Next-generation sequencing of tissue and circulating tumor DNA: Resistance mechanisms to EGFR targeted therapy in a cohort of patients with advanced non-small cell lung cancer

doi:10.21203/rs.3.rs-30958/v2

Download PDF

Research

Next-generation sequencing of tissue and circulating tumor DNA: Resistance mechanisms to EGFR targeted therapy in a cohort of patients with advanced non-small cell lung cancer

https://doi.org/10.21203/rs.3.rs-30958/v2

This work is licensed under a CC BY 4.0 License

Version 2

posted

You are reading this older preprint version

Read the latest preprint version →

Background: Epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI) has been considered as an effective treatment in epidermal growth factor receptor-mutant (EGFR-mutant) advanced non-small cell lung cancer (NSCLC). However, most patients develop acquired resistance eventually. Here, we compared and analyzed the genetic alterations between tissue assay and circulating tumor DNA (ctDNA), and further explored the resistance mechanisms after EGFR-TKI treatment.

Methods: Amplification refractory mutation system-polymerase chain reaction (ARMS-PCR), Cobas^® ARMS-PCR and next-generation sequencing (NGS) were performed on tissue samples after pathological diagnosis. Digital droplet PCR (ddPCR) and NGS were performed on plasma samples. The association between genetic alterations and clinical outcomes was analyzed retrospectively.

Results: Thirty-seven patients were included. The success rate of re-biopsy was 91.89% (34/37). The total detection rate of EGFR T790M was 62.16% (23/37) and the consistency between tissue and ctDNA was 78.26% (18/23). Thirty-four patients were analyzed retrospectively. Twenty-four patients harbored concomitant mutations. Moreover, tissue re-biopsy at resistance showed 21 patients (21/34, 61.76%) had concomitant T790M mutation, 4 with MET amplification and 4 with PIK3CA mutation. Patients with T790M mutation (p=0.010 & p=0.017) or third-generation EGFR-TKI treatment (p<0.0001 & p=0.073) showed better progression-free survival (PFS) and overall survival (OS). Interestingly, concomitant genetic alterations were significantly associated with a worse prognosis for patients with T790M mutation receiving third-generation EGFR-TKIs (p=0.037).

Conclusions: Multi-platforms are feasible and highly consistent for re-biopsy after EGFR-TKI resistance. Concomitant genetic alterations may be associated with a poor prognosis for patients with T790M mutation after third-generation EGFR-TKIs.

Trial Registration: The clinical trial registration was carried out on ClinicalTrials.gov. (NCT03309462), registered on September 1st 2017. https://register.clinicaltrials.gov/prs/app/action/SelectProtocol?sid=S0007G1B&selectaction=Edit&uid=U0001WUU&ts=2&cx=frpicv

Pulmonology

pidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI)

tissue biopsy

non-small cell lung cancer (NSCLC)

genetic alterations

clinical outcomes

Lung cancer is the leading cause of cancer‑relatedmortalityworldwide [1]. Among them, non-small cell lung cancer(NSCLC)accounts for about 85% lung cancer [2]. Epithelial growthfactorreceptor (EGFR) is the most common driver gene in NSCLC,occurringin an estimated 50% of adenocarcinoma cases in Asia [3].Exon 19deletion (19del) and exon 21 p.L858R (21L858R) mutationaccount forabout 90% of all EGFR activating mutations and are themostrelevant predictors of response to EGFR tyrosine kinaseinhibitor(EGFR-TKI) [4].

However, almost all patients eventually develop acquireddrugresistance after EGFR-TKI treatment. And the occurrence ofEGFRmutation p.T790M in exon 20 represents the most frequentmechanismof the acquired resistance [5]. The third-generationEGFR-TKI is anirreversible selective TKI, which specificallytargets EGFR T790Mand EGFR activating mutations, and has beenproven effective inpatients with EGFR T790M-positive NSCLCfollowing acquiredresistance to prior EGFR-TKIs [6, 7]. This makesre-biopsy widelyaccepted in clinical practice [8]. Through there-biopsy, it ispossible to effectively understand the cause ofdrug resistance andprovide a basis for follow-up treatment.

Unfortunately, resistance to third-generation EGFR-TKIsalsoinevitably occurs [9, 10]. In addition, the detection rateofmutations has increased significantly with the developmentofnext-generation sequencing (NGS), while the associationofconcomitant genetic alterations with the treatment responseisstill uncertain [11, 12]. Therefore, it is important to explorethestatus of re-biopsy to identify resistance mechanisms earlyinpatients with targeted drugs.

Patient selection

Patients with advanced NSCLC, who were evaluated asprogressivedisease (PD) after the first-generation and thesecond-generationEGFR-TKI treatment from September 2017 to January2019with initial diagnosis of EGFR 19del or EGFR 21L858Rmutation,were enrolled. The inclusion criteria included a.18-80years old;b. patients with initial diagnosis of advanced NSCLC andmolecularpathology confirmed the presence of EGFR-sensitivemutations; c.clinically accepted first and second-generationEGFR-TKI treatment;d. according to the criteria of solid tumorevaluation criteria(response evaluation criteria in solid tumors,RECIST), cliniciansbelieve that re-biopsy is necessary to guide thetreatment of thepatient; e. patient performance status (PS) ratedas score ≤ 2according to the Eastern Cooperative Oncology Group(ECOG). Theexclusion criteria were as follows: a. the patientreceived bloodtransfusion therapy within 1 month; b. patients withautoimmunediseases, including but not limited to systemiclupuserythematosus, rheumatoid arthritis, Sjogren's syndrome, etc.;c.patients with serious diseases were not suitable for biopsy;d.patients refused to participate in clinical trials; e.theinvestigator believed that the patient had other conditionsthatwere not suitable for this clinical trial. The study wasapprovedby the Ethics Committee of Shanghai Chest Hospital, and theethicalapproval number was KS1703. All patients were fully informedandsigned informed consent. The clinical trial registrationwascarried out on ClinicalTrials.gov. (NCT03309462).

