Clinical performance evaluation of the Idylla™ EGFR Mutation Test on formalin-fixed paraffin-embedded tissue of non-small cell lung cancer

RESEARCH ARTICLE Open Access Clinical pe o mance e alua ion o he Idylla™EGFR Mu a ion Tes on o malin- ixed pa a in-embedded issue o non- small cell lung cance Me cedes Delgado-Ga cía 1 , Bi gi Weynand 2 , Lou des Gómez-Izquie do 1 , Ma ía José He nández 1 , Ángela Ma ía Blanco 1 , Ma Va ela 3 , Xa ie Ma ias-Guiu 3,4 , E nes Nadal 5 , Bélgica Má quez-Lobo 6 , Ana Ala cão 7 , En ique de Ála a 1 and Michele Biscuola 1* Abs ac Backg ound: De ec ion o epide mal g ow h ac o ecep o (EGFR) mu a ions in exons 18–21 is ecommended in all pa ien s wi h ad anced Non-small-cell lung ca cinoma due o he demons a ed e iciency o he s anda d he apy wi h y osine kinase inhibi o s in EGFR-mu a ed pa ien s. The e o e, choosing a sui able echnique o es EGFR mu a ional s a us is c ucial o wa an a alid esul in a sho u na ound ime using he lowes possible amoun o issue ma e ial. The Idylla™EGFR Mu a ion Tes is a simple, as and eliable me hod designed o he de ec ion o EGFR mu a ions om o malin- ixed pa a in-embedded samples. The aim o his s udy was he Clinical Pe o mace E alua ion o he Idylla™EGFR Mu a ion Tes on he Idylla™ Sys em. Me hods: EGFR mu a ional s a us was de e mined on 132 a chi ed o malin- ixed pa a in-embedded issue sec ions wi h Idylla™ echnology. Resul s we e compa ed wi h he esul s p e iously ob ained by ou ine me hod in he e e ence lab (The asc een® EGFR RGQ PCR 2, Qiagen in Molecula Pa hology lab, Hospi al Uni e si a io Vi gen del Rocío de Se illa). Resul s: The o e all ag eemen be ween esul s ob ained wi h he Idylla™EGFR Mu a ion Tes and he Compa a o es me hod was 95.38% (wi h 1-sided 95% lowe limi o 91.7%) showing Posi i e Diagnos ic Ag eemen o 93.22% and Nega i e Diagnos ic Ag eemen o 97.18%, wi h a Limi O De ec ion ≤5%. Conclusions: The Idylla™EGFR Mu a ion Tes passed i s clinical alidi y pe o mance cha ac e is ics o accu acy. Keywo ds: Non-small-cell lung ca cinoma, EGFR, Mu a ions © The Au ho (s). 2020 Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence, and indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle's C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included in he a icle's C ea i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed in a c edi line o he da a. * Co espondence: michele.biscuola.sspa@jun adeandalucia.es 1 Depa men o Pa hology, Molecula Pa hology Labo a o y, Hospi al Uni e si a io Vi gen del Rocío-IBIS, A . Manuel Siu o , S/n, 41013 Se illa, Spain Full lis o au ho in o ma ion is a ailable a he end o he a icle Delgado-Ga cía e al. BMC Cance (2020) 20:275 h ps://doi.o g/10.1186/s12885-020-6697-7 Backg ound Non-small-cell lung ca cinoma (NSCLC) is one o he mos common cance s wo ldwide, con ibu ing o 13% o all cance ypes [1]. NSCLC ep esen he 85% o lung cance , mainly subdi ided in o wo ypes: squamous cell ca cinoma (SCC) and non-squamous cell ca cinoma (non-SCC) o which his ologically adenoca cinoma (ADC) is he mos p e alen one (50%) [2,3]. Mu a ions in he EGFR gene a e commonly obse ed in NSCLC pa icula ly in ADC [4]. The gene encodes a ansmemb ane glycop o ein, EGFR o HER1/E bB1, membe o he epide mal g ow h ac o y osine kinase (TK) ecep o s’ amily ERbB. Binding o EGFR o i s li- gands causes dime iza ion and y osine au ophospho yl- a ion. Downs eam cascade ac i a ion o pa hways, esul s in cellula p oli e a ion and su i al [5,6]. EGFR gene mu a ions, inc eased gene copy numbe and o e - exp ession o EGFR p o eins can howe e lead o cons i- u i e TK ac i i y and ca cinogenesis [7]. EGFR mu a ions a e p esen in a ound 15% o NSCLC [8]. EGFR mu a ion es ing o ac i a ing “ho -spo ”mu- a ions in exons 18–21 is ecommended in all pa ien s wi h ad anced NSCLC o he non-SCC sub ype [9]. Exon 19 dele ions, exon 21 (L858R, L861Q), and exon 18 (G719X) mu a ions a e associa ed wi h sensi i i y o EGFR- y osine kinase inhibi o s (TKIs) [10–12] whe e exon 21 L858R poin mu a ion and exon 19 dele ions a e he mos equen al e a ions (o e all 85–90%) [13]. Exon 20 (S768I) con e s a good ou come o i s gene - a ion EGFR TKIs (e lo inib, ge i inib), al hough i s asso- cia ion wi h sensi i i y o esis ance o TKIs is s ill no known [14]. Exon 20 inse ions may p edic esis ance o TKIs. Mo eo e , EGFR T790M mu a ion is one o he main causes o acqui ed esis ance o TKI he apy and has been epo ed in abou 55% o pa ien s wi h disease p og ession a e ini ial esponse o 1s o 2nd gene - a ion TKIs [10,11]. C797S is a second acqui ed esis - ance mu a ion, a ising in umo s ha ha e p og essed a e (osime inib) ea men o T790M+ disease. Ne e heless, his mu a ion is unusual and no cu en ly ea able [15]. Va ious comme cial assays a e used in ou ine p ac- ice o es he p esence o EGFR mu a ions in a u- mo al con ex and he majo i y o hose ha e been op imized o be compa ible wi h DNA ex ac ed om o malin- ixed pa a in-embedded (FFPE) samples (ob- ained om diagnos ic biopsies, su gical esec ions and e en cy ological specimens). Each assay is cha ac- e ized by a speci ic ange o co e ed mu a ions, di - e en le el o au oma ion and mul iplexing, a iable cos , high u na ound ime plus he need o special- ized equipmen and highly skilled s a bu mos o all, each o hem is cha ac e ized by speci ic sensi i i y, speci ici y and Limi o De ec ion (LOD) o each one o he es ed mu a ions. The Idylla™EGFR Mu a ion Tes , as he Idylla™KRAS and BRAF Mu a ion Tes s, is a ully-au oma ed eal- ime-PCR-based es designed o he de ec ion o EGFR mu a ions in a quick u na ound ime (app ox. 150 min) om FFPE sample o inal esul . The es is pe o med di ec ly on one FFPE issue sec ion, equi ing no be o e- hand sample p epa a ion and DNA ex ac ion and min- imal hands-on ime. The in e p e a ion o esul s is ully au oma ed. This s udy was conduc ed as a Clinical Pe o mance E alua ion (CPE) o e alua e he pe o mance o he Idylla™EGFR Mu a ion Tes on he Idylla™Sys em unde he condi ions o use [16,17], in acco dance wi h he quali y s anda d EN-13612 (2003) ‘Pe o mance e alu- a ion o in i o diagnos ic medical de ices’, in o de o demons a e he ollowings objec i es: a) o e all, posi i e and nega i e ag eemen o sensi izing and esis an mu- a ions; b) posi i e and nega i e ag eemen a he spe- ci ic geno ype call le el; c) posi i e and nega i e diagnos ic ag eemen o a leas 90% o he Idylla™EGFR Mu a ion Tes by compa ing i o a compa a o es (The asc een® EGFR RGQ PCR 2, Qiagen), in umo samples o subjec s wi h NSCLC. Me hods Pa ien s and samples Samples we e ec ui ed om ou di e en cen e s [Hospi al Uni e si a io Vi gen del Rocío de Se illa, Spain (HUVR), Uni e si ai Ziekenhuis Leu en, Belgium (UZL), Complejo Hospi ala io de Jaén, Spain (CHJ) and Hospi al Uni e si a io de Bell i ge, Spain (HUB)]. The CPE s udy was ca ied ou a wo si es: UZL and HUVR ( e e ence lab due o acc edi a ion unde he UNE-EN ISO 15189:2013 ela ed wi h his echnique). Samples used o his s udy we e slides o cu ls (slices) om a chi ed, app op ia ely s o ed and adequa ely iden i ied FFPE umo blocks. The ini ial s udy coho consis ed o 290 samples (42 om UZL and 248 om HUVR). UZL es ed 42 samples wi h Idylla™EGFR Mu a ion es and no samples wi h The asc een V2. HUVR es ed 137 samples wi h Idylla™EGFR Mu a ion es and 179 samples wi h The asc een V2. We excluded samples due o: a) in- clusion c i e ia no me (Table 1); b) absence o alid esul s o bo h es ; c) insu icien ma e ial (compa a- ion o an ini ial hema oxylin wi h he inal one). A - ending o his, he inal analysis was pe o med on 132 samples (Fig. 1) (Table 2). Pa ien s p o ided in o med consen o in es iga ional pu poses and he ins i u ional e hics commi ees o all hese cen e s app o ed he s udy. Main ea u es o he pa ien s a e shown in Table 3. Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 2 o 10 Once CPE s udy was inished, he e e ence lab in- cluded 82 ex a samples [ om HUVR, HUB, Hospi al Uni e si a io Mú ua Te assa, Spain (HUMT) and Facul- ade de Medicina da Uni e sidade de Coimb a, Po ugal (FMUC)]. A e he exclusion o 29 o hese samples o he same easons as desc ibed ea lie , 53 samples we e compa ed o Idylla™and compa a o es (supplemen- a y Table S1). Fi e-μm hick FFPE issue sec ions we e p epa ed as close as possible o he sec ions p e iously used o gen- e a e he e e ence esul s. Tumo con en , pe cen age o nec osis, p esence/absence o TAR (a la ge a ie y o o ganic and ino ganic chemicals gene a ed by bu ning obacco ha o ms a b own subs ance be ween lung cells; i is he main cause o lung and h oa cance in smoke s.) and a ea we e de e mined on a hema oxylin- eosin (HE)-s ained slide by a pa hologis . Mac o- dissec ion was pe o med o achie e umo cell con en o a leas 10%. The asc een® EGFR RGQ PCR ki e sion 2 used as e e ence me hod The asc een® EGFR RGQ PCR Ki ( e sion 2) was pe - o med acco ding o he manu ac u e ’s ins uc ions. This is an In-Vi o Diagnos ic (IVD) es o he de ec- ion o 29 soma ic mu a ions G719A/S/C in exon 18, 19 dele ions in exon 19, T790M, S768I and 3 inse ions in exon 20, and L858R, L861Q in exon 21 in he EGFR oncogene, using Sco pions® and ARMS® echnologies in eal- ime PCR. The The asc een® EGFR RGQ PCR Ki was es ed on DNA samples ex ac ed om FFPE umo issue om NSCLC pa ien s (Qiagen QIAamp® DNA FFPE-ki ), and un on a Ro o -Gene Q MDx ins umen . Idylla™EGFR mu a ion es The Idylla™EGFR Mu a ion Tes used in he s udy was an in es iga ional use only labeled p oduc as he IVD Table 1 Inclusion c i e ia o he Idylla™ es Inclusion c i e ia 1Male o emale pa ien s ≥18 yea s o age 2Samples can be used o in es iga ional pu poses acco ding o he applicable laws 3His ological con i med p ima y o me as a ic NSCLC wi h known EGFR s a us ( alid esul wi h he e sion 2 o he compa a o es ) 4Fo Idylla™ es ing: one slice o slide wi h a minimum o 10% umo cells o he o al issue used (i his is no ob ained, mac o-dissec ion is o be pe o med o each a leas 10% umo cells in o al issue a ea used) 5FFPE blocks om he ins i u e, which p e e ably had a maximum ixa ion ime o 48 h ( ou ine p ocedu e) and a e p e e ably no olde han 5 yea s a e he da e o collec ion, s o ed a ambien condi ions Fig. 1 The sample selec ion p ocess Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 3 o 10 e sion was a ha momen no ye comme cially a ail- able. This was he same p oduc as he IVD e sion ex- cep o i s labeling. The Idylla™EGFR Mu a ion Tes is a es o he quali a i e de ec ion o 51 EGFR mu a ions: exon 18 (G719A/S/C), 36 dele ions in exon 19 ( he se o 36 mu a ions in exon 19 de ec ed wi h Idylla™ ully o e lap he 19 mu a ions de ec ed wi h The asc een), exon 20 (T790M, S768I), 5 inse ions in exon 20 ( he 2 se s o mu a ions a e o e lapping in 2 mu a ions (c.2310_2311insGGT; p.D770_N771insG and c.2319_ 2320insCAC; p.H773_V774insH).) and exon 21 (L858R, L861Q) in he EGFR oncogene in FFPE human malig- nan lung cance issue. FFPE issue sec ions we e placed (one pe sample) di ec ly in o he ca idge o he ully au oma ed Idylla™pla o m (Bioca is, Mechelen, Belgium) ollowing he manu ac u e ’s ins uc ions, wi hou e- qui ing p io manual depa a iniza ion o FFPE p e- p ocessing. Wi h a hands-on ime o less han 2 min and a o al u na ound ime o 150 min, he ins u- men co e s ully in eg a ed sample p epa a ion (wi h a combina ion o eagen s, enzymes, hea , and high in ensi y ocused ul asound (HIFU) inducing depa a - iniza ion, dis up ion o he issue, and lysis o he cells) combined wi h PCR he mocycling ( ia mic o- luidic channels in he ca idge, nucleic acids a e anspo ed in o 5 sepa a ed chambe s wi h d ied o m PCR eagen s) and luo escence de ec ion o a ge sequences, using