: 283.3–230.3), 95% (peak area: 309.8–203.9), and 68% (peak area: 336.2–177.7). Finally, the range ± 2 σ was established as the criterion for considering a result as valid, since a 95% coverage was observed in that range (Figure 2) [49]. Foods 2023,12, 160 6 of 14 σ σ σ Figure 2. The normal distribution and range of variation in beer samples (gluten < LOD) spiked with 0.5 mg/kg PTW-Cubus. σ, coefficient of variation. 3.1.3. Accuracy and Precision According to the Eurachem guide, accuracy and precision indicate the ability of the method to obtain results within the established range of interest. According to these guidelines, the results must be obtained from an average of 6 to 15 replicates of each material, with the same equipment, analyst, and laboratory and in a short period [43]. In this study, accuracy was calculated by taking 18 measurements of 2 beer samples (<LOD) spiked with 0.5 mg/kg PTW-Cubus. These 18 measurements were performed on the same day and from 3 different extractions, each analyzed in triplicates (Table 2). Meanwhile, precision was achieved by performing 36 measurements of 2 beer samples (<LOD) spiked with 0.5 mg/kg PTW-Cubus. These 36 measurements were carried out on 2 different days, and each measurement was analyzed in triplicates. Table 2. Determination of accuracy and precision in beer samples (gluten < LOD) spiked with 0.5 mg/kg PTW-Cubus. Day Beer Extraction Analysis Peak Area Inside the Range Accuracy (%) Precision (%) Day 1 Beer I Beer Ia Beer Ia 277.5 yes 88.9% 94.5% Beer Ia 305.1 yes Beer Ia 290.4 yes Beer Ib Beer Ib 196.9 no Beer Ib 278.1 yes Beer Ib 267.3 yes Beer Ic Beer Ic 186.9 no Beer Ic 279.0 yes Beer Ic 217.9 yes Beer II Beer IIa Beer IIa 255.9 yes Beer IIa 259.2 yes Beer IIa 268.9 yes Beer IIb Beer IIb 248.2 yes Beer IIb 297.3 yes Beer IIb 265.4 yes Beer IIc Beer IIc 235.7 yes Beer IIc 221.7 yes Beer IIc 270.2 yes Foods 2023,12, 160 7 of 14 Table 2. Cont. Day Beer Extraction Analysis Peak Area Inside the Range Accuracy (%) Precision (%) Day 2 Beer I Beer Ia Beer Ia 282.0 yes 100% Beer Ia 278.5 Yes Beer Ia 270.2 Yes Beer Ib Beer Ib 248.3 Yes Beer Ib 303.7 yes Beer Ib 289.5 yes Beer Ic Beer Ic 301.9 yes Beer Ic 271.5 yes Beer Ic 226.6 yes Beer II Beer IIa Beer IIa 288.0 yes Beer IIa 232.3 yes Beer IIa 279.6 yes Beer IIb Beer IIb 248.2 yes Beer IIb 295.3 yes Beer IIb 268.7 yes Beer IIc Beer IIc 208.9 yes Beer IIc 283.3 yes Beer IIc 229.4 yes All the strips were positive on visual interpretation. Of the 18 replicates analyzed to calculate the accuracy, 16 were within the range (peak area: 203.9–309.8), indicating an accuracy of 88.9% on day 1 and 100% on day 2. In addition, 34 of the 36 replicates used to calculate the precision were within the established range (peak area: 203.9–309.8), indicating a precision of 94.5% (Table 2). 3.2. Determination of the Relative Affinity from Different Reference Standards To confirm the detection capacity of G12/A1 LFIA against the most widely used reference standards in immunological assays, we tested 33-mer peptide, PWG-gliadin, and PTW-Cubus (hydrolyzed wheat prolamin). The affinity of moAbs for the antigens was quantified by calculating the concentration of the antigen, resulting in a 50% reduction in the peak signal in the LFIA (IC 50 ) and CR. The results were compared with those obtained from R5 competitive ELISA. PWG-gliadin showed the highest reactivity with G12/A1 moAbs (IC 50 , 2.91; CR, 100%). In addition, the 33-mer peptide (IC 50 , 5.95; CR, 48.91%) and gliadin PTW-Cubus (IC 50 , 59.53; CR, 30.54%) were also detected with high sensitivity using these moAbs (Figure 3A,C). The results obtained using the R5 competitive ELISA demonstrated that this antibody was able to detect PWG-gliadin and gliadin PTW-Cubus, although with lower affinity than G12/A1 moAb, and the 33-mer peptide was not detected using moAb R5 (Figure 3B). Different