[Skip to Content]
[Skip to Content Landing]
Figure 1.
Assessment of Frequency of Tumor Protein p53 and Ataxia-Telangiectasia Mutated Comutation and Mutation Pattern of Driver Genes in Patients With Non–Small Cell Lung Cancer (NSCLC) From 3 Cohorts
Assessment of Frequency of Tumor Protein p53 and Ataxia-Telangiectasia Mutated Comutation and Mutation Pattern of Driver Genes in Patients With Non–Small Cell Lung Cancer (NSCLC) From 3 Cohorts

LUAD indicates lung adenocarcinoma; LUSC, lung squamous cell carcinoma; MSKCC, Memorial Sloan Kettering Cancer Center; and TCGA, The Cancer Genome Atlas.

Figure 2.
Tumor Mutation Burden of Samples From Patients With Non–Small Cell Lung Cancer From The Cancer Gene Atlas, Memorial Sloan Kettering Cancer Center (MSKCC), Geneplus, and POPLAR and OAK Cohorts.
Tumor Mutation Burden of Samples From Patients With Non–Small Cell Lung Cancer From The Cancer Gene Atlas,15 Memorial Sloan Kettering Cancer Center (MSKCC),19 Geneplus, and POPLAR4 and OAK20 Cohorts.

The height of the bars indicates median value; error bar, 95% CI. ATM indicates ataxia-telangiectasia mutated gene; NGS, next-generation sequencing; and TP53, tumor protein p53 gene.

Figure 3.
Association of TP53 and ATM Mutation Type With Prognosis in Patients Treated With Immune Checkpoint Inhibitors in the Memorial Sloan Kettering Cancer Center Cohort
Association of TP53 and ATM Mutation Type With Prognosis in Patients Treated With Immune Checkpoint Inhibitors in the Memorial Sloan Kettering Cancer Center Cohort23

ATM indicates ataxia-telangiectasia mutated gene; NSCLC, non–small cell lung cancer; TP53, tumor protein p53 gene; and crosses, patients who were censored.

Figure 4.
Association of TP53 and ATM Mutation Type With Prognosis and Response in the POPLAR and OAK Cohort
Association of TP53 and ATM Mutation Type With Prognosis and Response in the POPLAR4 and OAK20 Cohort

ATM indicates ataxia-telangiectasia mutated gene; NSCLC, non–small cell lung cancer; TP53, tumor protein p53 gene; and crosses, patients who were censored.

Supplement.

eMethods 1. Sample Processing and DNA Extraction

eMethods 2. Library Preparation, Target Capture and Next-Generation Sequencing

eMethods 3. Next-Generation Sequencing Analysis

eFigure 1. Data Sources

eFigure 2. Incidence of TP53 and ATM Comutation in the Cancer Genome Atlas and Geneplus Cohorts

eFigure 3. Assessment of Mutation Sites in TP53 and ATM Between Comutated and Solely Mutated Samples in Patients With Non–Small Cell Lung Cancer From the Cancer Genome Atlas and Geneplus Cohorts

eFigure 4. Tumor Mutation Burden of Samples Among Patients With Non–Small Cell Lung Cancer in the Cancer Genome Atlas, Memorial Sloan Kettering Cancer Center, Geneplus, and POPLAR and OAK Cohorts

eFigure 5. Association of TP53 and ATM Mutation Type With Overall Survival in Patients Treated With Immune Checkpoint Inhibitors in the Memorial Sloan Kettering Cancer Center Cohort

eFigure 6. Association of TP53 and ATM Mutation Type With Survival Among Patients in the POPLAR and OAK Cohort

eFigure 7. Gene Signatures Associated With TP53 and ATM Comutation in Patients With Non–Small Cell Lung Cancer in the Cancer Genome Atlas Cohort

eFigure 8. Gene Set Enrichment Analysis Identified Signaling Pathways Associated With TP53 and ATM Comutation in Patients With Non–Small Cell Lung Cancer in the Cancer Genome Atlas Cohort

