- Email: [email protected]
Clinical analysis by next-generation sequencing for NSCLC patients with MET amplification resistant to osimertinib
Accepted Manuscript Title: Clinical analysis by next-generation sequencing for NSCLC patients with MET amplification resistant to osimertinib Authors: Yubo Wang, Li Li, Rui Han, Lin Jiao, Jie Zheng, Yong He PII: DOI: Reference:
S0169-5002(18)30262-9 https://doi.org/10.1016/j.lungcan.2018.02.007 LUNG 5573
To appear in:
Lung Cancer
Received date: Revised date: Accepted date:
17-11-2017 8-2-2018 9-2-2018
Please cite this article as: Wang Yubo, Li Li, Han Rui, Jiao Lin, Zheng Jie, He Yong.Clinical analysis by next-generation sequencing for NSCLC patients with MET amplification resistant to osimertinib.Lung Cancer https://doi.org/10.1016/j.lungcan.2018.02.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Clinical analysis by next-generation sequencing for NSCLC patients with MET amplification resistant to osimertinib
SC R
IP T
Running title: Cohort study of MET amplification for osimertinib resistance
Authors and affiliations
A
Department of Respiratory Medicine, Daping Hospital, Third Military Medical
M
1
N
U
Yubo Wang1, Li Li1, Rui Han1, Lin Jiao1, Jie Zheng1, Yong He1
TE D
University, Chongqing, P.R. China
EP
Corresponding Author:
CC
Yong He
Department of Respiratory Medicine, Daping Hospital, Third Military Medical
A
University, 10# Changjiangzhilu Daping, Yuzhong District, 40042, Chongqing, P.R. China. Phone:86-02368757791; Fax:86-02368700970; E-mail: [email protected]
Highlights
MET amplification is a common mechanism of osimertinib resistance. MET amplification-induced osimertinib resistance might occur in the early phase. Profiling using NGS can disclose the underlying resistance mechanism. EGFR-TKI plus crizotinib are effective in patients with MET amp-induced resistance
IP T
SC R
Abstract
Introduction: The efficacy of osimertinib was compromised by the development of
U
resistance mechanisms, such as MET amplification. However, cohort studies of
N
osimertinib resistance mechanism, and the correlation of MET and progression-free
M
A
survival (PFS) after osimertinib resistance have been poorly investigated.
Objectives: This study was carried out to study the acquired MET amplification after
TE D
osimertinib resistance in advanced lung adenocarcinoma patients, and interrogate the correlation of clinical prognosis and MET amplification.
EP
Methods: We performed capture-based sequencing on longitudinal plasma and tissue
CC
samples obtained before osimertinib treatment and after resistance development from lung adenocarcinoma patients to investigate the underlying resistance mechanism. We
A
also investigated the correlation of MET amplification and patient prognosis after osimertinib resistance using Kaplan-Meier analysis.
Results: Paired biopsies before osimertinib treatment and after the resistance development revealed underlying resistance mechanisms. In addition, a cohort of 13
patients who developed disease progression after osimertinib resistance was investigated. Patients with MET amplification after osimertinib resistance commonly had inferior median progression-free survival (mPFS) than patients without MET amplification appearance or increase (3.5 months vs. 9.9 months, p=0.117). Patients in
IP T
MET amplification group also displayed poor median overall survival (mOS) compared to MET amplification negative group (15.6 months vs. 30.7 months,
SC R
p=0.885). Furthermore, combinatorial treatment of first/third-generation EGFR-TKI
and crizotinib was efficaciously administrated into two patients with newly acquired
U
MET amplification after osimertinib resistance. Partial responses were achieved by
A
N
them, both clinically and radiographically.
M
Conclusions: We investigated the osimertinib resistance mechanism in a small cohort of lung adenocarcinoma patients, and demonstrated MET amplification was correlated
TE D
with inferior PFS/OS after osimertinib treatment. Moreover, we reported the first clinical evidence of efficacy generated by combination of first-generation EGFR-TKI
CC
EP
icotinib and crizotinib after the resistance to osimertinib.
A
Key words:MET amplification, resistance, osimertinib, crizotinib
Introduction
Osimertinib, a third-generation epidermal growth factor receptor (EGFR)
tyrosine kinase inhibitor (TKI), benefits patients with T790M mutant non-small-cell lung cancer (NSCLC) who fail in treatment with first-generation EGFR-TKIs.1,
2
However, drug resistance limits the clinical application of osimertinib. At present, mechanisms of osimertinib resistance have been investigated in several studies. The
IP T
most common mechanism of osimertinib resistance was EGFR C797S mutation.3 MET amplification is another mechanism for acquired resistance to third generation
SC R
EGFR-TKIs.4 In addition, BRAF V600E mutation,5 HER2 amplification,6 EGFR amplification,7 and small-cell histological transformation8 were also been reported to
U
result in third-generation EGFR-TKI resistance.
A
N
MET is a tyrosine kinase receptor located at 7q21-q31. MET amplification is a
M
common gene alteration which occurs in about 1%~20% of NSCLC and associates with poor prognosis.9 MET amplification has been reported in approximately 5% of
TE D
resistant patients to first generation EGFR-TKI, such as erlotinib and gefitinib.10 In recent decades, a small number of cases have reported that MET amplification may
EP
also be induced after osimertinib treatment, suggesting that MET amplification can also be responsible for osimertinib resistance.6 However, cohort studies of osimertinib
CC
resistance mechanism, especially MET-associated, have been poorly investigated.
A
In this study, a retrospective investigation was performed by profiling plasma
and tissue biopsies from 13 patients with acquired resistance to osimertinib using NGS. Paired biopsies obtained before osimertinib treatment and after resistance development in 6 patients were analyzed for mutation spectrum and novel putative resistance mutations were identified. Furthermore, we provided two efficacious
clinical evidences of combinatorial therapy of EGFR-TKI and crizotinib to illustrate the personalized therapy design guided by dynamic ctDNA monitoring.
Patients and methods
IP T
Patient Information
Thirteen patients were collected and evaluated at our Hospital. This study was
SC R
approved by the Institutional Review Board at the Chongqing Daping Hospital. All patients were provided informed consent to this study and gave permission to the
U
entire study. Three sequencing panels, consisting of 168, 180 or 365 cancer-related
N
genes, were developed and validated for somatic variations identification.
M
A
Preparation of DNA of liquid and tissue samples
TE D
For cfDNA preparation of 168 cancer-related gene panel, 10 mL of peripheral blood was obtained and stored in tubes containing ethylenediaminetetraacetic acid, followed by incubating at room temperature for 2 hours. Circulating cfDNA was
EP
recovered from 4 to 5mL of plasma by using the QIAamp Circulating Nucleic Acid
CC
kit (Qiagen). Quantification of cfDNA was assessed with the Qubit 2.0 fluorimeter (Thermo Fisher Scientific). A minimum of 50 ng of cfDNA is required for
A
construction of an NGS library. For 365 cancer-related genes panel sequencing, genomic DNA in Formalin Fixed Paraffin Embedded (FFPE) sections was extracted with QIAamp DNA FFPE Tissue Kit (Qiagen), while cell-free circulating DNAs in plasma were extracted by QIAamp Circulating Nucleic Acid Kit (Qiagen) following the standard protocols. DNA was quantified by PicoGreen fluorescence assay
(Invitrogen). For 180 cancer-related genes panel sequencing, FFPE samples were extracted using QIAamp DNA FFPE Tissue Kit (Qiagen, Hilden, Germany). DNA concentration was measured using a Qubit fluorometer (Invitrogen, Carlsbad, CA USA) and the Qubit dsDNA HS (High Sensitivity) Assay Kit.