research groups have shown that the α -gliadin 33-mer is poorly identified with R5 competitive ELISA [ 23 , 50 , 51 ]. By contrast, analysis of T-cell reactivity and enzyme protein detoxification showed that the signals of G12 and A1 moAbs correlate with the potential toxicity of the sample for patients with CD [ 23 , 32 , 52 – 55 ]. The recognition profiles of G12 and A1 moAbs have been extensively characterized [ 23 , 52 , 53 ]. Moreno et al. [ 56 ] showed that most GIP (responsible for 80–95% of celiac T-cell immunoreactivity) react with G12 moAb. Foods 2023,12, 160 8 of 14 α Figure 3. Relative affinity of G12/A1 LFIA and R5 competitive ELISA for different reference standards. ( A ) G12/A1 LFIA. ( B ) R5 competitive ELISA. ( C ) Standard reference curves. IC 50 and CR were obtained for G12/A1 LFIA and R5 competitive ELISA. The IC 50 is defined as the concentration of the line that reduces the peak absorbance by 50% in the assay. These assays were performed in duplicates. In the figure the gray points correspond to PWG-gliadin; the green points, PTW-Cubus; and the red dots, 33-mer peptide. LFIA, lateral flow immunochromatographic assays; N.A., not applicable. 3.3. Gluten Analysis in Beer Gluten analysis was carried out in 107 beers, with 88 labeled gluten-free, 12 low in gluten or with traces of gluten (non-specific labeled), and 7 not labeled for gluten or artisanal. Additionally, we tested two high-gluten beer samples as positive controls. The beer samples were grouped according to their cereal content in their labeling; based on this classification, 65% (70/107) contained cereals with gluten, 16.8% (18/107) contained a mixture of cereals with and without gluten, and only 4.7% (5/107) of the beer samples contained exclusively gluten-free cereals (maize, millet, buckwheat, and rice). According to the type of yeast used in the preparation, 56.8% (49/107) were made with lager yeast, and 43% (46/107) were prepared with ale yeast (Table 3). Table 3. Composition and characteristics of beer samples. Number of Beers Positive for Gluten % (Positive/Number of Beer) % (Positive/Total Beer Samples) Total beer samples 107 Yeast style Lager 49 10 20 (10/49) 9 (10/107) Ale 46 16 35 (16/46) 15 (16/107) Unknown 12 5 42 (5/12) 5 (5/107) Label Low in gluten or with traces of gluten 12 4 33 (4/12) 4 (4/107) Gluten-free 88 26 30 (26/88) 24 (26/107) Unknown 7 1 14 (1/7) 1 (1/107) Ingredients Contains cereals with gluten 70 24 34 (24/70) 22 (24/107) Barley 59 16 27 (16/59) 15 (16/107) Wheat 1 0 0 (0/1) 0 (0/107) Oats 1 1 100 (1/1) 1 (1/107) Barley and wheat 4 3 75 (3/4) 3 (3/107) Barley and rye 1 0 0 (0/1) 0 (0/107) Barley and oats 2 2 100 (2/2) 2 (2/107) Wheat, barley, and rye 1 0 0 (0/1) 0 (0/107) Wheat, barley, and oats 1 1 100 (1/1) 1 (1/107) Foods 2023,12, 160 9 of 14 Table 3. Cont. Number of Beers Positive for Gluten % (Positive/Number of Beer) % (Positive/Total Beer Samples) Unknown 15 6 40 (6/15) 6 (6/107) Contains cereals with gluten and gluten-free 18 2 11.1 (2/18) 3 (3/107) Barley and maize 7 1 14 (1/7) 1 (1/107) Barley and millet 2 0 0 (0/2) 0 (0/107) Wheat and millet 1 0 0 (0/1) 0 (0/107) Barley, wheat, and quinoa 1 0 0 (0/1) 0 (0/107) Barley and rice 3 0 0 (0/3) 0 (0/107) Wheat, barley, rye, and rice 1 0 0 (0/1) 0 (0/107) Barley, oats, and maize 1 0 0 (0/1) 0 (0/107) Barley, rice, and maize 1 0 0 (0/1) 0 (0/107) Barley, rice, and quinoa 1 1 100 (1/1) 1 (1/107) 1 Contains only gluten-free cereals 5 2 40 (2/5) 2 (2/107) Maize 2 0 0 (0/2) 0 (0/107) Millet 2 1 50 (1/2) 1 (1/107) Buckwheat and rice 1 1 100 (1/1) 1 (1/107) Country of origin Argentina 1 0 0 (0/1) 0 (0/107) Belgium 10 1 10 (1/10) 1 (1/107) Czech Republic 1 0 0 (0/1) 0 (0/107) England 1 0 0 (0/1) 0 (0/107) Finland 15 8 53 (8/15) 7 (8/107) France 3 1 33 (1/3) 1 (1/107) Germany 3 1 33 (1/3) 1 (1/107) Italy 5 0 0 (0/5) 0 (0/107) Luxembourg 2 2 100 (2/2) 2 (2/107) Mexico 3 0 0 (0/3) 0 (0/107) Poland 4 0 0 (0/4) 0 (0/107) Portugal 1 0 0 (0/1) 0 (0/107) Scotland 2 1 50 (1/2) 1 (1/107) Spain 43 16 37 (16/43) 15 (16/107) Switzerland 13 1 8 (1/13) 8 (1/107) The gluten content was below the LOD (0.5 mg/kg gluten) in 71% (76/107) of the samples analyzed using G12/A1 LFIA. Our results showed that 29% (31/107) of the analyzed samples were positive in the range of 0.5–256 mg/kg of gluten (Figure 4), although they were labeled as gluten-free or low-gluten. In addition, 6.5% (7/107) contained levels higher than 20 mg/kg of gluten. This could be related to the brewing process, ingredients, the yeast used, and other factors. The addition or absence of malted cereal and the use of clarification processes, such as filtration, centrifugation, or the cold step, can influence the gluten content in beer [ 19 ]. Therefore, the inclusion of additional steps in the production process and/or the use of enzymes to hydrolyze gluten (e.g., PEP) could modify the amount of gluten present [ 34 ]. It should be noted that 74.2% (23/31) of the analyzed beer samples contained gluten-containing cereals on their labels. Furthermore, the number of beers with a positive result for gluten and ale yeast was higher than the number of those containing lager yeast (51.6% (16/31) and 32.3% (10/31), respectively). The influence of yeast fermentation on gluten content in the final product was previously noted by Tanner et al. [ 57 ] and Fernandez-Gil et al. [ 30 ], who highlighted the importance of selecting lager yeasts or excluding cereals with gluten in the brewing process to reduce the gluten content of the final product. Additionally, the results obtained with G12/A1 LFIA were compared with those of R5 competitive ELISA, which revealed that 85% (91/107) of the samples were below the limit of quantification (LOQ) (<10 mg/kg of gluten) and 15% (16/107) showed positive results ranging from 10.8 to 63.1 mg/kg of gluten (Figure 4). This was a predictable result, as G12/A1 LFIA had a lower LOD than R5 competitive ELISA. Therefore, G12/A1 LFIA detected gluten in 15 more beer samples (beers Q to EE), 73% (11/15) of which contained gluten-containing cereals on their labels. 119 RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 120 CAPÍTULO 7 7.1. Estudio longitudinal prospectivo: uso de péptidos inmunogénicos del gluten en heces para el seguimiento de niños diagnosticados de enfermedad celíaca durante la transición a una dieta sin gluten. El único tratamiento actual de la EC es una DSG durante toda la vida del paciente con el objetivo de aliviar los síntomas, lograr la recuperación de la mucosa intestinal y prevenir complicaciones asociadas. Una vez confirmado el diagnóstico de EC y establecida la DSG, se debe realizar un seguimiento clínico-analítico periódico del paciente con el fin de detectar posibles complicaciones y patologías asociadas (Bascuñán et al., 2018; Lebwohl et al., 2018; Singh et al., 2018; Caio et al., 2019; Lindfors et al., 2019; Cabanillas, 2020; Therrien et al., 2020). Sin embargo, no se disponen de herramientas de monitorización de la adherencia a la DSG que sean objetivas, fiables y no invasivas. Se ha demostrado que la sintomatología, serología o los cuestionarios de adherencia no son eficaces para asegurar la ausencia de lesión histológica y evitar el desarrollo de complicaciones a largo plazo en estos pacientes (Silvester et al., 2017; Al-Toma et al., 2019; Ruiz-Carnicer et al., 2020; Wieser et al., 2020; Fernández-Bañares et al., 2021; Galli et al., 2021; Tye-Die, 2022; Garzón et al., 2023). Recientemente, se ha confirmado que el análisis cuantitativo de GIP en muestras de heces de pacientes adultos es un método preciso y no invasivo que permite una evaluación directa del consumo del gluten después de su ingesta (Comino et al., 2012b, 2016). No obstante, no se ha determinado la utilidad