Sample preparation

Liquid biopsy specimens were collected via standardvenipuncturetechniques into two tubes. And circulating tumor DNA(ctDNA) wasextracted from the plasma fraction of EDTA blood sampleswithin twohours of collection as recommended (QIAamp CirculatingNucleic AcidKit; Qiagen, Germantown, MD). Tissue samples wereobtained fromsmall biopsies with multiple techniques, including butnot limitedto transbronchial biopsy (TBB), transbronchial lungbiopsy (TBLB),transbronchial needle aspiration (TBNA),transthoracic needleaspiration (TTNA), and percutaneous lymph nodeneedle biopsy. Thenthe obtained tissue samples were fixed andembedded. After that,the formalin-fixed, paraffin-embedded (FFPE)specimens underwenthistological review by hematoxylin eosinstaining before subjectingto nucleic acid extraction. DNA waspurified with the use of aQIAamp DNA FFPE Kit (Qiagen, Valencia,CA), and the quantity of DNAwas verified with the use of Qubit 3.0with a dsDNA HS Assay Kit(Life Technologies).

Gene testing methods

ARMS-PCR for detecting EGFR mutations was performed via EGFR21Mutations Detection Kit (Amoy Diagnostics, Xiamen, China). DNAwasextracted from 10 to 15 unstained FFPE sections, each5μmthickness, using QIAamp DNA FFPE tissue kit (Qiagen, Valencia,CA,USA) according to the manufacturer’s instructions.Theconcentration of DNA was measured by SMA4000spectrophotometer(Merinton, Beijing, China). Cobas^®ARMS-PCR wasperformed with 10-15 slides of 5μm paraffin sectionssubjected toDNA extraction using a Cobas^® DNA samplepreparation kit(Roche Diagnostics, Indianapolis, USA), and DNAconcentration andpurity were examined using a Nanodrop UV-Visspectrophotometer. DNAintegrity was examined by agarose gelelectrophoresis. Genetictesting was performed using the human EGFRgene kitCobas^® EGFR Mutation Test v2 (RocheDiagnostics,Indianapolis, USA) for EGFR 18-21 exons mutations. ForNGS, thelibrary was constructed based on OncoAim™ cancer 223 genepanel(Singlera Genomics, Inc., Shanghai, China) with a total of1300exon regions, 456 hotspots, 21 intron regions, and 1 genepromoter.DNA shearing was performed using Covaris M220, followed byendrepair, phosphorylation and adaptor ligation. Fragments ofsize200-400bp were selected using Agencourt AMPure beads(BeckmanCoulter, Brea, CA, USA) followed by hybridization withcaptureprobes baits, hybrid selection with magnetic beads andPCRamplification. A bio-analyzer high-sensitivity DNA assaywasperformed to assess the quality and size of the fragments. Fiftyngof DNA was used for library construction. Twelve PCR cycleswereused for library amplification. The indexed samples weresequencedon Hiseq500 sequencer (Illumina, Inc., San Diego, CA, USA)withpaired reads (read length 150 bp). The sequencing data in theFASTQformat were mapped to the human genome (hg19) using BWAaligner0.7.10. Local alignment optimization, variant callingandannotation were performed using GATK 3.2, MuTect, andVarScan,respectively. DNA translocation analysis was performedusing bothTophat2 and Factera 1.4.3. Gene-level copy numbervariation wasassessed using a t statistic after normalizing readsdepth at eachregion by total reads number and region size, andcorrectingGC-bias using a LOESS algorithm.

Follow-up

Clinical follow-up assessments including physicalexaminations,imaging, and routine laboratory tests were performedevery 4 weeks.Tumor response was assessed according to the RECIST(version) 1.1by experienced investigators and categorized ascomplete response(CR), partial response (PR), stable disease (SD)or PD.Progression-free survival (PFS) was defined as the durationbetweenthe initiation of EGFR-TKIs and progression of disease ordeath forany cause, whichever occurred first. Similarly, overallsurvival(OS) was calculated from the date of starting EGFR-TKIs tothe dateof death for any cause or last follow-up (censoredpatient).Patients who were still alive were censored on their dateof lastfollow-up per chart review. The cut-off date for analysiswas April25, 2019.

Statistical analysis

Frequency, percentage, average ± SD, median (range)werepresented as appropriate. Survival was estimated withKaplan–Meiermethodology and compared with the use of log-rank testbetweendifferent groups. P values were calculated using Fisher’sexacttest or Pearson test for categorical variables orcontinuousvariables, respectively. Wilcoxon test was used forcomparingcontinuous variables to binary variables. Statisticalanalysis wasperformed using SPSS v25 (IBM Corporation, NY, USA). APvalue less than 0.05 was considered statisticallysignificant.

Patients’ characteristics

In this study, 39 patients were enrolled and 34 patientswereincluded in the retrospective analysis. The enrollment processisshown in Figure 1. Two patients were excluded, due to one19delfalse positive and one T790M mutation at baseline.Thirty-sevenpatients were successfully tested for plasma with asuccess rate of100% (37/37), while three patients failed to performtissuere-biopsy due to few tumor cells by pathological examination.Thecharacteristics of initial and secondary biopsy ofenrolledpatients were summarized in Table 1 and 2. Thirty-fourpatients’clinical characteristics for retrospective analysis wereshown inTable 3. Pathology revealed only one adenosquamous cellcarcinoma,the others were adenocarcinoma (97.06%, 33/34). And thecommon EGFRactivating alterations were largely represented (67.65%,23/34,19del; 32.35%, 11/34, 21L858R). During first-line treatment,medianPFS resulted 13 months. After progression, 70.59% (24/34)ofpatients received third-generation EGFR-TKI treatment, andmedianPFS resulted 6 months. The first patient startedthird-generationEGFR-TKI treatment on October 24, 2017 and the lastone assumed thefirst dose on August 17, 2018. The data cut-off forthis analysiswas April 25, 2019.