allele speci ic p ime s. A sam- ple p ocessing con ol (SPC) is included in each un and he p esence o a mu an geno ype is de e mined by calcula ing he ΔCq (quan i ica ion cycle) be ween he EGFR SPC and he EGFR mu an signal(s). All e- qui ed consumables a e p o ided in he ca idge and he Idylla™Console and he Idylla™ins umen s a e CE ma ked. E alua ion o samples and in e e ences Al hough inclusion c i e ia we e well es ablished, an as- sessmen was made o di e en cha ac e is ics o sam- ples o a oid in alid o alse esul s, including: i) Age o p epa ed FFPE blocks: 14 samples wi h an unknown p epa a ion da e and 9 blocks olde han 5 yea s. ii) Mac o-dissec ion: needed o inc ease he pe cen age o umo nuclei o each a leas 10%. iii) Tissue a ea: issue a ea o samples was be ween 1 and 567 mm 2, since he e was no minimum issue a ea equi emen inpu o he EGFR Mu a ion es . i ) O he in e e ences: he p esence o nec o ic issue and TAR. Ne e heless, i an in alid esul was ob ained, bo h es s we e epea ed once. In alid esul s may be caused by a a ie y o easons including p esence o inhibi o s in he sample, insu icien DNA, inco ec placemen o a sample in a ca idge and/o sample olume ou o ange. A his poin , i is impo an o claim ha he epe i ion o in alid esul s in The - asc een was pa o he s udy, bu i is also pa o ou ou ine diagnos ic p o ocol, usually modi ying DNA concen a ion and/o epea ing ex ac ion in o de o a oid nec osis o TAR. Analysis o disco dan esul s A hi d me hod was used o u he analyze some o he samples ha ing an Idylla™EGFR Mu a ion Tes esul no conco dan wi h he esul o he e e ence me hod. Nex gene a ion sequencing (NGS) and/o D ople Digi al™PCR (ddPCR) we e used depending on he quan i y o le o e ma e ial a ailable (sec ions close o hose used o he o he es s we e p o ided). NGS was done (wi h a minimum amoun o 8 slices) by a alida ed wo k low o he Tumo Ho spo MASTR™ Plus ki (Mul iplicom) on he Illumina MiSeq Dx ins u- men . NGS and he subsequen da a-analyses pipeline was done by His ogenex (minimal o al mean ead dep h o 185.000, exon co e age o 500x mean ead dep h). The ddPCR was pe o med a Bioca is. ddPCR was done on lique ied FFPE ma e ial using comme cially a ailable ddPCR assays (D ople Digi al™PCR Assays Table 2 CPE s udy coho CPE s udy Posi i e EGFR Nega i e EGFR To al Se illa 27 47 74 Jaen 6 0 6 Leu en 19 22 41 Bell i ge 11 0 11 To al 63 69 132 Table 3 Demog aphic and clinical cha ac e is ics o pa ien s. * Mean age (yea ) a umo collec ion da e To al numbe (NSCLC) 132 Gende Male: 56% Female: 43% Missing: 1% Mean age (yea )* 65 Sd: 10.12 median: 65 min: 44 max: 85 Missing: 11 Tissue loca ion Numbe o pa ien s (%) P ima y 85 (64%) Dis an me as ases 15 (11.5%) Me as asis in lymph nodes o pleu a 15 (11.5%) Unknown 17 (13%) Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 4 o 10 and QX200 ddPCR sys em, Bio-Rad Labo a o ies, Inc.). These p edesigned assays con ain p obes o he de ec- ion o bo h WT and all he speci ic mu a ions. Samples we e conside ed posi i e by ddPCR when he % mu an was ≥1%, excep o T790M which was conside ed posi- i e when he % mu an was ≥5%. Fu he mo e, i was necessa y o analyze he deg ee o agmen a ion o see i he DNA in he disco dan sam- ples was hea ily agmen ed o no , which could be a p oblem o a PCR based analysis me hod like Idylla™.A 5-plex PCR was de eloped and execu ed by Bioca is. Samples we e lique ied on he pla o m ollowing a PCR eac ion o 5 housekeeping genes: β-ac in (321 bp), ABCB (213 bp), TFRC (149 bp), HPRT (105 bp) and RNaseP (63 bp). A sample is conside ed agmen ed when he size o he amplicons de ec ed wi h he 5-plex PCR is smalle han he amplicons ha would be needed o he EGFR es (EGFR PCR p oduc s ange om 67 o 170 bp). S a is ical analysis Nine y- i e pe cen wo sided con idence in e al based on Wilson’s sco e me hod [18] a he dicho omous le el (“mu a ion de ec ed” e sus “no mu a