IP T
NGS sequencing of liquid biopsies
SC R
For liquid biopsies sequencing of 168 cancer-related gene panel, fragments with a size of 200-400 base pairs (bp) were selected by using Agencourt AMPure beads
U
(Beckman Coulter, Fullerton, CA). Hybridization was performed with capture probes
N
baits, followed by hybrid selection with magnetic beads, and PCR amplification. A
A
high-sensitivity DNA assay was performed to assess the quality and size of the
M
fragments, and indexed samples were sequenced on a Nextseq500 sequencer
TE D
(Illumina, Inc., San Diego, CA) with paired-end reads. Sequence data were mapped to the human genome (hg19) using BWA aligner.
For 365-cancer related gene panel sequencing, construction of sequencing
EP
libraries 50-200 ng of DNAs was fragmented to around ~200 bp by sonication
CC
(Covaris), and constructed into the libraries with KAPA Hyper Prep Kit (Kapa
A
Biosystems) for sequencing.
For 180-cancer related gene panel sequencing, library construction was prepared
according to IlluminaTruSeq DNA Library Preparation Kit (Illumina, San Diego, CA) using 1g of DNA sheared by an ultrasonoscope to generate fragments with a peak of 250 bps, followed by end repair, A tailing and ligation. Libraries were hybridized to
custom-designed biotinylated oligonucleotide probes (Roche NimbleGen, Madison, WI, USA) and DNA sequencing was carried out with the HiSeq 3000 Sequencing System (Illumina, San Diego, CA) with 2×100-bp paired-end reads.
IP T
Copy number variation algorithms In the panel of 168 cancer-related genes for plasma ctDNA sequencing, CNV
SC R
were identified based on depth of coverage data of capture intervals. Coverage depth data were firstly corrected for the sequencing bias due to GC content and target probe
U
density. The average coverage of all capture regions was calculated as internal control,
N
which was utilized to normalize the coverage of different samples to comparable
A
scales. The coverage of MET with copy number gain would be significantly larger
M
than the internal control. The difference of adjusted coverage depth for each of genes
TE D
from plasma samples and reference was evaluated by t-test. Genes with p-value<5e-3 in the t-test and CN>2.25 were considered as amplification.
For the 365-cancer related genes panel, a statistically rigorous and
EP
computationally efficient algorithm called BIC-seq was used for detecting CNVs. We
CC
obtained a log-ratio profile of the sample by normalizing the sequence coverage obtained at all exons against a process-matched normal control. This profile was
A
corrected for GC-bias and log2ratio>0.7 was considered to be copy number gain and abstract copy number was estimated by log2ratio normalized by tumor purity.11 cfDNA fraction of 5% is required for MET amplification detection of minimum 2.5 copy. In total, for gene copy gain events whose ratios were larger than 3.5 copy, the
sensitivity of CNV detection in ctDNA achieved 83.3%.
For the 180-cancer related genes panel for tissue sequencing, somatic copy-number alterations were identified with CONTRA (v2.0.8).12 Base-level log-ratios between the case and control were taken into consideration to eliminated
IP T
GC bias and imbalanced library size effect. Target regions of <10bp with depth of
SC R
coverage <10 were discarded. For the remaining regions, the mean log-ratio was
calculated and the significant P value was then assigned. Then the circular binary segmentation was applied to the log-ratios of regions with default settings.
U
Segmentation results were then used for the subsequent analysis. GISTIC37 algorithm
A
N
was used to infer recurrently amplified or deleted genomic regions, using copy
M
numbers in 100-kb windows. G scores represented the frequency and amplitude of amplifications or deletions of each genomic region. MET copy number gain was
TE D
identified using thresholds of 2.3 copies normalized to control.13
Statistical Analysis
EP
All these survival analyses were conducted in software R with survival package.
CC
We performed Kaplan-Meier analysis to investigate the survival functions. Log-rank test was used to determine the difference of survival curve between groups. P-value of
A
< 0.05 was considered statistically significant.
Results
Patient cohort and NGS overview
To characterize potential mechanisms of resistance to third-generation EGFR-TKI osimertinib, a retrospective analysis was performed by enrolling 37 NSCLC patients from August 2015 to May 2017 in our hospital. Prior to treatment of osimertinib, all these patients displayed resistance to the first-generation of
IP T
EGFR-TKI and the classic resistant mutation EGFR T790M mutation were detected. Of these, 16 patients (43.2%) developed disease progression after osimertinib
SC R
resistance, and either plasma or tissue biopsies from 13 patients were profiled by capture-based targeted ultra-deep sequencing after osimertinib resistance. Of the
U
cohort of 13 patients, the average age was 57.7 years, ranging from 44 to 72 years. 6
N
were males and 7 were females. 3 patients had a history of smoking and 10 patients
A
were nonsmokers. All the 13 patients were classified as stage IV adenocarcinoma with
M
distant metastasis. The clinical characteristics of the cohort was summarized in Table
TE D
1.
Moreover, paired specimens (before and after osimertinib treatment) from 6 of
EP
the 13 patients were sequenced to investigate the underlying novel potential resistant mechanisms and monitor the dynamic genomic alteration status. The sequencing
CC
panels, consisting of 168, 180 and 365 critical exons and introns for genomic variants,
A
covers multiple classes of somatic mutations for detection and quantification of genetic alterations.
Potential newly acquired resistance mechanisms to osimertinib
To identify the underlying mechanisms of resistance to osimertinib, we deemed
that gene aberrations which drive resistance should generate in the process of osimertinib therapy. Therefore, we identified gene alterations that were absent before osimertinib treatment but newly emerged or increased in relative abundance after the resistance development. We performed capture-based ultradeep targeted sequencing
IP T
on paired biopsies from 6 patients before osimertinib treatment and after resistance development, consisting of 1 paired tissue sample and 5 paired plasma samples. We
SC R
identified several putative mechanisms through dynamically monitoring (Figure 1A,
1B). MET amplification was the most frequent mechanism and was observed in 4
U
patients, one with increased MET copy-number abundance and three with newly
N
emerged MET amplification. BRAF single nucleotide variant V600E, which had been
M
of the four patients.
A
reported to induce osimertinib resistance in previous studies,5 was also detected in one
TE D
Of the other two patients without MET amplification, patients 05 who originally harbored EGFR sensitizing mutation L747_P753delinsS and resistant mutation
EP
T790M newly obtained four missense variants (PTEN R346C, NTRK1 R761W, TSC2 M815I, ARID1A A1872T), which may be the putative mutations which result in
CC
osimertinib resistance and drive tumor progression. Patient 06 harbored EGFR L858R
A
and H870R, accompanied with MSH3 L297R mutation, which may also be a potential mechanism for resistance after osimertinib treatment. Further validation studies were needed to explore the potential underlying mechanism of these newly identified mutations.