Genetic alterationsinctDNA and tissue of enrolledpatients

Of the 37 patients who had plasma sent for ctDNA NGS, 34(91.89%)also had tissue sent for solid tumor NGS. Out of 34patients withsamples sent for both ctDNA and tissue NGS, 34 hadshared at least 1alteration identified by both tissue NGS andctDNA analysis. InFigure 2A, the most frequent alterationsdetected by ctDNA NGS aredisplayed. The three most frequent ctDNAalterations involved thefollowing genes: TP53 (67.57%, 25/37),followed by KRAS (8.11%,3/37) and amplification of c-Met (5.41%,2/37) (Figure 2B). The mostfrequent alterations detected by tissueNGS involved the followinggenes: TP53 (52.94%, 18/34),amplification of c-Met (11.76%, 4/34)and PIK3CA (11.76%, 4/34)(Figure 2C&D). In total, 83.78%(31/37) of patients harboredconcomitant mutations, 70.27% (26/37)by ctDNA and 70.59% (24/34)by tissue NGS. Patients with a historyof smoking (85.71% [6/7] vs66.67% [18/27]) was found in tissue witha higher incidence ofconcomitant mutations, but not in ctDNA (37.5%[3/8] vs 79.31%[23/29]).

A total of 23 cases of EGFR T790M mutation in plasmas andtissueswere detected after first-line EGFR-TKI treatment. Thetotalpositive rate was 62.16% (23/37). The positive rate oftissuesamples to detect EGFR T790M mutation was 64.71% (22/34), andthepositive rate of plasma samples to detect EGFR T790M mutationwas51.35% (19/37). One of the patients had a positive mutationinplasma, while the tissue was negative. This patient wastreatedwith Osimertinib in the follow-up treatment. The bestefficacy wasevaluated as PR, and PFS was 7.0 months. Therefore, wecombined theefficacy and test results to consider the tissue testresults asfalse negative, suggesting that tissue and ctDNA assayprovidedcomplementary results. The remaining 18 patients withpositiveblood tests were consistent with those having positivetissue tests(Figure 3A). Therefore, the consistency rate betweentissue andplasma for EGFR T790M mutation was 78.26% (18/23). Inaddition,EGFR T790M mutation in plasma samples were tested andcomparedwith ddPCR and NGS as illustrated by the Venn diagrams.Tissueswere detected and compared by ARMS-PCR, Cobas^®ARMS-PCRand NGS (Figure 3B&C). The consistency rate in plasmaandtissue for different platforms were 84.21% (16/19) and78.26%(18/23), respectively.

Treatment outcome

Of the 34 patients in the retrospective analysis, 24receivedthird-generation EGFR-TKI treatment and 10 received othersafterprogression with first-line EGFR-TKI treatment.Twenty-four(70.59%, 24/34) of patients exhibited prominent tumorshrinkageduring treatment. Among them, 22 received third-generationEGFR-TKItreatment, revealing the strength of third-generationEGFR-TKIs andradio/chemotherapy for patients with resistance tofirst-linetreatment (Figure 4). Thirteen partial responses wereobserved andall occurred in T790M-positive patients (contained onewho wasdetected only by ctDNA assay and one was not detected by thetissueNGS but was detected on other platforms), the objectiveresponserate (ORR) was 54.17% as shown in Table 4(p=0.003).Comparing with third-generation EGFR-TKItreatment,radio/chemotherapy(others) significantly showedshorterPFS(p<0.001, median survival, 4.0 months, ratio, 0.25[95%CI,0.19-0.85 months] vs 10.0 months, ratio, 4[95%CI, 1.17-5.34];HR,4.49[95%CI, 1.36-14.76])and OS (p=0.058, HR,4.72[95%CI,0.61-36.57]). Similarly, T790M-negative patientssignificantlyshowed shorter PFS than patients with T790Mmutation(p=0.010, median survival, 5.0 months, ratio,0.5[95% CI,0.23-1.08] vs 10.0 months, ratio, 2.5[95%CI, 0.92-4.33];HR,2.37[95%CI, 0.92-6.09]), while OS was not statisticallysignificant(Table 4). There was significant difference in theresponse ratebetween the T790M-positive and T790M-negative patients(57.14% vs7.69%, respectively; p= 0.004, Chi-squaredtest).Furthermore, radio/chemotherapy (others) wassignificantlyassociated with poor PFS and OS in the multivariateanalysis(p=0.073 and p=0.049; age, sex, historyofsmoking, type of treatment, EGFR status, EGFR T790M mutation,andconcomitant mutations were entered into the multivariableCoxproportional hazards regression model) (Table 4).Interestingly,concomitant genetic alterations were significantlyassociated witha poor PFS for patients receiving third-generationEGFR-TKIs withT790M mutation as shown in Figure 5(p=0.0374, mediansurvival, 13.0 months, ratio,1.37 [95%CI, 0.39-4.76] vs 9.5months, ratio, 0.73[95%CI, 0.21-2.54]; HR,0.33 [95%CI,0.12-0.87]).

Resistance mechanisms to third-generationEGFR-TKItreatment

After third-generation EGFR-TKI treatment, 22 patientspresentedPD and two patients presented SD. The analysis ofputativemechanisms of resistance in these patients showed thattheactivation of known by-pass signaling pathways was observedinseven EGFR T790M-positive patients with PD, including threewithPIK3CA mutation, two with MET amplification, one with PTENdeletionand one with STK11 mutation, excepting one with KRASactivatingmutation in EGFR T790M-negative patients. Nineprogressive patientspresented other concomitant geneticalterations, including TP53,RB1, NOTCH1, FANCA, CTNNB1, and BRCA1.Notably, both patientswithout disease progression had co-mutationsin the TP53. Moreover,in five patients (30%) (#1, #3, #6, #7, #40),no mechanisms ofresistance to third-generation EGFR-TKIs weredetected (Table5).