de dicho biomarcador en la población pediátrica. Basándose en estos antecedentes, en el capítulo 3 de esta Tesis Doctoral (Comino et al., 2019) se evaluó el contenido de GIP en heces como un marcador no invasivo de monitorización de la DSG en pacientes pediátricos con EC de reciente diagnóstico, y se contrastaron con los métodos tradicionales utilizados en la práctica clínica. Para ello, se realizó un estudio observacional prospectivo multicéntrico donde se incluyeron a 64 niños (21 varones y 43 mujeres) con una mediana de edad de 4 años (rango: 1,5-9 años). Los niveles de GIP se cuantificaron utilizando el kit iVYLISA GIP Stool de la línea iVYDAL In Vitro Diagnostics® (Biomedal S.L., Sevilla). Para ello, las muestras de heces fueron obtenidas de pacientes pediátricos en dieta con gluten (DCG) en el momento de diagnóstico (visita basal) y durante las 3 visitas siguientes (visitas a los 6, 12 y 24 meses) a DSG. Además, en la visita basal, a los pacientes se les preguntaba por la sintomatología clínica, incluyendo síntomas gastrointestinales típicos, así como también RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 121 por síntomas atípicos. Por otro lado, se compararon los resultados obtenidos mediante los GIP con los métodos serológicos anti-tTG y anti-DGP, y cuestionarios dietéticos en las distintas visitas de seguimiento a los 0, 6, 12 y 24 meses (Figura 1). La participación de los pacientes en el estudio disminuyó progresivamente con el tiempo, alcanzando una tasa del 94% a los 6 meses, del 78% a los 12 meses y del 55% a los 24 meses. Las razones más comunes de pérdida de seguimiento fueron el cambio de residencia fuera del área de estudio, la ausencia en las visitas de seguimiento y la omisión en la recolección de muestras. Aunque nuestro estudio reveló una tasa de seguimiento baja, esta fue comparable a la observada en otros estudios diferentes llevados a cabo en centros pediátricos (Mozer-Glassberg et al., 2011; Herman et al., 2012; Mouslih et al., 2023). Figura 1. Diseño del estudio clínico prospectivo. Los participantes incluidos en el estudio fueron instruidos para recolectar entre 2 y 4 g de heces. El día de la visita se les realizó una extracción de sangre para medir los marcadores serológico y un cuestionario alimentario estructurado para registrar los alimentos consumidos durante los 4 días previos a la recolección de las muestras. Además, se les preguntó por la sintomatología durante ese periodo. EC, enfermedad celíaca; GIP, péptidos inmunogénicos del gluten. En este estudio se demostró que antes de comenzar la DSG, el 97% de los pacientes pediátricos mostraron niveles detectables de GIP en sus muestras fecales (Figura 2). Además, el 83% de los participantes reportaron tener síntomas clásicos antes del diagnóstico de EC. Después del diagnóstico y de instaurar la DSG, el 23% de los pacientes presentaron niveles detectables de GIP en al menos una visita, mientras que el 20% no cumplió según el cuestionario dietético. Durante el estudio, observamos que el porcentaje de transgresiones fue del 13% (7 de RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 122 54 pacientes) a los 6 meses, aumentó al 18% (7 de 39 pacientes) a los 12 meses y alcanzó el 25% (6 de 24 pacientes) a los 24 meses, como se muestra en la Figura 2. Figura 2. Determinación del contenido de los GIP en pacientes con EC pediátricos en cada visita de seguimiento (0, 6, 12 y 24 meses). GIP, péptidos inmunogénicos del gluten; LOQ, límite de cuantificación. Además, la tasa de GIP positivo en heces durante el seguimiento aumentó con la edad (P = 0,041), siendo detectadas en el 6% de los niños con EC antes de los 2 años, e incremento al 24% y 35% en aquellos diagnosticados a los 2-6 años y a una edad mayor, 7-18 años, respectivamente (Figura 3). Este incremento puede atribuirse posiblemente a comportamientos más arriesgados, especialmente