ion de ec ed”) was used o he es ima ion o o al, posi i e and nega i e ag eemen . Speci ici y and sensi i i y we e de ined as he p opo - ion o conco dan esul s agains he sum o conco dan and disco dan esul s ( ue posi i es / ( ue posi i es + alse nega i es) and ue nega i es / ( ue nega i es + alse posi i es). Analyses we e pe o med in R so wa e 3.2.5 (R Co e De elopmen Team, 2016). Resul s Compa a o es and Idylla™EGFR mu a ion es In a p ima y analysis on dicho omous le el, he in alid uns a i s es ing we e excluded, esul ing in a o al o 122 samples. A e epea es ing, 10 mo e samples p o- ided a alid esul o Idylla™and/o Compa a o es V2, leading o a o al o 132 samples included o he seconda y analysis. The EGFR mu a ional s a us o 132 e ospec i e clin- ical FFPE samples om pa ien s wi h p ima y o me a- s a ic NSCLC was es ed wi h Idylla™Sys em (Idylla™), and esul s we e compa ed wi h he o iginal assessmen s made by The asc een® EGFR RGQ PCR Ki ( e sion 2). Idylla™ esul s we e no used o any diagnos ic o he a- peu ic pu poses. We ob ained 57 posi i e samples o EGFR mu a ion and 75 wild- ype cases. Idylla™demons a ed ag eemen wi h ou ine me hod in 121 ou o 132 samples (91.7%). Two samples had an Idylla™ esul wi h a posi i e mu a- ion de ec ed ha was no de ec ed by he The asc een® es . Fou samples had a The asc een® esul wi h a posi i e mu a ion ha was no de ec ed by he Idylla™ es . Th ee samples had di e en posi i e mu a ions de- ec ed by bo h es s, and wo samples con ained a spe- ci ic exon 20 inse ion: c.2311_2319dupAACCCCCAC; p.Asn771_His773dup [p.N771_H773dup] ha was no a ge ed by he Idylla™EGFR Mu a ion Tes and we e he e o e disco dan by design (Table 4). Samples cha ac e is ics The age o he FFPE blocks should be p e e ably maximum 5 yea s a e he da e o collec ion. This was he case o all he blocks excep o 14 samples o which he p epa a ion da e is “unknown”and 9 blocks which we e olde han 5 yea s. All samples ha e been included in he analysis, since no age- ela ed end ega ding in alid es s was iden i ied (supplemen a y Table 2). Tissue a ea o samples was be ween 1 and 567 mm 2 . No issue size- ela ed end was obse ed in alid and disco dan a e. The samples wi h a issue a ea below 10mm 2 s ill led o alid and conco dan esul s in > 90% o he samples (supplemen a y Table 3). The in luence o he p esence o nec o ic issue and TAR on he esul s was e alua ed in all samples whe e possible. Resul s showed ha nei he nec o ic issue- ela ed no TAR- ela ed ends we e appa en in he ob- ained esul s (supplemen a y Table 4). Compa ison be ween bo h me hods A alid esul wi h bo h me hods was ob ained o 132 ou 179 o ini ially selec ed samples. A e he exclusion o wo disco dan esul s by design, he ag eemen be- ween Idylla™and he compa a o es was calcula ed based on he dicho omous esponse whe he a mu a ion was de ec ed o no . Table 5shows he aw da a used o calcula e i . The o e all ag eemen (de ined as he p o- po ion o conco dan esul s in all esul s) o 130 sam- ples was 95.38% wi h a lowe limi o he 95% con idence in e al (CI) o 91.32. Posi i e ag eemen was calcula ed o be 93.22% CI o 85.73, and, he nega i e ag eemen was calcula ed o be 97.18% CI o 91.84 (Table 6). In alid esul s we e ob ained o 47 ou 179 selec ed samples: wi h Idylla™o wi h compa a o es o wi h bo h echnologies: in alid esul only o The asc een® (34 ou 47); in alid esul only o Idylla™(6 ou 47); in- alid esul o bo h es s (7 ou 47). An addi ional pos -CPE s udy was done wi h 53 ex a samples wi h he same cha ac e is ics desc ibed p e i- ously. Idylla™ esul s showed ag eemen wi h ou ine me hod in 49 samples. We ound ha 3 samples had a The asc een® esul wi h a posi i e mu a ion de ec ed ha was no de ec ed by he Idylla™ es , and 1 sample had di e en posi i e mu a ions de ec ed by he wo Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 5 o 10 Table 4 Ag eemen able a he mu a