EGFR mutation has long been considered as the mostimportantprognostic factor in patients with advanced NSCLC in theEGFR-TKIera. And EGFR-TKIs were more effective in treating advancedNSCLCthan chemotherapy. However, as resistance develops, re-biopsywillplay a significant role in guiding the use ofthird-generationEGFR-TKIs. Also, it provides a better understandingof theunderlying resistance mechanisms for third-generationEGFR-TKIs tooptimize clinical outcomes. Therefore, we prospectivelyenrolled 39patients to study the status of re-biopsy andretrospectivelyanalyzed 34 EGFR-mutant patients, who were diagnosedwith advancedNSCLC after failure to first-line EGFR-TKItreatment.

A total of 34 pairs of tissue-to-blood paired specimensweresuccessfully biopsied, and histopathological typing andgenedetection of tissue and blood were performed with a successrate of91.89% (34/37). The success rate reported in previousbiopsystudies ranged from 73% to 95%, and our results are similartoprevious studies [13-16]. The reason for our higher successratemay be that we performed more than three biopsy times perlesion.For the three cases that failed to obtain enough tumortissues forgene detection, the puncture sites were all primary lunglesions,and the pathological characteristics suggested that bloodclots andsome scattered lung tissues were presented under themicroscope,but no tumor cells were observed. This may be due to thefact thatthe lesion contains more fibrous tissue components ornecrosis andblood clots after target treatment, which may increasethedifficulty in the process of re-biopsy. Even if someheterotypiccells may exist in the on-site cytology, the number andproportionof tumor cells cannot be guaranteed. This also indicatesthedifficulty of re-biopsy after the first-generationEFGR-TKIresistance. Another reason for the lack of tumor cells inthere-biopsy samples may be due to the heterogeneity of thetumortissue. Previous studies have shown that larger differencescanoccur between primary tumors and their metastases or tumorsofdifferent pathological subtypes of the same tissue [17]. Xie etal.proved that there was no significant difference in PFSbetweenpatients with partially matched mutational profiles betweenprimarytumors and metastatic lymph nodes and those having 100%concordancerate. It indicated that the genetic profiles of bothprimarylesions and metastatic lymph nodes could be a guidance ofNSCLCtargeted therapy. As well as other studies, it was confirmedthatthe genetic mutations in the primary lung and metastasesweresimilar in patients with advanced lung cancer[18-20].Unfortunately, we were unable to confirm another re-biopsyin thesethree patients, so we were not sure whether we couldobtainadequate, effective and qualified tissue specimensforhistopathological typing and genetic testing. However, it maybeconsidered that if the primary lung lesion biopsy fails to obtainaqualified specimen, other metastatic lesions and metastaticlymphnodes may be considered for genetic testing to guidefollow-uptreatment.

In this study, we detected and identified potentiallyactionablealterations using both blood-derived ctDNA and tumortissue bydifferent platforms, especially for EGFR T790M mutation.Comparedwith traditional detection methods, NGS technology hasbetterdetection efficiency and higher throughput through tissue andbloodsamples. It can detect mutations, insertions, rearrangementsandcopy number variations quantitatively at the same time, andalsosignificantly save cost and time in large-scalesequencing.Concordance rates between tissue and ctDNA NGS appearedhigh here,although there were differences in genomic alterationsdetected.However, compared with the efficacy of genetic testing fortissuebiopsy specimens, the efficacy of blood test for T790M stillhasmore false negatives, resulting in lower sensitivity.Meanwhile,there was also one patient with T790M mutation in plasmabutnegative in tissue. We combined the efficacy with the testresultsto consider that the tissue test results were also falsenegative,suggesting that tissue and ctDNA assay providedcomplementaryresults. Therefore, domestic and foreign guidelinesrecommend thatwhen the patient's physical condition is tolerableand tissuespecimens are available, tissue biopsy should bepreferred as thefirst choice for re-biopsy after EGFR-TKIresistance. When tissuespecimens are not available or tissuespecimens are unqualified andpatients refuse to perform, in thecase of a re-biopsy, liquidbiopsy techniques can be a powerfulcomplement.

For genetic alterations and resistance mechanisms,weretrospectively analyzed 34 patients with mutations detectedbyboth ctDNA and tissue NGS. Among them, 88.23% (30/34) ofpatientsharbored concomitant mutations detected by at least one NGSassay.And co-occurring genetic alterationswerenegatively correlated with the response to thetreatment,which may be plausibly due to the bypass activation ofsurvivalsignaling pathways or tumor heterogeneity. Consistent withrecentstudies [12], patients with 19del survived longer thanpatientswith 21L858R mutation. Of interest, a higher incidenceofconcomitant mutations in patients was also detected with21L858R,although there was no significant difference which was notshown inthe results. As we know, the presence of EGFR T790Mmutation inresistant patients after first-line EGFR-TKI therapywassignificantly associated with better efficacy ofthird-generationEGFR-TKIs, which provides a rationale for thesuperiority ofthird-generation EGFR-TKI therapy over othertherapies. Inmultivariable analysis, the EGFR T790M mutation wasstillsignificantly associated with survival.