Of the 6 patients monitored dynamically by NGS, EGFR T790M mutation,
which was targeted by osimertinib, diminished in 4 patients and decreased in AF from 6.27% to 0.22% in 1 patient after osimertinib therapy. The concomitancy of lost T790M and newly developed gene aberrance post-osimertinib, suggested that the resistance may arise from T790 wildtype subclones and reflected the clinical feature
SC R
Mutation spectrum of 13 patients with osimertinib resistance
IP T
of tumor heterogeneity.
Biopsies from 13 NSCLC patients, consisting of 4 paired tissue samples and 9
U
paired plasma samples, were subjected to capture-based targeted ultra-deep
N
sequencing. The mutation spectrum which demonstrated the somatic mutations was
A
presented in Figure 1B. Overall, 70 unique somatic aberrations, consisting of 26 genes,
M
were identified. 19 copy-number variants (CNVs) were detected: 4 MET
TE D
amplifications and 5 EGFR amplifications were observed; Four copy number deletion also existed in this cohort. In addition to CNVs, 33 missense variants were identified and 5 frame-shift variants was detected. In the cohort of 13 patients, MET
EP
amplification accounted for 30.8% (4/13) and the Integrative Genomics Viewer (IGV)
CC
of these amplifications from NGS results were presented in Figure 2.
A
Harboring of MET amplification affect the PFS after osimertinib treatment
To further investigate MET amplification as a mechanism of resistance to
osimertinib, we next sought to explore the prognosis difference between patients with MET copy number amplification and patients did not harboring MET amplification after osimertinib resistance development. 13 patients developed osimertinib resistance
were enrolled in this analysis. We observed progression-free survival (PFS) difference between patients with or without MET amplification. Patients with MET amplification (n=4, mPFS=3.5 months) after osimertinib resistance commonly had a shorter mPFS than patients without MET CNV appearance or increase (n=9, mPFS=9.9 months,
IP T
p=0.117, Figure 3A). The median overall survival (mOS) of MET amplification group was also shorter than non-MET amplification group (15.6 moths vs. 30.7 months,
SC R
p=0.885, Figure 3B). The high p value may be attributed to the limited patient number. This finding indicated that the presence of MET CNV as the resistance mechanisms
U
following third-generation TKI osimertinib was associated with an inferior survival.
A
N
Clinical evidence of combinatorial therapy of EGFR-TKI with crizotinib
M
Of the 13 patients developed osimertinib resistance, we further explored the
TE D
clinical efficacy of combinatorial therapy of first/third generation EGFR-TKI with MET inhibitor crizotinib on two patients.
A 44-year old female patient was diagnosed as stage IV lung adenocarcinoma.
EP
After commencement of osimertinib (80 mg once daily) due to the appearance of
CC
EGFR T790M, she obtained stable disease (SD) 42 days after the initiation of treatment and developed with progressive disease (PD) 5 months later. Plasma ctDNA
A
profiling revealed the harboring of EGFR exon 19 deletion and MET amplification with a copy number of 2.31 at the PD stage. Considering of the presence of MET amplification and absence of EGFR T790M, a combinatorial treatment, consisting of crizotinib (250mg twice daily) with first-generation EGFR-TKI icotinib (125mg
three times), was administrated. The patient obtained significant improvement of symptoms such as cough, asthma, dizziness and hemiplegia within 42 days. Partial response was confirmed 42 days after the initiation of the treatment by radiography. CT scan revealed the clinical response with shrinkage of 54.1% of the lung lesion and
IP T
68.2% of the head lesion (Figure 4A). Interestingly, re-biopsy of plasma ctDNA revealed the negative for all panel-included cancer-related mutations, which indicated
SC R
the efficacy of the combinatorial therapy of crizotinib and icotinib. The response
sustained for 3 months without progressive disease (PD) until the patient gave up
U
treatment due to her own reasons. And we were unable to obtain the follow-up record
A
N
afterwards.
M
A 51-year old male patient presented at our hospital with stage IV lung adenocarcinoma. After osimertinib administration and resistance development, plasma
TE D
ctDNA monitored by NGS revealed the decreased allelic fraction of EGFR exon 19 deletion (from 5.24% to 2.87%), T790M (from 6.27% to 0.22%), accompanied with
EP
MET copy number gain (CN=2.33). Taken into account of the co-existence of T790M and MET amplification, the combination of crizotinib (250mg twice daily) and
CC
osimertinib (80mg once daily) were administrated to the patient. He achieved partial
A
response 42 days after the initiation of treatment, with a shrinkage of 46.5% tumor size in lung lesion (Figure 4B). He also obtained significant improvement of symptoms in cough, expectoration. He was still under disease control and has been advised to attend regular follow-up visits.
Discussion
Osimertinib is a third-generation irreversible EGFR-TKI that overcomes EGFR T790M, the first generation of EGFR-TKI resistance mutation. However, gene alterations lead to resistance of osimertinib inevitable appeared after a period of treatment. EGFR C797, osimertinib binding site, has been confirmed to be a common
IP T
resistance mechanism when the cysteine mutated to serine.14 The C797S mutation can be targeted and effectively inhibited by a combination treatment of first- and
SC R
third-generation EGFR-TKIs if the C797S and T790M occurred in separate allele of the chromosome.15
U
Another mechanism of resistance to third generation EGFR-TKIs is the MET
A
N
amplification. Chabon et al. identified MET copy-number gain was the most frequent
M
mechanism to third-generation of EGFR-TKI rociletinib, occurring in 11 patients of a cohort of 43 NSCLC patient (26%).16 In a 2017 ASCO abstract, Piotrowska et al.
TE D
reported that 30% (7/23) patients were detected with MET amplification after osimertinib resistance. In our study, 4 patients in the cohort were detected with MET
EP
amplification after osimertinib treatment with an incidence of 30.7% (4/13), in accordance with incidence reported in previous publications. Moreover, several novel
CC
putative resistance mutations were also identified. Although further validations were
A
needed, this finding can provide valuable knowledge for the underlying resistance mechanism of osimertinib.
Of the sequenced 13 patients who treated with osimertinib due to T790M and further developed with osimertinib resistance, we analyzed the correlation of MET amplification and progression-free survival. We found that patients with MET CNV
commonly associated with shorter PFS and OS than patients without MET CNV after osimertinib treatment. Earlier studies reported that patients with MET amplification combined with T790M mutations appear to have an earlier resistance to third-generation of EGFR-TKI,16 suggesting that MET amplification induced
IP T
osimertinib resistance might occur in the early phase of drug administration.
SC R
Our study demonstrated that ctDNA profiling from noninvasive biopsy using NGS can dynamically monitor tumor clonal progress and disclose the underlying
resistance mechanism. This application of ctDNA has important implications for
U
therapeutic strategy design and treatment efficacy evaluation. In our study, one patient
A
N
developed with MET amplification and EGFR exon 19 deletion, without T790M,
M
displayed partial response to the combinatorial therapy of crizotinib and first-generation of EGFR-TKI icotinib, both clinically and radiographically. Another
TE D
patient developed with MET amplification after osimertinib therapy, accompanied with EGFR exon 19 deletion and T790M. He responded well to the combination of
EP
crizotinib and osimertinib. This indicated that personalized targeted therapy guided by ctDNA monitoring and analysis may be of significant clinical utility. Moreover, to the
CC
best of our knowledge, this was the first efficacious clinical evidence of a NSCLC
A
patient who achieved partial response to the combination of first-generation EGFR-TKI icotinib and crizotinib, after the previous treatment history of third-generation TKI osimertinib resistance.