durante la adolescencia. Este período, conocido por su desafío en la adherencia a la DSG debido a la estigmatización y la mayor exposición al gluten fuera del hogar, se identificó como especialmente crítico. Resultados similares han sido reportados en un estudio retrospectivo realizado por Mouslih et al. (2023), en el que se correlacionó la adherencia a la DSG según cuestionarios con la edad. Los hallazgos de Simón et al. (2023) sugieren que, en general, los adolescentes presentan una menor adherencia a la DSG, mientras que los niños menores de 12 años muestran un mejor cumplimiento con esta dieta. Además, se observó que la mayoría de las ocasiones en las que no se cumplía la DSG eran atribuibles a causas no intencionadas, especialmente en el caso de los niños. En contraste, los adolescentes mostraban una mayor propensión a no seguir la DSG de manera intencional, debido a las dificultades para implementarla en su entorno social o para identificar alimentos libres de gluten (Pohoreski et al., 2023). RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 123 Figura 3. Adherencia a la DSG según el contenido de GIP en los diferentes grupos etario de pacientes con EC pediátricos. GIP, péptidos inmunogénicos del gluten. Nuestros resultados mostraron diferencias entre la detección de GIP y la evaluación de cuestionarios dietéticos. En general, el 11% de todos los pacientes con EC no cumplían con la DSG según ambos métodos, mientras que el 48% estaban libres de gluten según ambos métodos (κ = 0,5). Sin embargo, el 39% mostró discrepancias entre los resultados de GIP y de los cuestionarios. Es relevante destacar que el 46% de los participantes no adherentes a la DSG tenían dos o más muestras positivas para GIP a DSG, lo que sugiere una exposición crónica o repetida al gluten. Nuestros hallazgos sobre la ingesta de gluten en pacientes con EC a DSG se respaldan con estudios posteriores que han identificado altas tasas de no adherencia a la DSG en pacientes adultos y adolescentes. Además, son comparables con estudios previos que han identificado altas tasas de no adherencia a la DSG en pacientes pediátricos y adultos, documentando cifras de hasta el 69% y el 64%, respectivamente. Estas evaluaciones han utilizado diversos métodos, como cuestionarios de adherencia, registros dietéticos y pruebas serológicas, así como la determinación de GIP en heces y orina (Segura et al., 2021; Wieser et al., 2021) (Tabla 1). Tabla 1. Estudios que han utilizado la determinación de GIP en muestras de heces y/u orina para el seguimiento de una DSG. DSG, dieta sin gluten; GIP, péptidos inmunogénicos del gluten. (modicado de Coto et al., 2021). 6,2 23,5 34,8 93,8 75 65,2 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% < 2 años 2-6 años 7-18 años Porcentaje GIP evolutivo Negativo Positivo RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 124 Tipo de muestras Población de estudio Diseño Referencias Heces Niños Estudio casos-controles Comino et al., 2012b Estudio de cohortes Gerasimidis et al., 2018 Estudio prospectivo Comino et al., 2019 Estudio transversal Roca et al., 2019 Estudio prospectivo Coronel-Rodríguez et al., 2019 Revisión sistemática Myleus et al., 2020 Estudio prospectivo Roca et al., 2021 Estudio descriptivo observacional Fernández-Miaja et al., 2021 Niños y Adultos Estudio prospectivo Comino et al., 2016 Estudio transversal Porcelli et al., 2020 Adultos Estudio prospectivo observacional Laserna-Mendieta et al., 2020 Estudio prospectivo Fernández-Bañares et al., 2021 Estudio transversal Skodje et al., 2022 Estudio prospectivo observacional Stefanolo et al., 2023 Estudio prospectivo Stefanolo et al., 2024 Orina Niños Estudio transversal Mello et al., 2024 Estudio observacional prospectivo Seetharaman et al., 2023 Niños y Adultos Estudio controlado Soler et al., 2016 Estudio controlado aleatorizado Moreno et al., 