ion speci ic le el The asc een® Idylla™DelEx19 DelEx19,T790M L858R L861Q G719X G719X, S768I L858R,T790M InsEx20 WT To als DelEx19 20 20 DelEx19,T790M 5 5 L858R 22 1 23 L861Q 2 2 G719X 1 1 G719X,S768I 1 1 L858R,T790M 1 1 T790M 1 1 2 DelEx19, S768I 1 1 S768I 1 1 WT 1 2 1 2 69 75 To als 21 7 24 2 2 1 2 2 71 132 Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 6 o 10 es s. Howe e , he o e all conco dance be ween he Idylla™EGFR Mu a ion Tes and he e e ence ou ine me hod was ound o be 94.34%, wi h a nega i e ag ee- men o 100% and a posi i e ag eemen o 89.26% (sup- plemen a y Tables 5, 6 and 7). Disco dan esul s we e no analyzed. Disco dan esul s Disco dan samples we e es ed wi h NGS and/o ddPCR o in es iga e he oo -cause o he disco dances. A disco dan esul was obse ed o 11 samples (Table 4). Two disco dan esul s we e he esul o a speci ic inse ion (p.Asn771_His773dup) ha is no a - ge ed in he design o he Idylla™EGFR Mu a ion Tes (disco dan by design) and we e no aken in o accoun in he ag eemen calcula ions. Fou disco dan samples had insu icien ma e ial o NGS analysis. The emaining i e disco dan samples we e included in NGS analyses, oge he wi h six andom conco dan samples (Table 7). All six conco dan samples we e con i med using NGS (da a no shown). Discussion Molecula diagnosis o ac i a ing “ho -spo ”mu a ions in EGFR exons 18 o 21 is ecommended in all pa ien s wi h ad anced NSCLC o a non-SCC sub ype since TKI he apy p o ides signi ican imp o emen in su i al and quali y o li e [9]. Classical me hods o es ing EGFR in FFPE ma e ial equi e issue depa a iniza ion, manual isola ion o DNA and DNA quan i ica ion [9]. The asc een® EGFR RGQ PCR Ki ( e sion 2) wi h a o al u na ound ime o 20 h (including DNA ex ac ion, pu i ica ion and PCR), is conside ed a obus me hod o he de ec ion o ‘ho spo ’mu a ions p edic i e o TKI esponse, and is widely u ilized [19,20]. The Idylla™EGFR Mu a ion Tes , pe o med on he Bioca is Idylla™Sys em, is an in i o diagnos ic ully au oma ed eal- ime PCR based es o he quali a i e de ec ion o EGFR mu a ions, om FFPE sample o inal esul . The es is pe o med di ec ly on one FFPE issue sec ion, equi ing no be o ehand sample p epa a ion and minimal hands-on ime, yielding esul s wi hin 2.5 h. This es encompasses 51 EGFR clinically ele an mu a- ions om exons 18 o 21 in human lung cance FFPE ma e ial (acco ding o he la es IASLC a las o EGFR es ing in lung cance ) [21]. The pe o mance o he Idylla™pla o m has been p e- iously epo ed in a ious s udies [16,17,22–25]. This CPE s udy compa ed he The asc een® EGFR RGQ PCR 2 o he Idylla™EGFR Mu a ion Tes , showed an o e all p esence o EGFR mu a ions in he samples o 45.38% when measu ed wi h The asc een® EGFR RGQ PCR 2 and o 43.85% measu ed wi h he Idylla™EGFR Mu a ion Tes , demons a ing ha he sensi i i y o he Idylla™EGFR Mu a ion Tes is compa - able o he sensi i i y o he ou ine e e ence me hod. The o e all conco dance be ween he Idylla™EGFR Mu- a ion Tes and he e e ence ou ine me hod was ound o be 95.38% [95% CI: 91.32%-100], indica ing a good conco dance, showing he nea -equi alence o bo h measu ing echniques, and, consequen ly, es obus ness o Idylla™. A e he exclusion o 2 esul s disco dan by design (wi h he speci ic inse ion p.Asn771_His773dup ha is no a ge ed in he design o Idylla™EGFR Mu a ion Tes ), 9 disco dan esul s we e ound be ween bo h me hods. Using NGS and/o ddPCR and analyzing he deg ee o DNA agmen a ion, he main easons o hese disc epancies we e s udied, including insu icien ma e - ial, low sample inpu and/o low allelic equency (in he cases wi h enough le o e ma e ial and good sample quali y) (Table 7). Besides, se e al o he p e-analy ical pa ame e s we e e alua ed, such as he age o he FFPE blocks, issue a ea, pe cen age o nec o ic issue, and he p esence o TAR. Fo hese 4 pa ame e s, no co ela ion could be ound, indica ing ha he Idylla™EGFR