There was also limitation in this study. Contrary to our findings in patients, preclinical studies have suggested that resistance to third-generation EGFR-TKIs
would primarily owing to the additional mutations in EGFR itself (for example, C797S).17,
18
However, we observed MET amplification to be the most highly
appearance resistance mechanism to osimertinib in our study. This observation indicated that larger cohort study and further investigation of comprehensive
SC R
IP T
resistance profiling were required to explore the underlying resistance mechanism.
Conflict of Interest:
A
N
U
Authors declare no conflict of interest.
M
Acknowledgement
TE D
We owe thanks to the patients and their family. We thank the staffs at Chongqing
A
CC
EP
Daping Hospital.
References: 1.
Remon J, Caramella C, Jovelet C, et al. Osimertinib benefit in EGFR-mutant NSCLC patients with
T790M-mutation detected by circulating tumour DNA. Ann Oncol 2017;28:784-790. 2.
Mok TS, Wu YL, Ahn MJ, et al. Osimertinib or Platinum-Pemetrexed in EGFR T790M-Positive Lung
Cancer. N Engl J Med 2017;376:629-640. Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutation mediates resistance to
IP T
3.
AZD9291 in non-small cell lung cancer harboring EGFR T790M. Nat Med 2015;21:560-562. 4.
Ortiz-Cuaran S, Scheffler M, Plenker D, et al. Heterogeneous Mechanisms of Primary and
5.
SC R
Acquired Resistance to Third-Generation EGFR Inhibitors. Clin Cancer Res 2016;22:4837-4847.
Ho CC, Liao WY, Lin CA, et al. Acquired BRAF V600E Mutation as Resistant Mechanism after
Treatment with Osimertinib. J Thorac Oncol 2017;12:567-572. 6.
Planchard D, Loriot Y, Andre F, et al. EGFR-independent mechanisms of acquired resistance to
AZD9291 in EGFR T790M-positive NSCLC patients. Ann Oncol 2015;26:2073-2078.
Knebel FH, Bettoni F, Shimada AK, et al. Sequential liquid biopsies reveal dynamic alterations of
U
7.
osimertinib in NSCLC. Lung Cancer 2017;108:238-241.
Ham JS, Kim S, Kim HK, et al. Two Cases of Small Cell Lung Cancer Transformation from EGFR
A
8.
N
EGFR driver mutations and indicate EGFR amplification as a new mechanism of resistance to
Mutant Adenocarcinoma During AZD9291 Treatment. J Thorac Oncol 2016;11:e1-4. Sattler M, Reddy MM, Hasina R, et al. The role of the c-Met pathway in lung cancer and the
M
9.
potential for targeted therapy. Ther Adv Med Oncol 2011;3:171-184. 10. Ou SH, Agarwal N, Ali SM. High MET amplification level as a resistance mechanism to osimertinib
TE D
(AZD9291) in a patient that symptomatically responded to crizotinib treatment post-osimertinib progression. Lung Cancer 2016;98:59-61.
11. Xi R, Hadjipanayis AG, Luquette LJ, et al. Copy number variation detection in whole-genome sequencing data using the Bayesian information criterion. Proc Natl Acad Sci U S A 2011;108:E1128-1136.
EP
12. Li J, Lupat R, Amarasinghe KC, et al. CONTRA: copy number analysis for targeted resequencing. Bioinformatics 2012;28:1307-1313. 13. Yang X, Chu Y, Zhang R, et al. Technical Validation of a Next-Generation Sequencing Assay for
CC
Detecting Clinically Relevant Levels of Breast Cancer-Related Single-Nucleotide Variants and Copy Number Variants Using Simulated Cell-Free DNA. J Mol Diagn 2017;19:525-536. 14. Yu HA, Tian SK, Drilon AE, et al. Acquired Resistance of EGFR-Mutant Lung Cancer to a
A
T790M-Specific EGFR Inhibitor: Emergence of a Third Mutation (C797S) in the EGFR Tyrosine Kinase Domain. JAMA Oncol 2015;1:982-984. 15. Wang Z, Yang JJ, Huang J, et al. Lung Adenocarcinoma Harboring EGFR T790M and In Trans C797S Responds to Combination Therapy of First- and Third-Generation EGFR TKIs and Shifts Allelic Configuration at Resistance. J Thorac Oncol 2017. 16. Chabon JJ, Simmons AD, Lovejoy AF, et al. Circulating tumour DNA profiling reveals heterogeneity of EGFR inhibitor resistance mechanisms in lung cancer patients. Nat Commun 2016;7:11815. 17. Ercan D, Choi HG, Yun CH, et al. EGFR Mutations and Resistance to Irreversible Pyrimidine-Based
EGFR Inhibitors. Clin Cancer Res 2015;21:3913-3923. 18. Niederst MJ, Hu H, Mulvey HE, et al. The Allelic Context of the C797S Mutation Acquired upon Treatment with Third-Generation EGFR Inhibitors Impacts Sensitivity to Subsequent Treatment
A
CC
EP
TE D
M
A
N
U
SC R
IP T
Strategies. Clin Cancer Res 2015;21:3924-3933.
Figure Legends
Figure 1: Overall mutation spectrum of the cohort. A. Mutation spectrum of 6 paired biopsies before and after osimertinib treatment. B. Mutation spectrum of the 13
IP T
patient cohort after osimertinib resistance. P1-1 means the mutation spectrum of patient 1 before osimertinib treatment while P1-2 stands for patient 1 after osimertinib
SC R
treatment. The top bar demonstrated the number of mutations detected in an individual patient. The side bar stands for the patient number harboring the
U
corresponding mutation. The bottom category indicates the sequencing panel used for
A
CC
EP
TE D
M
A
N
each tissue or plasma sample.
Figure 2: MET copy-number amplification after osimertinib resistance in 4 patients.
IP T SC R U N A
M
Figure 3: Kaplan-Meier analysis of patients with MET amplification and without MET amplification appearance or increase after osimertinib resistance patients. A.
A
CC
EP
TE D
PFS analysis. B. OS analysis.
IP T SC R U N A
M
Figure 4: Computed tomography scan images of lung and magnetic resonance scan images of head prior and post-crizotinib and EGFR-TKI treatment. A. After 42 days
TE D
treatment of crizotinib and icotinib, partial response was confirmed at lung lesion and head lesion. B. After 42 days treatment of crizotinib and osimertinib, partial response
A
CC
EP
was confirmed at lung lesion.
EP
CC
A TE D
IP T
SC R
U
N
A
M
Table 1: Summary of baseline patient characteristics.