2017 Adultos Estudio transversal Peláez et al., 2020 Estudio prospectivo Ruiz-Carnicer et al., 2020 Estudio prospectivo Moreno et al., 2021 Estudio transversal Coto et al., 2021 Estudio transversal Palanski et al., 2022 Estudio prospectivo cruzado Burger et al., 2022 Estudio prospectivo Garzón-Benavides et al., 2023 Estudio prospectivo Segura et al., 2023 Estudio transversal Lombardo et al., 2023 Heces y orinas Niños Estudio transversal Horton et al., 2023 Estudio prospectivo Guz-Mark et al., 2023 Niños y Adultos Metaanálisis Syage et al., 2018 Adultos Estudio prospectivo Costa et al., 2019 Estudio prospectivo Silvester et al., 2020a, 2020b Estudio prospectivo Stefanolo et al., 2021 Estudio prospectivo Rusell et al., 2024 Al evaluar la respuesta clínica como un indicador de la adherencia a la DSG se observó una rápida mejoría sintomática inicial. Sin embargo, a medida que pasaba el tiempo desde el RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 125 diagnóstico la proporción de muestras fecales positivas para GIP aumentaban sugiriendo una posible tendencia a relajar la adherencia a la DSG. Un estudio reciente apoya la noción de que la mayoría de los niños experimentan una resolución de sus síntomas clínicos en un lapso de 3 meses (Mouslih et al., 2023). No obstante, es importante señalar que la exposición al gluten durante la DSG puede no necesariamente provocar síntomas (Dewar et al., 2012; Baggus et al., 2019; Penny et al., 2020). Respecto a la correlación entre los niveles de GIP en heces y los anticuerpos séricos se observó que los niveles de anticuerpos séricos anti-tTG y anti-DGP disminuyeron durante el seguimiento, mientras que el porcentaje de muestras fecales positivas para GIP mostraron una tendencia ascendente (Figura 4). Sin embargo, no se encontró correlación estadísticamente significativa entre los valores absolutos de ambos tipos de anticuerpos y los niveles de GIP en heces, ni siquiera al utilizar términos dicotómicos. La disminución absoluta en el título de anticuerpos tTG después de 6 y 12 meses a DSG fue sustancial, siendo más pronunciada en el grupo de pacientes adherentes a la DSG. Esto demuestra que la serología una vez negativizada no vuelve a positivizarse en una gran proporción de pacientes que cometen transgresiones. Por tanto, la normalización de los marcadores serológicos no refleja el cumplimiento estricto de una DSG y es un mal predictor de las transgresiones dietéticas (Husby et al., 2019). Como han señalado otros autores, los marcadores serológicos muestran una eficacia limitada para monitorizar la adherencia a la DSG de los pacientes adultos o pediátricos con EC (Silvester et al., 2017; Comino et al., 2016; Ruiz-Carnicer et al., 2020; Garzón et al., 2023) (Tabla 1). RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 126 Figura 4. Comparativa de los resultados obtenidos con los marcadores serológicos en las visitas de seguimiento realizada a los pacientes pediátricos con EC en relación con los resultados de GIP positivo, en muestras fecales, obtenidos en cada una de estas visitas. anti-tTG, antitransglutaminasa tisular; anti-DGP, anti-péptidos de gliadina desaminada; GIP, péptidos inmunogénicos del gluten. En este estudio, se determinó, también, la ingesta de gluten en pacientes con EC que estaban siguiendo una DSG mediante la cuantificación de GIP en muestras de heces. Este análisis se realizó utilizando un modelo matemático desarrollado por Syage et al. (2018), que fue aplicado en un metaanálisis de datos, permitiendo así una estimación precisa de la ingesta inadvertida de gluten en individuos con EC que estaban siguiendo una DSG. Durante el período de seguimiento, se observó una tendencia al aumento en la ingesta de gluten a medida que pasaban los meses, con niveles de 144/99 mg/día a los 6 meses, 452/105 mg/día a los 12 meses y 606/117 mg/día a los 24 meses después