Mu a ion Tes is a obus es . The numbe o in alid esul s, ob ained wi h one o he wo me hods o wi h bo h, ep esen s an in e es ing aspec o his s udy. We ob ained a The asc een® in alid esul o 34 ou 47 samples. As we de ailed be o e, Table 6 Measu es o ag eemen Measu e Ra e Poin es ima e 95% lowe limi (1-sided) 95% uppe limi (1-sided) O e all Diagnos ic Ag eemen 124/130 95.38 91.32 100 Posi i e Diagnos ic Ag eemen 55/59 93.22 85.73 100 Nega i e Diagnos ic Ag eemen 69/71 97.18 91.84 100 Table 5 Ag eemen able a he dicho omous le el o alid, non-missing esul s The asc een® Idylla™Mu an WT To als Mu an 55 2 57 WT 4 69 73 To als 59 71 130 a a n= 130 samples a e excluding wo disco dan by design Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 7 o 10 Table 7 Resul s o disco dance analysis and conclusions o he oo cause The asc een® V2 Idylla™NGS ddPCR F agmen a ion Resul / Roo case 1Wild ype T790M T790M Wild ype o T790M No Con i med wi h NGS 2Wild ype S768I ND a ND a ND No le o e ma e ial o analysis 3G719X Wild ype G719X b low inpu Yes Sample quali y ( agmen a ion) + low sample inpu 4DelEx19 Wild ype p.Glu746_Al a750del low inpu No Low sample inpu + low allelic equency in NGS (6%) T790M wild ype low inpu Low sample inpu 5L858R Wild ype L858R low inpu No Low sample inpu 6L858R Wild ype ND a low inpu yes Sample quali y ( agmen a ion) + low sample inpu 7L858R L858R ND a low inpu No Low sample inpu , insu icien ma e ial o NGS T790M wild ype o T790M 8DelEx19 DelEx19 DelEx19 DelEx19 no es ed No Conco dan esul Wild ype S768I wild ype 6,7% S768I Low allelic equency (as con i med wi h ddPCR) 9DelEx19 T790M ND a Wild ype o Del15 & Del 18 No Con i med wi h ddPCR, insu icien ma e ial o NGS T790M 6,2% T790M a No enough ma e ial a ailable; b Va ian Allele F equency below LOD and p esen in only one o wo amplicons co e ing he a ian posi ion; ND No de e mined Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 8 o 10 epe i ions wi h his es occu du ing ou ou ine p ac- ice and, o hose cases, we usually modi y DNA con- cen a ion and/o epea ex ac ion in o de o a oid nec osis o TAR. Howe e , al hough we modi ied he s a ing ma e ial in he epe i ions, his s a egy was no enough o ob ain an e aluable esul o hese samples es ed wi h he compa a o es . In alid esul s o Idylla™(6 ou 47 samples) we e mainly due o he absence o ampli ica ion o he in- e nal con ol. We epea ed he es a leas one ime o each one o he in alid cases ob aining he same esul . We conside ed o y o a oid nec osis o TAR wi h mac odissec ion bu , as p e iously demons a ed, hese cha ac e is ics do no in e e e wi h he inal esul . Finally, in alid esul s o bo h es we e ob ained o 7 ou 47 samples, and, al hough we epea ed hese samples wi h bo h echniques once, we ob ained he same esul . We a gued ha his kind o samples p esen ed some in- insic a e ac s no con olled in ou lab, o example, an imp ope ixa ion ime and/o a mo e ele a ed p esence o speci ic PCR inhibi o s due o he p ocess, concluding ha he quali y o hese samples was no adequa e o s udy he mu a ional s a us o he EGFR gene. The small numbe o in alid samples ob ained wi h Idylla ep esen s an impo an aspec o he EGFR s a- us de e mina ion because mo e pa ien s could be sc eened o hese mu a ions in o de o ecei e a mo e pe sonalized ea men . The ad an ages o his sys em ha e been clea ly exposed be o e [16,17,22–25], he load o FFPE samples di ec ly in o single use ca idges wi h minimal sample p epa a ion (and as consequence, minimal p obabili y o con amina- ions), a quick u na ound ime and a ully au oma ed in e - p e a ion o esul s. Some disad an ages and/o cau ions need o be kep in mind, such as a limi ed h oughpu (due o hesys emonlyp ocessingonesamplea a ime)and, consequen ly, he di icul ies o labo a o ies wi h high daily sample wo kload. Howe e , is possible o add up o eigh di e en modula sys ems o he same console. A his