Gender
Age (years)
Smoking history
Past chemotherapy
1
female
62
No
No
2
male
53
Yes
Yes
3
female
70
No
No
4
female
66
No
No
5
female
58
No
No
6
male
45
No
No
7
male
70
Yes
No
8
female
59
No
No
9
female
44
No
No
10
male
52
No
No
11
female
53
No
No
12
male
47
Yes
No
13
male
72
No
Yes
A
CC
EP
TE D
M
A
N
U
SC R
IP T
Patient
S0169-5002(18)30262-9 https://doi.org/10.1016/j.lungcan.2018.02.007 LUNG 5573
To appear in:
Lung Cancer
Received date: Revised date: Accepted date:
17-11-2017 8-2-2018 9-2-2018
Please cite this article as: Wang Yubo, Li Li, Han Rui, Jiao Lin, Zheng Jie, He Yong.Clinical analysis by next-generation sequencing for NSCLC patients with MET amplification resistant to osimertinib.Lung Cancer https://doi.org/10.1016/j.lungcan.2018.02.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Clinical analysis by next-generation sequencing for NSCLC patients with MET amplification resistant to osimertinib
SC R
IP T
Running title: Cohort study of MET amplification for osimertinib resistance
Authors and affiliations
A
Department of Respiratory Medicine, Daping Hospital, Third Military Medical
M
1
N
U
Yubo Wang1, Li Li1, Rui Han1, Lin Jiao1, Jie Zheng1, Yong He1
TE D
University, Chongqing, P.R. China
EP
Corresponding Author:
CC
Yong He
Department of Respiratory Medicine, Daping Hospital, Third Military Medical
A
University, 10# Changjiangzhilu Daping, Yuzhong District, 40042, Chongqing, P.R. China. Phone:86-02368757791; Fax:86-02368700970; E-mail: [email protected]
Highlights
MET amplification is a common mechanism of osimertinib resistance. MET amplification-induced osimertinib resistance might occur in the early phase. Profiling using NGS can disclose the underlying resistance mechanism. EGFR-TKI plus crizotinib are effective in patients with MET amp-induced resistance
IP T
SC R
Abstract
Introduction: The efficacy of osimertinib was compromised by the development of
U
resistance mechanisms, such as MET amplification. However, cohort studies of
N
osimertinib resistance mechanism, and the correlation of MET and progression-free
M
A
survival (PFS) after osimertinib resistance have been poorly investigated.
Objectives: This study was carried out to study the acquired MET amplification after
TE D
osimertinib resistance in advanced lung adenocarcinoma patients, and interrogate the correlation of clinical prognosis and MET amplification.
EP
Methods: We performed capture-based sequencing on longitudinal plasma and tissue
CC
samples obtained before osimertinib treatment and after resistance development from lung adenocarcinoma patients to investigate the underlying resistance mechanism. We
A
also investigated the correlation of MET amplification and patient prognosis after osimertinib resistance using Kaplan-Meier analysis.
Results: Paired biopsies before osimertinib treatment and after the resistance development revealed underlying resistance mechanisms. In addition, a cohort of 13
patients who developed disease progression after osimertinib resistance was investigated. Patients with MET amplification after osimertinib resistance commonly had inferior median progression-free survival (mPFS) than patients without MET amplification appearance or increase (3.5 months vs. 9.9 months, p=0.117). Patients in
IP T
MET amplification group also displayed poor median overall survival (mOS) compared to MET amplification negative group (15.6 months vs. 30.7 months,
SC R
p=0.885). Furthermore, combinatorial treatment of first/third-generation EGFR-TKI
and crizotinib was efficaciously administrated into two patients with newly acquired
U
MET amplification after osimertinib resistance. Partial responses were achieved by
A
N
them, both clinically and radiographically.
M
Conclusions: We investigated the osimertinib resistance mechanism in a small cohort of lung adenocarcinoma patients, and demonstrated MET amplification was correlated
TE D
with inferior PFS/OS after osimertinib treatment. Moreover, we reported the first clinical evidence of efficacy generated by combination of first-generation EGFR-TKI
CC
EP
icotinib and crizotinib after the resistance to osimertinib.
A
Key words:MET amplification, resistance, osimertinib, crizotinib
Introduction
Osimertinib, a third-generation epidermal growth factor receptor (EGFR)
tyrosine kinase inhibitor (TKI), benefits patients with T790M mutant non-small-cell lung cancer (NSCLC) who fail in treatment with first-generation EGFR-TKIs.1,
2
However, drug resistance limits the clinical application of osimertinib. At present, mechanisms of osimertinib resistance have been investigated in several studies. The
IP T
most common mechanism of osimertinib resistance was EGFR C797S mutation.3 MET amplification is another mechanism for acquired resistance to third generation
SC R
EGFR-TKIs.4 In addition, BRAF V600E mutation,5 HER2 amplification,6 EGFR amplification,7 and small-cell histological transformation8 were also been reported to
U
result in third-generation EGFR-TKI resistance.
A
N
MET is a tyrosine kinase receptor located at 7q21-q31. MET amplification is a
M
common gene alteration which occurs in about 1%~20% of NSCLC and associates with poor prognosis.9 MET amplification has been reported in approximately 5% of
TE D
resistant patients to first generation EGFR-TKI, such as erlotinib and gefitinib.10 In recent decades, a small number of cases have reported that MET amplification may
EP
also be induced after osimertinib treatment, suggesting that MET amplification can also be responsible for osimertinib resistance.6 However, cohort studies of osimertinib
CC
resistance mechanism, especially MET-associated, have been poorly investigated.
A
In this study, a retrospective investigation was performed by profiling plasma
and tissue biopsies from 13 patients with acquired resistance to osimertinib using NGS. Paired biopsies obtained before osimertinib treatment and after resistance development in 6 patients were analyzed for mutation spectrum and novel putative resistance mutations were identified. Furthermore, we provided two efficacious
clinical evidences of combinatorial therapy of EGFR-TKI and crizotinib to illustrate the personalized therapy design guided by dynamic ctDNA monitoring.
Patients and methods
IP T
Patient Information
Thirteen patients were collected and evaluated at our Hospital. This study was
SC R
approved by the Institutional Review Board at the Chongqing Daping Hospital. All patients were provided informed consent to this study and gave permission to the
U
entire study. Three sequencing panels, consisting of 168, 180 or 365 cancer-related
N
genes, were developed and validated for somatic variations identification.
M
A
Preparation of DNA of liquid and tissue samples
TE D
For cfDNA preparation of 168 cancer-related gene panel, 10 mL of peripheral blood was obtained and stored in tubes containing ethylenediaminetetraacetic acid, followed by incubating at room temperature for 2 hours. Circulating cfDNA was
EP
recovered from 4 to 5mL of plasma by using the QIAamp Circulating Nucleic Acid
CC
kit (Qiagen). Quantification of cfDNA was assessed with the Qubit 2.0 fluorimeter (Thermo Fisher Scientific). A minimum of 50 ng of cfDNA is required for
A
construction of an NGS library. For 365 cancer-related genes panel sequencing, genomic DNA in Formalin Fixed Paraffin Embedded (FFPE) sections was extracted with QIAamp DNA FFPE Tissue Kit (Qiagen), while cell-free circulating DNAs in plasma were extracted by QIAamp Circulating Nucleic Acid Kit (Qiagen) following the standard protocols. DNA was quantified by PicoGreen fluorescence assay
(Invitrogen). For 180 cancer-related genes panel sequencing, FFPE samples were extracted using QIAamp DNA FFPE Tissue Kit (Qiagen, Hilden, Germany). DNA concentration was measured using a Qubit fluorometer (Invitrogen, Carlsbad, CA USA) and the Qubit dsDNA HS (High Sensitivity) Assay Kit.