del diagnóstico. Este incremento posiblemente estaba relacionado con una relajación en la adherencia a la DSG por parte de los pacientes, lo que llevaba a un consumo más elevado de alimentos con gluten. Es importante destacar que la ingesta estimada de gluten en los pacientes que seguían una DSG (mediana de 0,104 gramos por día) fue más de 30 veces menor que la cantidad típica utilizada en las pruebas de provocación con gluten (3 a 10 g/día) (Bruins et al., 2013). Esto sugiere que la ingestión de 0,1 a 0,5 g de gluten puede no generar síntomas en una proporción significativa de pacientes no adherentes. Aunque se ha descrito en la literatura que una ingesta diaria de menos de 50 mg parece ser segura para la mayoría de los pacientes con EC (Catassi et al., 2007), otros autores han disminuido este nivel a 30 mg para evitar anomalías en la mucosa intestinal (Collin et al., 2004). Dado que existe una gran diversidad en la sensibilidad al gluten entre los individuos, establecer un umbral inofensivo de ingesta diaria de gluten para la población con EC sigue siendo una tarea problemática (Catassi et al., 2007; Hall et al., 2013). Estudios realizados en adultos también demostraron resultados similares (Silvester et al., 2020a, 2020b). Por otra parte, conforme a las recomendaciones de la Sociedad Europea de Gastroenterología, Hepatología y Nutrición Pediátrica (ESPGHAN) (Husby et al., 2020), se sugiere que, en pacientes pediátricos y adolescentes jóvenes, la primera consulta de seguimiento se lleve a cabo entre 3 y 6 meses después del diagnóstico de EC, y que se repita cada 6 meses hasta que los niveles de anticuerpos anti-tTG se normalicen. Posteriormente, las visitas pueden espaciarse cada 12 o 24 meses. Sin embargo, este estudio ha evidenciado que, a pesar de seguir una DSG, los pacientes han estado expuestos al gluten. En este sentido, el análisis secuencial de GIP en heces no solo ha confirmado directamente la ingestión de gluten días antes del RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 127 diagnóstico de la EC, sino también ha documentado una disminución significativa en la exposición al gluten después del diagnóstico y la implementación de la dieta. Por consiguiente, la cuantificación de GIP en las heces como una herramienta de evaluación de la adherencia a la DSG puede mejorar la gestión clínica y disminuir la necesidad de pruebas invasivas adicionales en el seguimiento de los pacientes pediátricos con EC. Este hallazgo sugiere que el tratamiento implica restricciones y cambios en la vida social y hábitos de los pacientes, lo que podría afectar la adherencia a la DSG y llevar a transgresiones o incluso al abandono de la misma (Fernández- Miaja et al., 2020). En conclusión, los resultados descritos en este trabajo demuestran que el tratamiento de la DSG conlleva restricciones y cambios en la vida social y hábitos de los pacientes, lo que podría comprometer la adherencia a este tipo de dieta y resultar en transgresiones incluso en el abandono de la misma. 7.2. Evaluación de la utilidad de un inmunoensayo automatizable para el monitoreo de la enfermedad celíaca mediante la cuantificación de péptidos inmunogénicos de gluten en la orina. Los avances en la detección de biomarcadores a través de muestras de orina han transformado la monitorización de las enfermedades, proporcionado una herramienta invaluable para la investigación médica y el diagnóstico clínico (Broza et al., 2019). La orina, debido a su recolección no invasiva, permite la posibilidad de obtener grandes cantidades de y ser recolectada repetidamente durante periodos prolongados de tiempo, lo que ha convertido este tipo de muestra en un recurso fundamental en el campo científico y clínico (Adachi et al., 2006). Sin embargo, a pesar de su utilidad, las muestras de orina presentan desafíos significativos debido a su alta heterogeneidad y bajo