To study the potential mechanisms of resistance, re-biopsyatprogression to third-generation EGFR-TKIs was performed in24patients. Activation of known by-pass signaling pathwaysasmechanism of resistance was represented, and one patient withSTK11mutation, one with PTEN deletion, one with KRAS mutation, twowithMET amplification and three with PIK3CA mutation wereobserved.STK11 mutation presented accompanying with TP53 mutationin an EGFRT790M-positive patient, which had been reported as amediator ofthe cold tumor immune microenvironment and a majordriver ofprimary resistance to PD-1 axis inhibitors in non-squamouslungadenocarcinoma [21]. PTEN loss was previously described asamechanism of resistance to first-generation EGFR-TKIs [22].WhileKim et al [23]. reported a following increase of theproportion oftumors with PTEN deletions in post-treatment tumorsand the gradualincrease of PTEN deletions might contribute to focalprogression toOsimertinib. MET amplification was already mentionedas a possiblemechanism of resistance to Osimertinib, which has beenconsideredas the very common findings of acquired resistanceunderfirst-generation EGFR-TKIs [24-26], and described in theliteratureat frequencies ranging from 5% to 50% [27, 28].Particularly,activating mutations of the catalytic subunit alpha(PIK3CA) ofPI3K lipid kinases family was associated with poor PFSin ourcohort including a case of small cell lung cancer(SCLC)transformation, since activation of PI3K/AKT/mTOR signalingpathwaywas not mutually exclusive with other carcinogenicdrivingmechanisms. The shorter median survival time in patientswithco-existing PIK3CA and EGFR mutations suggested apossiblesynergistic effect due to stronger activation ofrelevantdownstream signals [29, 30]. The results suggested apotential roleof PIK3CA-inhibitor, alone or in combination, toaccuratelyovercome this resistance.

The main limitations of this study were sample size,singlecenter design and lack of continuous biopsy. Furthermore,data fromour NGS hotspots (excluding other mutations, copy numberchanges,or chromosomal abnormalities) may represent anunderestimation ofconcomitant mutations and impede further analysisof signalingpathways or cloning.

In conclusion, using multi-platforms to perform detection ofEGFRT790M mutation on EGFR-TKI resistant re-biopsy tissue andbloodsamples is feasible, and the consistency between tissue andbloodsamples is high. They can provide complementary resultsmutually.Our study also highlights the importance of re-biopsy andmoleculardiagnosis during disease progression in patients treatedbyEGFR-TKIs for oncogene-addicted NSCLC. And concomitantgeneticalterations may affect response to treatment and decisionofsequential therapy strategy.

Ethics approval and consenttoparticipate

The study was approved by the Ethics Committee of ShanghaiChestHospital, and the ethical approval number was KS1703. Allpatientswere fully informed and signed informed consent. Theclinical trialregistration was carried out onClinicalTrials.gov.(NCT03309462).

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current studyareavailable from the corresponding author on reasonablerequest.

Competing Interests

The authors declare that they have no competing interests.

Funding

This work was supported by National Key R&D ProgramofChina (2017YFC0112700); Shanghai Municipal HealthandMedical Talents Training Program(2018BR09); ShanghaiMunicipal Education Commission-GaofengClinical Medicine GrantSupport (20181815).

Authors’ Contributions

J. S. and L. X. contributed to the study design,subjectsenrolled, acquisition of specimens, data analysis andrevising themanuscript. Y. Z. contributed to the data collection,analysis andinterpretation, drafting and revising the manuscript.F. X.contributed to analysis and revising the manuscript. X. Z. andY.L. contributed to the acquisition of specimens and reviewingthemanuscript. L. Z. contributed to the pathological analysisandreviewing the manuscript. All authors read and approved thefinalmanuscript.

Acknowledgements

We would like to express our gratitude to Dr. Xueyan Zhang,Dr.Huimin Wang and Dr. Yuqing Lou, who have offered usvaluableresource and suggestions during this study. We thankShanghaiTongshu Biotechnology Co., Ltd. for technical support.

ARMS-PCR: amplification refractory mutationsystem-polymerasechain reaction;

CR: complete response;

ctDNA: circulating tumor DNA;

ddPCR: digital droplet PCR;

ECOG: Eastern Cooperative Oncology Group;

EGFR: epidermal growth factor receptor;

EGFR-TKI: epidermal growth factor receptor-tyrosinekinaseinhibitor;

NGS: next-generation sequencing;

NSCLC: non-small cell lung cancer;

ORR: objective response rate;

OS: overall survival;

PD: progressive disease;

PFS: progression-free survival;

PR: partial response;

PS: performance status;

RECIST: Response Evaluation Criteria in Solid Tumors;

ROSE: rapid on-site evaluation;

SD: stable disease;

TBB: transbronchial biopsy;

TBLB: transbronchial lung biopsy;

TBNA: transbronchial needle aspiration;

TTNA: transthoracic needle aspiration.

Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, JemalA.Global cancer statistics 2018: GLOBOCAN estimates of incidenceandmortality worldwide for 36 cancers in 185 countries. CA CancerJClin. 2018;68:394-424. https://doi.org/10.3322/caac.21492
Govindan R, Page N, Morgensztern D, et al. Changingepidemiologyof small-cell lung cancer in the United States overthe last 30years: analysis of the surveillance, epidemiologic, andend resultsdatabase. J Clin Oncol.2006;24:4539-4544.https://doi.org/10.1200/JCO.2005.04.4859
Lindeman NI, Cagle PT, Beasley MB, et al. Moleculartestingguideline for selection of lung cancer patients for EGFR andALKtyrosine kinase inhibitors: guideline from the College ofAmericanPathologists, International Association for the Study ofLungCancer, and Association for Molecular Pathology. J ThoracOncol.2013;8:823-859.https://doi.org/10.1097/JTO.0b013e318290868f
Fang S, Wang Z. EGFR mutations as a prognostic andpredictivemarker in non-small-cell lung cancer. Drug Des DevelTher.2014;8:1595-1611. https://doi.org/10.2147/DDDT.S69690
Jackman D, Pao W, Riely GJ, et al. Clinical definitionofacquired resistance to epidermal growth factor receptortyrosinekinase inhibitors in non-small-cell lung cancer. J ClinOncol.2010;28:357-360.https://doi.org/10.1200/JCO.2009.24.7049
Cross DAE, Ashton SE, Ghiorghiu S, et al. AZD9291,anirreversible EGFR TKI, overcomes T790M-mediated resistance toEGFRinhibitors in lung cancer. Cancer Discov.2014;4:1046-1061.https://doi.org/10.1158/2159-8290.CD-14-0337
Attili I, Karachaliou N, Conte P, Bonanno L, RosellR.Therapeutic approaches for T790M mutation positivenon-small-celllung cancer. Expert Rev Anticancer Ther.2018;18:1021-1030.https://doi.org/10.1080/14737140.2018.1508347
National Comprehensive Cancer Network. NCCN clinicalpracticeguidelines in oncology. Non-small cell lungcancer.https://www.nccn.org/professionals/physician_gls/default.aspx#nscl;2020Accessed 29 February, 2020.
Tetsu O, Hangauer MJ, Phuchareon J, Eisele DW, McCormick F.DrugResistance to EGFR Inhibitors in Lung Cancer.Chemotherapy.2016;61:223-235.https://doi.org/10.1159/000443368
Hata A, Katakami N, Yoshioka H, et al. SpatiotemporalT790MHeterogeneity in Individual Patients withEGFR-MutantNon-Small-Cell Lung Cancer after Acquired Resistance toEGFR-TKI. JThorac Oncol.2015;10:1553-1559.https://doi.org/10.1097/jto.0000000000000647
Blakely CM, Watkins TBK, Wu W, et al. Evolution andclinicalimpact of co-occurring genetic alterations inadvanced-stageEGFR-mutant lung cancers. Nat Genet.2017;49:1693-1704.https://doi.org/10.1038/ng.3990
Hong S, Gao F, Fu S, et al. Concomitant Genetic AlterationsWithResponse to Treatment and Epidermal Growth Factor ReceptorTyrosineKinase Inhibitors in Patients With EGFR-Mutant AdvancedNon–SmallCell Lung Cancer. JAMA Oncol.2018;4:739-742.https://doi.org/10.1001/jamaoncol.2018.0049
Nosaki K, Satouchi M, Kurata T, et al. Re-biopsy statusamongnon-small cell lung cancer patients in Japan: Aretrospectivestudy. Lung Cancer.2016;101:1-8.https://doi.org/10.1016/j.lungcan.2016.07.007
Kirita K, Izumo T, Matsumoto Y, Hiraishi Y, TsuchidaT.Bronchoscopic Re-biopsy for Mutational Analysis of Non-smallCellLung Cancer. Lung.2016;194:371-378.https://doi.org/10.1007/s00408-016-9864-5
Chouaid C, Dujon C, Do P, et al. Feasibility and clinicalimpactof re-biopsy in advanced non small-cell lung cancer: aprospectivemulticenter study in a real-world setting (GFPC study12-01). LungCancer.2014;86:170-173.https://doi.org/10.1016/j.lungcan.2014.08.016
Bosc C, Ferretti GR, Cadranel J, et al. Rebiopsy duringdiseaseprogression in patients treated by TKI for oncogene-addictedNSCLC.Target Oncol.2015;10:247-253.https://doi.org/10.1007/s11523-014-0332-y
McGranahan N, Swanton C. Clonal Heterogeneity andTumorEvolution: Past, Present, and the Future. Cell.2017;168:613-628.https://doi.org/10.1016/j.cell.2017.01.018
Xie F, Zhang Y, Mao X, et al. Comparison of geneticprofilesamong primary lung tumor, metastatic lymph nodes andcirculatingtumor DNA in treatment-naive advanced non-squamousnon-small celllung cancer patients. Lung Cancer.2018;121:54-60.https://doi.org/10.1016/j.lungcan.2018.05.002
Sherwood J, Dearden S, Ratcliffe M, Walker J. Mutationstatusconcordance between primary lesions and metastatic sitesofadvanced non-small-cell lung cancer and the impact ofmutationtesting methodologies: a literature review. J Exp ClinCancer Res.2015;34:92.https://doi.org/10.1186/s13046-015-0207-9
Wang S, Wang Z. Meta-analysis of epidermal growthfactorreceptor and KRAS gene status between primary andcorrespondingmetastatic tumours of non-small cell lung cancer. ClinOncol (RColl Radiol).2015;27:30-39.https://doi.org/10.1016/j.clon.2014.09.014
Skoulidis F, Goldberg ME, Greenawalt DM, et al.STK11/LKB1Mutations and PD-1 Inhibitor Resistance in KRAS-MutantLungAdenocarcinoma. Cancer Discov.2018;8:822-835.https://doi.org/10.1158/2159-8290.CD-18-0099
Sos ML, Koker M, Weir BA, et al. PTEN loss contributestoerlotinib resistance in EGFR-mutant lung cancer by activationofAkt and EGFR. Cancer Res.2009;69:3256-3261.https://doi.org/10.1158/0008-5472.CAN-08-4055
Kim TM, Song A, Kim DW, et al. Mechanisms of AcquiredResistanceto AZD9291: A Mutation-Selective, Irreversible EGFRInhibitor. JThorac Oncol.2015;10:1736-1744.https://doi.org/10.1097/JTO.0000000000000688
Sequist LV, Waltman BA, Dias-Santagata D, et al. Genotypicandhistological evolution of lung cancers acquiring resistance toEGFRinhibitors. Sci Transl Med.2011;3:75ra26.https://doi.org/10.1126/scitranslmed.3002003
Yu HA, Arcila ME, Rekhtman N, et al. Analysis of tumorspecimensat the time of acquired resistance to EGFR-TKI therapy in155patients with EGFR-mutant lung cancers. Clin CancerRes.2013;19:2240-2247.https://doi.org/10.1158/1078-0432.ccr-12-2246
Arcila ME, Oxnard GR, Nafa K, et al. Rebiopsy of lungcancerpatients with acquired resistance to EGFR inhibitors andenhanceddetection of the T790M mutation using a locked nucleicacid-basedassay. Clin Cancer Res.2011;17:1169-1180.https://doi.org/10.1158/1078-0432.ccr-10-2277
Oxnard GR, Hu Y, Mileham KF, et al. Assessment ofResistanceMechanisms and Clinical Implications in Patients WithEGFRT790M-Positive Lung Cancer and Acquired Resistance toOsimertinib.JAMA Oncol.2018;4:1527-1534.https://doi.org/10.1001/jamaoncol.2018.2969
Lin CC, Shih JY, Yu CJ, et al. Outcomes in patientswithnon-small-cell lung cancer and acquired Thr790Met mutationtreatedwith osimertinib: a genomic study. Lancet RespirMed.2018;6:107-116.https://doi.org/10.1016/S2213-2600(17)30480-0
Chaft JE, Arcila ME, Paik PK, et al. Coexistence of PIK3CAandother oncogene mutations in lung adenocarcinoma-rationaleforcomprehensive mutation profiling. Mol Cancer Ther.2012;11:485-491.https://doi.org/10.1158/1535-7163.MCT-11-0692
Ludovini V, Bianconi F, Pistola L, etal.Phosphoinositide-3-Kinase Catalytic Alpha and KRAS MutationsareImportant Predictors of Resistance to Therapy with EpidermalGrowthFactor Receptor Tyrosine Kinase Inhibitors in PatientswithAdvanced Non-small Cell Lung Cancer. J ThoracOncol.2011;6:707-715.https://doi.org/10.1097/JTO.0b013e31820a3a6b