IP T
NGS sequencing of liquid biopsies
SC R
For liquid biopsies sequencing of 168 cancer-related gene panel, fragments with a size of 200-400 base pairs (bp) were selected by using Agencourt AMPure beads
U
(Beckman Coulter, Fullerton, CA). Hybridization was performed with capture probes
N
baits, followed by hybrid selection with magnetic beads, and PCR amplification. A
A
high-sensitivity DNA assay was performed to assess the quality and size of the
M
fragments, and indexed samples were sequenced on a Nextseq500 sequencer
TE D
(Illumina, Inc., San Diego, CA) with paired-end reads. Sequence data were mapped to the human genome (hg19) using BWA aligner.
For 365-cancer related gene panel sequencing, construction of sequencing
EP
libraries 50-200 ng of DNAs was fragmented to around ~200 bp by sonication
CC
(Covaris), and constructed into the libraries with KAPA Hyper Prep Kit (Kapa
A
Biosystems) for sequencing.
For 180-cancer related gene panel sequencing, library construction was prepared
according to IlluminaTruSeq DNA Library Preparation Kit (Illumina, San Diego, CA) using 1g of DNA sheared by an ultrasonoscope to generate fragments with a peak of 250 bps, followed by end repair, A tailing and ligation. Libraries were hybridized to
custom-designed biotinylated oligonucleotide probes (Roche NimbleGen, Madison, WI, USA) and DNA sequencing was carried out with the HiSeq 3000 Sequencing System (Illumina, San Diego, CA) with 2×100-bp paired-end reads.
IP T
Copy number variation algorithms In the panel of 168 cancer-related genes for plasma ctDNA sequencing, CNV
SC R
were identified based on depth of coverage data of capture intervals. Coverage depth data were firstly corrected for the sequencing bias due to GC content and target probe
U
density. The average coverage of all capture regions was calculated as internal control,
N
which was utilized to normalize the coverage of different samples to comparable
A
scales. The coverage of MET with copy number gain would be significantly larger
M
than the internal control. The difference of adjusted coverage depth for each of genes
TE D
from plasma samples and reference was evaluated by t-test. Genes with p-value<5e-3 in the t-test and CN>2.25 were considered as amplification.
For the 365-cancer related genes panel, a statistically rigorous and
EP
computationally efficient algorithm called BIC-seq was used for detecting CNVs. We
CC
obtained a log-ratio profile of the sample by normalizing the sequence coverage obtained at all exons against a process-matched normal control. This profile was
A
corrected for GC-bias and log2ratio>0.7 was considered to be copy number gain and abstract copy number was estimated by log2ratio normalized by tumor purity.11 cfDNA fraction of 5% is required for MET amplification detection of minimum 2.5 copy. In total, for gene copy gain events whose ratios were larger than 3.5 copy, the
sensitivity of CNV detection in ctDNA achieved 83.3%.
For the 180-cancer related genes panel for tissue sequencing, somatic copy-number alterations were identified with CONTRA (v2.0.8).12 Base-level log-ratios between the case and control were taken into consideration to eliminated
IP T
GC bias and imbalanced library size effect. Target regions of <10bp with depth of
SC R
coverage <10 were discarded. For the remaining regions, the mean log-ratio was
calculated and the significant P value was then assigned. Then the circular binary segmentation was applied to the log-ratios of regions with default settings.
U
Segmentation results were then used for the subsequent analysis. GISTIC37 algorithm
A
N
was used to infer recurrently amplified or deleted genomic regions, using copy
M
numbers in 100-kb windows. G scores represented the frequency and amplitude of amplifications or deletions of each genomic region. MET copy number gain was
TE D
identified using thresholds of 2.3 copies normalized to control.13
Statistical Analysis
EP
All these survival analyses were conducted in software R with survival package.
CC
We performed Kaplan-Meier analysis to investigate the survival functions. Log-rank test was used to determine the difference of survival curve between groups. P-value of
A
< 0.05 was considered statistically significant.
Results
Patient cohort and NGS overview
To characterize potential mechanisms of resistance to third-generation EGFR-TKI osimertinib, a retrospective analysis was performed by enrolling 37 NSCLC patients from August 2015 to May 2017 in our hospital. Prior to treatment of osimertinib, all these patients displayed resistance to the first-generation of
IP T
EGFR-TKI and the classic resistant mutation EGFR T790M mutation were detected. Of these, 16 patients (43.2%) developed disease progression after osimertinib
SC R
resistance, and either plasma or tissue biopsies from 13 patients were profiled by capture-based targeted ultra-deep sequencing after osimertinib resistance. Of the
U
cohort of 13 patients, the average age was 57.7 years, ranging from 44 to 72 years. 6
N
were males and 7 were females. 3 patients had a history of smoking and 10 patients
A
were nonsmokers. All the 13 patients were classified as stage IV adenocarcinoma with
M
distant metastasis. The clinical characteristics of the cohort was summarized in Table
TE D
1.
Moreover, paired specimens (before and after osimertinib treatment) from 6 of
EP
the 13 patients were sequenced to investigate the underlying novel potential resistant mechanisms and monitor the dynamic genomic alteration status. The sequencing
CC
panels, consisting of 168, 180 and 365 critical exons and introns for genomic variants,
A
covers multiple classes of somatic mutations for detection and quantification of genetic alterations.
Potential newly acquired resistance mechanisms to osimertinib
To identify the underlying mechanisms of resistance to osimertinib, we deemed
that gene aberrations which drive resistance should generate in the process of osimertinib therapy. Therefore, we identified gene alterations that were absent before osimertinib treatment but newly emerged or increased in relative abundance after the resistance development. We performed capture-based ultradeep targeted sequencing
IP T
on paired biopsies from 6 patients before osimertinib treatment and after resistance development, consisting of 1 paired tissue sample and 5 paired plasma samples. We
SC R
identified several putative mechanisms through dynamically monitoring (Figure 1A,
1B). MET amplification was the most frequent mechanism and was observed in 4
U
patients, one with increased MET copy-number abundance and three with newly
N
emerged MET amplification. BRAF single nucleotide variant V600E, which had been
M
of the four patients.
A
reported to induce osimertinib resistance in previous studies,5 was also detected in one
TE D
Of the other two patients without MET amplification, patients 05 who originally harbored EGFR sensitizing mutation L747_P753delinsS and resistant mutation
EP
T790M newly obtained four missense variants (PTEN R346C, NTRK1 R761W, TSC2 M815I, ARID1A A1872T), which may be the putative mutations which result in
CC
osimertinib resistance and drive tumor progression. Patient 06 harbored EGFR L858R
A
and H870R, accompanied with MSH3 L297R mutation, which may also be a potential mechanism for resistance after osimertinib treatment. Further validation studies were needed to explore the potential underlying mechanism of these newly identified mutations.