poin , i is impo an o emphasize he c ucial impo ance o he pa hologis ’s p e-analy ical e alua ion in assessing he pe cen age o umo cells, and o he cha ac e is ics. P e ious (and manda o y) e alua ion o issue helps o a oid alse and/o in alid esul s. Conclusions All he cha ac e is ics exposed be o e, oge he wi h he high conco dance wi h he e e ence me hod [The asc een® EGFR RGQ PCR Ki ( e sion 2)], indica e ha Idylla™EGFR Mu a ion Tes on he Idylla™Sys em is a ully-au oma ed me hod, designed o he de ec ion o EGFR mu a ions in a quick u na ound ime om FFPE sample o inal esul and can be a sui able clinical es o ou ine use in diagnos- ic p ocedu es. Supplemen a y in o ma ion Supplemen a y in o ma ion accompanies his pape a h ps://doi.o g/10. 1186/s12885-020-6697-7. Addi ional ile 1 Supplemen a y Table 1. pos -CPE s udy coho . Supplemen a y Table 2. age o p epa ed FFPE blocks. Supplemen a y Table 3. assessmen o he issue a ea. Supplemen a y Table 4. e alu- a ion o nec o ic issue and a . Supplemen a y Table 5: pos -CPE e- sul s. Supplemen a y Table 6. ag eemen able a he dicho omous le el o alid, non-missing esul s. Supplemen a y Table 7. pos -CPE measu es o ag eemen . Abb e ia ions ADC: Adenoca cinoma; CPE: Clinical pe o mance e alua ion; ddPCR: D ople digi al™PCR; EGFR: Epide mal g ow h ac o ecep o ; FFPE: Fo malin- ixed pa a in-embedded; HE: Hema oxylin-eosin; HIFU: High in ensi y ocused ul asound; IVD: In- i o diagnos ic; LOD: Limi o de ec ion; NGS: Nex gene a ion sequencing; Non-SCC: Non-squamous cell ca cinoma; NSCLC: Non-small-cell lung ca cinoma; SCC: Squamous cell ca cinoma; SPC: Simple p ocessing con ol; TK: Ty osine kinase; TKIs: Ty osine kinase inhibi o s Acknowledgemen s We hank HUVR-IBiS Biobank (Andalusian Public Heal h Sys em Biobank) o i s help and suppo o clinical samples, used in his wo k. We also acknow- ledge oncologis s R.B and A.L.G o ec ui ing pa ien s. Finally, we would also like o hank e iewe s o cons uc i ely e ising his manusc ip . Au ho s’con ibu ions M.D-G., B.W., L.G-I., M-J.H., A-M.B., E.dA. and M.B. con ibu ed o he s udy concep ion and design, pa icipa ed in sample collec ion, p ocessing, and da a acquisi ion, and ook pa in he s a is ical analysis and in e p e a ion o esul s. M.V., X.M-G., E.N., B.M-L., A.A. pa icipa ed eclui ing pa ien s and in sample collec ion. M.D-G. and M.B. w o e he manusc ip . All au ho s e iewed he manusc ip c i ically o in ellec ual con en and app o ed he inal e sion o be published. Au ho s’in o ma ion No applicable. Funding Suppo ed by each clinical cen e . Bioca is p o ided ee-o -cha ge Idylla™ EGFR Mu a ion Tes ca idges. A ailabili y o da a and ma e ials The da ase s used and/o analyzed du ing he cu en s udy a e a ailable om he co esponding au ho upon easonable eques . E hics app o al and consen o pa icipa e This s udy was app o ed by he E hical commi ee o he Hospi al Uni e si a io Vi gen del Rocío wi h e e ence 33160034. All pa ien s olde han 18 consen o ha e esidual ma e ial es ed o scien i ic pu poses acco ding o he Spanish legisla ion o 3 July 2007, unless hey s a ed di e en ly by w i en ag eemen . Consen o publica ion No applicable. Compe ing in e es s All he au ho s decla e ha no compe ing in e es s exis . Au ho de ails 1 Depa men o Pa hology, Molecula Pa hology Labo a o y, Hospi al Uni e si a io Vi gen del Rocío-IBIS, A . Manuel Siu o , S/n, 41013 Se illa, Spain. 2 Depa men o Pa hology, Uni e si ai Ziekenhuis Leu en, Leu en, Belgium. 3 Depa men o Pa hology, Hospi al Uni e si a i de Bell i ge, Idibell, Oncobell, Ba celona, Spain. 4 Depa men o Pa hology, Hospi al Uni e si a i A nau de Vilano a, Uni e si y o Lleida, IRBLleida, CIBERONC, Lleida, Spain. 5 Depa men o Medical Oncology, Ca alan Ins i u e o Oncology, Idibell, Oncobell, Ba celona, Spain. 6 Depa men o Pa hology, Complejo Hospi ala io Delgado-Ga cía e al. BMC Cance (2020) 20:275 Page 9 o 10