contenido proteico. Estas características hacen que el desarrollo de inmunoensayos para la detección de biomarcadores sea una tarea compleja. En este contexto, la detección de GIP en la orina de pacientes con EC ha sido objeto de investigación durante muchos años. Hace aproximadamente dos décadas, se logró demostrar la presencia de GIP en muestras de orina de pacientes con EC utilizando técnicas de cromatografía. Posteriormente, este descubrimiento fue confirmado mediante métodos basados en AcMos (Reichelt et al., 1998; Diaz-Amigo y Popping, 2013; Soler et al., 2016; Moreno et al., 2017). Entre estos destacan el G12 y el A1 (Morón et al., 2008a, 2008b), que han demostrado su capacidad para detectar los GIP, fragmentos de gluten resistentes a la digestión gastrointestinal y que desencadenan la respuesta inmune en los pacientes con EC (Cebolla et al., RESUMEN GLOBAL DE LOS RESULTADOS Y DISCUSIÓN 128 2018). Estudios posteriores han revelado una estrecha relación entre la estimulación de células T en pacientes con EC y la reactividad de estos AcMos frente a distintos tipos de cereales (Comino et al., 2011, 2012a; Real et al., 2012, 2014). Además, se ha observado que el AcMo G12 es capaz de unirse a la mayoría de los GIP presentes en las cervezas y la gliadina hidrolizada, como se demostró en un estudio de inmunocaptura (Moreno et al., 2016). Estos hallazgos respaldan la eficacia de los AcMos G12 y A1 en la detección de GIP en muestras de orina y otras matrices, lo cual proporciona una herramienta valiosa para el diagnóstico y seguimiento de la EC. Se han desarrollado inmunoensayos basados en los AcMos G12 y A1 para controlar la adherencia a la dieta (Comino et al., 2012b; Moreno et al., 2017). Actualmente, el método que resultó ser más eficaz fue el LFIA, conocido como test rápido. El tratamiento de la orina con una solución de acondicionamiento permitió detectar los GIP entre 15-30 minutos después de realizar el ensayo, dependiendo de la concentración de GIP de la muestra, con una sensibilidad y especificidad diagnóstica del 97% y 100%, respectivamente, tanto en pacientes con EC como en controles sanos (Moreno et al., 2017; Roca et al., 2019). Aunque el resultado proporcionado por este test es de tipo cualitativo (positivo/negativo), el uso de un lector de LFIA permitió hacer una cuantificación de los GIP (Moreno et al., 2017). Basándose en estos antecedentes, en el capítulo 4 de esta Tesis Doctoral (Segura et al., 2023) se ha desarrollado un método analítico cuantitativo, automatizable y de altamente sensible para la detección de los GIP y su posible aplicación en laboratorios clínicos. El estudio se enfocó en la optimización, validación y aplicación de un método ELISA altamente sensible y automatizable para la detección cuantitativa de GIP en muestras de orina. El desarrollo de un prototipo de método ELISA destinado a detectar GIP en orina, se realizó con el objetivo de ofrecer una alternativa polivalente y mejorada en comparación con el LFIA. Para ello, se utilizó un ELISA tipo sándwich previamente desarrollado, en nuestro grupo de investigación, que empleaba el AcMo G12 y que se utiliza para analizar GIP en heces (Comino et al., 2016; Comino et al., 2019). Con el objetivo de aumentar la sensibilidad se establecieron nuevas condiciones de trabajo y aplicaciones, y se usó el péptido de 33-mer como estándar de referencia. Este péptido es reconocido por las células T de los pacientes con EC y se ha demostrado que es uno de los péptidos más inmunogénicos de las α-gliadinas del trigo (Cebolla et al., 2018; Lindfors et al., 2019). Se estudiaron varios parámetros, como la solución de dilución del ensayo, el anticuerpo monoclonal de recubrimiento, el estándar de la curva y la concentración de captura del anticuerpo, con el fin de mejorar las condiciones y metodología del ELISA en la detección de GIP en la matriz de orina.