Table 1
Initial biopsy baselineinformation (n = 39).
Characteristics	Number
Gender, n (%)
Male	21 (53.85%)
Female	18 (46.15%)
Age, median (range),years	61 (38–79)
Smoking history, n(%)
Yes	12 (30.77%)
No	27 (69.23%)
Pathology, n (%)
Adenocarcinoma	38 (97.44%)
adenosquamouscarcinoma	1 (2.56%)
Molecular pathology, n(%)
EGFR 19del	26 (66.67%)
EGFR 21L858R	13 (33.33%)
Treatment, n (%)
First-generationEGFR-TKI	36 (92.31%)
Chemotherapy + first-generationEGFR-TKI	3 (7.69%)
PFS^*, median (range), months	13.0 (1-100)
Sites, n (%)
Lung lesions	26 (66.67%)
Intrathoracic metastatic lymphnodes	3 (7.69%)
Extrathoracic metastatic lymphnodes	6 (15.38%)
Pleural effusion	4 (10.26%)
Methods of initial biopsy, n(%)
TBB	10 (25.64%)
EBUS-TBNA	3 (7.69%)
TBLB	1 (2.56%)
TTNA	10 (25.64%)
Percutaneous needle aspirationof lymph nodes	6 (15.38%)
Surgery	5 (12.82%)
Pleural effusion	4 (10.26%)
^*PFS was determined from the startingdate of the first-generation EGFR-TKI treatment to the date ofdisease progression.
Abbreviations: EBUS-TBNA,endobronchial ultrasound transbronchial needle aspiration; EGFR,epidermal growth factor receptor; EGFR-TKI, epidermal growth factorreceptor- tyrosine kinase inhibitor; PFS, progression-freesurvival; TBB, transbronchial biopsy; TBLB, transbronchial lungbiopsy; TTNA, transthoracic needle aspiration

Table2
Re-biopsy baseline information (n = 37).
Characteristics	Number
Pathology, n (%)
Adenocarcinoma	30 (81.08%)
Squamous cell lungcancer	3 (8.11%)
Small cell lungcancer	1 (2.7%)
Inadequate tumorcells	3 (8.11%)
Clinical staging, n(%)
IIIA	3 (8.11%)
IIIB	2 (5.41%)
IV	32 (86.49%)
Metastasis, n
Bone metastasis	11 (29.73%)
Brain metastasis	16 (43.24%)
• Pleural metastasis	8 (21.62%)
Other distantmetastases	18 (48.65%)
Treatment, n (%)
Third-generationEGFR-TKI	27 (72.97%)
Radio/chemotherapy	10 (27.03%)
PFS^*, median (range), months	6 (0–23)
OS, median (range),months	13 (0–23)
Sites, n (%)
Lung lesions	23 (62.16%)
Intrathoracic metastatic lymphnodes	8 (21.62%)
Extrathoracic metastatic lymphnodes	6 (16.22%)
Methods of re-biopsy, n(%)
TBB	9 (24.32%)
EBUS-TBNA	8 (21.62%)
TBLB	6 (16.22%)
TTNA	8 (21.62%)
Percutaneous needle aspirationof lymph nodes	6 (16.22%)
^*PFS was determined from the startingdate of the third-generation EGFR-TKI treatment and/or othertreatments to the date of disease progression, or the lastfollow-up time for those who have not reached diseaseprogression.
Abbreviations: EBUS-TBNA,endobronchial ultrasound transbronchial needle aspiration;EGFR-TKI, epidermal growth factor receptor- tyrosine kinaseinhibitor; OS, overall survival; PFS, progression-free survival;TBB, transbronchial biopsy; TBLB, transbronchial lung biopsy; TTNA,transthoracic needle aspiration.