Of the 6 patients monitored dynamically by NGS, EGFR T790M mutation,
which was targeted by osimertinib, diminished in 4 patients and decreased in AF from 6.27% to 0.22% in 1 patient after osimertinib therapy. The concomitancy of lost T790M and newly developed gene aberrance post-osimertinib, suggested that the resistance may arise from T790 wildtype subclones and reflected the clinical feature
SC R
Mutation spectrum of 13 patients with osimertinib resistance
IP T
of tumor heterogeneity.
Biopsies from 13 NSCLC patients, consisting of 4 paired tissue samples and 9
U
paired plasma samples, were subjected to capture-based targeted ultra-deep
N
sequencing. The mutation spectrum which demonstrated the somatic mutations was
A
presented in Figure 1B. Overall, 70 unique somatic aberrations, consisting of 26 genes,
M
were identified. 19 copy-number variants (CNVs) were detected: 4 MET
TE D
amplifications and 5 EGFR amplifications were observed; Four copy number deletion also existed in this cohort. In addition to CNVs, 33 missense variants were identified and 5 frame-shift variants was detected. In the cohort of 13 patients, MET
EP
amplification accounted for 30.8% (4/13) and the Integrative Genomics Viewer (IGV)
CC
of these amplifications from NGS results were presented in Figure 2.
A
Harboring of MET amplification affect the PFS after osimertinib treatment
To further investigate MET amplification as a mechanism of resistance to
osimertinib, we next sought to explore the prognosis difference between patients with MET copy number amplification and patients did not harboring MET amplification after osimertinib resistance development. 13 patients developed osimertinib resistance
were enrolled in this analysis. We observed progression-free survival (PFS) difference between patients with or without MET amplification. Patients with MET amplification (n=4, mPFS=3.5 months) after osimertinib resistance commonly had a shorter mPFS than patients without MET CNV appearance or increase (n=9, mPFS=9.9 months,
IP T
p=0.117, Figure 3A). The median overall survival (mOS) of MET amplification group was also shorter than non-MET amplification group (15.6 moths vs. 30.7 months,
SC R
p=0.885, Figure 3B). The high p value may be attributed to the limited patient number. This finding indicated that the presence of MET CNV as the resistance mechanisms
U
following third-generation TKI osimertinib was associated with an inferior survival.
A
N
Clinical evidence of combinatorial therapy of EGFR-TKI with crizotinib
M
Of the 13 patients developed osimertinib resistance, we further explored the
TE D
clinical efficacy of combinatorial therapy of first/third generation EGFR-TKI with MET inhibitor crizotinib on two patients.
A 44-year old female patient was diagnosed as stage IV lung adenocarcinoma.
EP
After commencement of osimertinib (80 mg once daily) due to the appearance of
CC
EGFR T790M, she obtained stable disease (SD) 42 days after the initiation of treatment and developed with progressive disease (PD) 5 months later. Plasma ctDNA
A
profiling revealed the harboring of EGFR exon 19 deletion and MET amplification with a copy number of 2.31 at the PD stage. Considering of the presence of MET amplification and absence of EGFR T790M, a combinatorial treatment, consisting of crizotinib (250mg twice daily) with first-generation EGFR-TKI icotinib (125mg
three times), was administrated. The patient obtained significant improvement of symptoms such as cough, asthma, dizziness and hemiplegia within 42 days. Partial response was confirmed 42 days after the initiation of the treatment by radiography. CT scan revealed the clinical response with shrinkage of 54.1% of the lung lesion and
IP T
68.2% of the head lesion (Figure 4A). Interestingly, re-biopsy of plasma ctDNA revealed the negative for all panel-included cancer-related mutations, which indicated
SC R
the efficacy of the combinatorial therapy of crizotinib and icotinib. The response
sustained for 3 months without progressive disease (PD) until the patient gave up
U
treatment due to her own reasons. And we were unable to obtain the follow-up record
A
N
afterwards.
M
A 51-year old male patient presented at our hospital with stage IV lung adenocarcinoma. After osimertinib administration and resistance development, plasma
TE D
ctDNA monitored by NGS revealed the decreased allelic fraction of EGFR exon 19 deletion (from 5.24% to 2.87%), T790M (from 6.27% to 0.22%), accompanied with
EP
MET copy number gain (CN=2.33). Taken into account of the co-existence of T790M and MET amplification, the combination of crizotinib (250mg twice daily) and
CC
osimertinib (80mg once daily) were administrated to the patient. He achieved partial
A
response 42 days after the initiation of treatment, with a shrinkage of 46.5% tumor size in lung lesion (Figure 4B). He also obtained significant improvement of symptoms in cough, expectoration. He was still under disease control and has been advised to attend regular follow-up visits.
Discussion
Osimertinib is a third-generation irreversible EGFR-TKI that overcomes EGFR T790M, the first generation of EGFR-TKI resistance mutation. However, gene alterations lead to resistance of osimertinib inevitable appeared after a period of treatment. EGFR C797, osimertinib binding site, has been confirmed to be a common
IP T
resistance mechanism when the cysteine mutated to serine.14 The C797S mutation can be targeted and effectively inhibited by a combination treatment of first- and
SC R
third-generation EGFR-TKIs if the C797S and T790M occurred in separate allele of the chromosome.15
U
Another mechanism of resistance to third generation EGFR-TKIs is the MET
A
N
amplification. Chabon et al. identified MET copy-number gain was the most frequent
M
mechanism to third-generation of EGFR-TKI rociletinib, occurring in 11 patients of a cohort of 43 NSCLC patient (26%).16 In a 2017 ASCO abstract, Piotrowska et al.
TE D
reported that 30% (7/23) patients were detected with MET amplification after osimertinib resistance. In our study, 4 patients in the cohort were detected with MET
EP
amplification after osimertinib treatment with an incidence of 30.7% (4/13), in accordance with incidence reported in previous publications. Moreover, several novel
CC
putative resistance mutations were also identified. Although further validations were
A
needed, this finding can provide valuable knowledge for the underlying resistance mechanism of osimertinib.
Of the sequenced 13 patients who treated with osimertinib due to T790M and further developed with osimertinib resistance, we analyzed the correlation of MET amplification and progression-free survival. We found that patients with MET CNV
commonly associated with shorter PFS and OS than patients without MET CNV after osimertinib treatment. Earlier studies reported that patients with MET amplification combined with T790M mutations appear to have an earlier resistance to third-generation of EGFR-TKI,16 suggesting that MET amplification induced
IP T
osimertinib resistance might occur in the early phase of drug administration.
SC R
Our study demonstrated that ctDNA profiling from noninvasive biopsy using NGS can dynamically monitor tumor clonal progress and disclose the underlying
resistance mechanism. This application of ctDNA has important implications for
U
therapeutic strategy design and treatment efficacy evaluation. In our study, one patient
A
N
developed with MET amplification and EGFR exon 19 deletion, without T790M,
M
displayed partial response to the combinatorial therapy of crizotinib and first-generation of EGFR-TKI icotinib, both clinically and radiographically. Another
TE D
patient developed with MET amplification after osimertinib therapy, accompanied with EGFR exon 19 deletion and T790M. He responded well to the combination of
EP
crizotinib and osimertinib. This indicated that personalized targeted therapy guided by ctDNA monitoring and analysis may be of significant clinical utility. Moreover, to the
CC
best of our knowledge, this was the first efficacious clinical evidence of a NSCLC
A
patient who achieved partial response to the combination of first-generation EGFR-TKI icotinib and crizotinib, after the previous treatment history of third-generation TKI osimertinib resistance.