Table3
Patients’ clinical characteristics for retrospective analysis.
Characteristic	NO. (%)
Age, mean ± SD, years	58.5 ± 9.31
Sex
Male	17 (50%)
Female	17 (50%)
Smoking history
Never smoker	27 (79.41%)
Smoker	7 (20.59%)
Clinical stage
Ⅲ	5 (14.71%)
Ⅳ	29 (85.29%)
T stage
T1	1 (2.94%)
T2	22 (64.71%)
T3	2 (5.88%)
T4	9 (26.47%)
Lymph node metastasis
N0	1 (2.94%)
N2	16 (47.06%)
N3	17 (50%)
Distant metastasis
No	5 (14.71%)
Yes	29 (85.29%)
Metastatic sites
Brain	10 (29.41%)
Bone	17 (50%)
Liver	2 (5.88%)
Adrenal	4 (11.76%)
Pericardium	2 (5.88%)
ECOG PS
0	12 (35.29%)
1	22 (64.71%)
Baseline pathologicalclassification
Adenocarcinoma	33 (97.06%)
Adenosquamouscarcinoma	1 (2.94%)
Re-biopsy pathologicalclassification
Adenocarcinoma	30 (88.24%)
Squamous carcinoma	3 (8.82%)
Small cell lungcancer	1 (2.94%)
EGFR mutation
19del	23 (67.65%)
L858R	11 (32.35%)
First line EGFR-TKI
gefitinib	8 (23.53%)
erlotinib	3 (8.82%)
icotinib	21 (61.76%)
others	2 (5.88%)
Post-TKI treatment
Third-generationEGFR-TKI	24 (70.59%)
Radio/chemotherapy	10 (29.41%)
Abbreviations: ECOG PS: EasternCooperative Oncology Group performance status; EGFR, epidermalgrowth factor receptor; EGFR-TKI, epidermal growth factor receptor-tyrosine kinase inhibitor

Table4
Treatment outcomes for patients with EGFR-mutant advanced non-smallcell lung cancer treated with EGFR TKIs.
Variable		NO. (%)	Objective Response		Progression-FreeSurvival				Overall Survival
					Univariate		Multivariable*		Univariate		Multivariable*
			NO. (%)	P Value	HR (95% CI)	P Value	HR (95% CI)	P Value	HR (95% CI)	P Value	HR (95% CI)	P Value
Post-TKI treatments
	Third-generationEGFR-TKIs	24(70.59%)	13(54.17%)	0.003	1(Reference)	< 0.0001	1(Reference)	0.073	1(Reference)	0.058	1(Reference)	0.049
	Radio/Chemotherapy	10(29.41%)	0(0.0%)	0.003	4.49(1.36–14.76)	< 0.0001	2.98(0.91–9.82)	0.073	4.72(0.61–36.57)	0.058	15.93(1.01-250.57)	0.049
EGFR status^#
	Exon 19 deletion	23(67.65%)	9 (39.13%)	0.877	1(Reference)	0.272	1(Reference)	0.727	1(Reference)	0.874	1(Reference)	0.740
	Exon 21 mutation	11(32.35%)	4 (36.36%)	0.877	1.46(0.66–3.23)	0.272	0.85(0.34–2.13)	0.727	1.15(0.19–7.12)	0.874	1.38(0.2–9.42)	0.740
EGFR T790M mutation^#
	Positive	21(61.76%)	12(57.14%)	0.004	1(Reference)	0.010	1(Reference)	0.017	1(Reference)	0.584	1(Reference)	0.324
	Negative	13(38.24%)	1(7.69%)	0.004	2.37(0.92–6.09)	0.010	4.92(1.33–18.16)	0.017	0.44(0.07–2.72)	0.584	0.3(0.03–3.33)	0.324
Concomitant mutation^#
	Yes	24(70.59%)	9(37.5%)	0.891	1(Reference)	0.479	1(Reference)	0.245	1(Reference)	0.553	1(Reference)	0.706
	No	10(29.41%)	4(40%)	0.891	0.78(0.37–1.64)	0.479	0.58(0.23–1.46)	0.245	0.53(0.08–3.44)	0.553	1.50(0.18–12.5)	0.706
Abbreviations: EGFR, epidermalgrowth factor receptor; TKI, tyrosine kinase inhibitor
*Age (stratified by65 years old), sex, history of smoking, type of TKI, EGFRstatus, EGFR T790M mutation, and concomitant mutations were enteredinto the multivariable Cox proportional hazards regressionmodel.
# Based on tissue NGSresults

Table5
Results of performed tissue re-biopsies after progression tothird-generation EGFR-TKIs.

patients	Mutations	Best Response	progression	PFS	patients	Mutations	Best Response		progression		PFS
#1	n.d	PR	yes	17	#7	n.d		SD		yes	3
#3	n.d	PR	yes	12	#22	TP53, KRAS		SD		yes	6
#4	c-MET	PD	yes	0	#38	TP53		SD		yes	6
#5	PTEN	SD	yes	3
#6	n.d	SD	yes	13
#10	TP53, STK11	PR	yes	9
#11	PIK3CA, NOTCH1	SD	yes	11
#12	TP53, RB1	PR	yes	14
#14	BRCA1	PR	yes	10
#15	PIK3CA, TP53	PR	yes	5
#16	FANCA, CTNNB1	SD	yes	12
#18	TP53	SD	yes	13
#20	TP53	PR	no	12
#24	PIK3CA, FANCA, CTNNB1	PR	yes	7
#26	TP53	PR	yes	11
#28	TP53	SD	yes	9
#30	TP53	PR	yes	10
#32	TP53, RB1	SD	yes	5
#34	c-MET, TP53	PR	yes	6
#36	TP53	PR	no	9
#40	n.d	PR	yes	7
Abbreviations: EGFR, epidermalgrowth factor receptor; n.d, no detection; PD, progressive disease;PFS, progression-free survival; PR, partial response; SD, stabledisease

Download PDF

Version 2

posted

You are reading this older preprint version

Read the latest preprint version →

Next-generation sequencing of tissue and circulating tumor DNA: Resistance mechanisms to EGFR targeted therapy in a cohort of patients with advanced non-small cell lung cancer

Status:

Version 2

Abstract

Figures

Background

Methods

Results

Discussion

Conclusions

Declarations

Abbreviations

References

Tables

Status:

Version 2