There was also limitation in this study. Contrary to our findings in patients, preclinical studies have suggested that resistance to third-generation EGFR-TKIs
would primarily owing to the additional mutations in EGFR itself (for example, C797S).17,
18
However, we observed MET amplification to be the most highly
appearance resistance mechanism to osimertinib in our study. This observation indicated that larger cohort study and further investigation of comprehensive
SC R
IP T
resistance profiling were required to explore the underlying resistance mechanism.
Conflict of Interest:
A
N
U
Authors declare no conflict of interest.
M
Acknowledgement
TE D
We owe thanks to the patients and their family. We thank the staffs at Chongqing
A
CC
EP
Daping Hospital.
References: 1.
Remon J, Caramella C, Jovelet C, et al. Osimertinib benefit in EGFR-mutant NSCLC patients with
T790M-mutation detected by circulating tumour DNA. Ann Oncol 2017;28:784-790. 2.
Mok TS, Wu YL, Ahn MJ, et al. Osimertinib or Platinum-Pemetrexed in EGFR T790M-Positive Lung
Cancer. N Engl J Med 2017;376:629-640. Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutation mediates resistance to
IP T
3.
AZD9291 in non-small cell lung cancer harboring EGFR T790M. Nat Med 2015;21:560-562. 4.
Ortiz-Cuaran S, Scheffler M, Plenker D, et al. Heterogeneous Mechanisms of Primary and
5.
SC R
Acquired Resistance to Third-Generation EGFR Inhibitors. Clin Cancer Res 2016;22:4837-4847.
Ho CC, Liao WY, Lin CA, et al. Acquired BRAF V600E Mutation as Resistant Mechanism after
Treatment with Osimertinib. J Thorac Oncol 2017;12:567-572. 6.
Planchard D, Loriot Y, Andre F, et al. EGFR-independent mechanisms of acquired resistance to
AZD9291 in EGFR T790M-positive NSCLC patients. Ann Oncol 2015;26:2073-2078.
Knebel FH, Bettoni F, Shimada AK, et al. Sequential liquid biopsies reveal dynamic alterations of
U
7.
osimertinib in NSCLC. Lung Cancer 2017;108:238-241.
Ham JS, Kim S, Kim HK, et al. Two Cases of Small Cell Lung Cancer Transformation from EGFR
A
8.
N
EGFR driver mutations and indicate EGFR amplification as a new mechanism of resistance to
Mutant Adenocarcinoma During AZD9291 Treatment. J Thorac Oncol 2016;11:e1-4. Sattler M, Reddy MM, Hasina R, et al. The role of the c-Met pathway in lung cancer and the
M
9.
potential for targeted therapy. Ther Adv Med Oncol 2011;3:171-184. 10. Ou SH, Agarwal N, Ali SM. High MET amplification level as a resistance mechanism to osimertinib
TE D
(AZD9291) in a patient that symptomatically responded to crizotinib treatment post-osimertinib progression. Lung Cancer 2016;98:59-61.
11. Xi R, Hadjipanayis AG, Luquette LJ, et al. Copy number variation detection in whole-genome sequencing data using the Bayesian information criterion. Proc Natl Acad Sci U S A 2011;108:E1128-1136.
EP
12. Li J, Lupat R, Amarasinghe KC, et al. CONTRA: copy number analysis for targeted resequencing. Bioinformatics 2012;28:1307-1313. 13. Yang X, Chu Y, Zhang R, et al. Technical Validation of a Next-Generation Sequencing Assay for
CC
Detecting Clinically Relevant Levels of Breast Cancer-Related Single-Nucleotide Variants and Copy Number Variants Using Simulated Cell-Free DNA. J Mol Diagn 2017;19:525-536. 14. Yu HA, Tian SK, Drilon AE, et al. Acquired Resistance of EGFR-Mutant Lung Cancer to a
A
T790M-Specific EGFR Inhibitor: Emergence of a Third Mutation (C797S) in the EGFR Tyrosine Kinase Domain. JAMA Oncol 2015;1:982-984. 15. Wang Z, Yang JJ, Huang J, et al. Lung Adenocarcinoma Harboring EGFR T790M and In Trans C797S Responds to Combination Therapy of First- and Third-Generation EGFR TKIs and Shifts Allelic Configuration at Resistance. J Thorac Oncol 2017. 16. Chabon JJ, Simmons AD, Lovejoy AF, et al. Circulating tumour DNA profiling reveals heterogeneity of EGFR inhibitor resistance mechanisms in lung cancer patients. Nat Commun 2016;7:11815. 17. Ercan D, Choi HG, Yun CH, et al. EGFR Mutations and Resistance to Irreversible Pyrimidine-Based
EGFR Inhibitors. Clin Cancer Res 2015;21:3913-3923. 18. Niederst MJ, Hu H, Mulvey HE, et al. The Allelic Context of the C797S Mutation Acquired upon Treatment with Third-Generation EGFR Inhibitors Impacts Sensitivity to Subsequent Treatment
A
CC
EP
TE D
M
A
N
U
SC R
IP T
Strategies. Clin Cancer Res 2015;21:3924-3933.
Figure Legends
Figure 1: Overall mutation spectrum of the cohort. A. Mutation spectrum of 6 paired biopsies before and after osimertinib treatment. B. Mutation spectrum of the 13
IP T
patient cohort after osimertinib resistance. P1-1 means the mutation spectrum of patient 1 before osimertinib treatment while P1-2 stands for patient 1 after osimertinib
SC R
treatment. The top bar demonstrated the number of mutations detected in an individual patient. The side bar stands for the patient number harboring the
U
corresponding mutation. The bottom category indicates the sequencing panel used for
A
CC
EP
TE D
M
A
N
each tissue or plasma sample.
Figure 2: MET copy-number amplification after osimertinib resistance in 4 patients.
IP T SC R U N A
M
Figure 3: Kaplan-Meier analysis of patients with MET amplification and without MET amplification appearance or increase after osimertinib resistance patients. A.
A
CC
EP
TE D
PFS analysis. B. OS analysis.
IP T SC R U N A
M
Figure 4: Computed tomography scan images of lung and magnetic resonance scan images of head prior and post-crizotinib and EGFR-TKI treatment. A. After 42 days
TE D
treatment of crizotinib and icotinib, partial response was confirmed at lung lesion and head lesion. B. After 42 days treatment of crizotinib and osimertinib, partial response
A
CC
EP
was confirmed at lung lesion.
EP
CC
A TE D
IP T
SC R
U
N
A
M
Table 1: Summary of baseline patient characteristics.
Gender
Age (years)
Smoking history
Past chemotherapy
1
female
62
No
No
2
male
53
Yes
Yes
3
female
70
No
No
4
female
66
No
No
5
female
58
No
No
6
male
45
No
No
7
male
70
Yes
No
8
female
59
No
No
9
female
44
No
No
10
male
52
No
No
11
female
53
No
No
12
male
47
Yes
No
13
male
72
No
Yes
A
CC
EP
TE D
M
A
N
U
SC R
IP T
Patient