eTable 1. Detailed List of Genes in the Geneplus 59 Panel and 1021 Panel

eTable 2. Characteristics of Patients Treated With Immune Checkpoint Inhibitors in the Memorial Sloan Kettering Cancer Center Cohort

eTable 3. Characteristics of Patients Treated With Immune Checkpoint Inhibitors in the POPLAR and OAK Cohort

eTable 4. Univariate Analysis of Factors Associated With Survival Among Patients With Non–Small Cell Lung Cancer Treated With Immune Checkpoint Inhibitors in the Memorial Sloan Kettering Cancer Center Cohort

eTable 5. Univariate Analysis of Factors Associated With Survival Among Patients With Any Cancer Treated With Immune Checkpoint Inhibitors in the Memorial Sloan Kettering Cancer Center Cohort

eTable 6. Univariate Analysis of Factors Associated With Survival Among Patients Treated With Immune Checkpoint Inhibitors in the POPLAR and OAK Cohort

eTable 7. Multivariable Analysis of Factors Associated With Survival of Patients Treated With Immune Checkpoint Inhibitors in the POPLAR and OAK Cohort

eReferences

1.
Reck  M, Rodríguez-Abreu  D, Robinson  AG,  et al; KEYNOTE-024 Investigators.  Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.  N Engl J Med. 2016;375(19):1823-1833. doi:10.1056/NEJMoa1606774PubMedGoogle ScholarCrossref
2.
Carbone  DP, Reck  M, Paz-Ares  L,  et al; CheckMate 026 Investigators.  First-line nivolumab in stage IV or recurrent non-small-cell lung cancer.  N Engl J Med. 2017;376(25):2415-2426. doi:10.1056/NEJMoa1613493PubMedGoogle ScholarCrossref
3.
Herbst  RS, Baas  P, Kim  DW,  et al.  Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial.  Lancet. 2016;387(10027):1540-1550. doi:10.1016/S0140-6736(15)01281-7PubMedGoogle ScholarCrossref
4.
Fehrenbacher  L, Spira  A, Ballinger  M,  et al; POPLAR Study Group.  Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial.  Lancet. 2016;387(10030):1837-1846. doi:10.1016/S0140-6736(16)00587-0PubMedGoogle ScholarCrossref
5.
Borghaei  H, Paz-Ares  L, Horn  L,  et al.  Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer.  N Engl J Med. 2015;373(17):1627-1639. doi:10.1056/NEJMoa1507643PubMedGoogle ScholarCrossref
6.
 Nivolumab approved for lung cancer.  Cancer Discov. 2015;5(5):OF1. doi:10.1158/2159-8290.CD-NB2015-042PubMedGoogle ScholarCrossref
7.
Langer  CJ, Gadgeel  SM, Borghaei  H,  et al; KEYNOTE-021 Investigators.  Carboplatin and pemetrexed with or without pembrolizumab for advanced, non-squamous non-small-cell lung cancer: a randomised, phase 2 cohort of the open-label KEYNOTE-021 study.  Lancet Oncol. 2016;17(11):1497-1508. doi:10.1016/S1470-2045(16)30498-3PubMedGoogle ScholarCrossref
8.
Hellmann  MD, Rizvi  NA, Goldman  JW,  et al.  Nivolumab plus ipilimumab as first-line treatment for advanced non-small-cell lung cancer (CheckMate 012): results of an open-label, phase 1, multicohort study.  Lancet Oncol. 2017;18(1):31-41. doi:10.1016/S1470-2045(16)30624-6PubMedGoogle ScholarCrossref
9.
Sacher  AG, Gandhi  L.  Biomarkers for the clinical use of PD-1/PD-L1 inhibitors in non-small-cell lung cancer: a review.  JAMA Oncol. 2016;2(9):1217-1222. doi:10.1001/jamaoncol.2016.0639PubMedGoogle ScholarCrossref
10.
Ready  N, Hellmann  MD, Awad  MM,  et al.  First-line nivolumab plus ipilimumab in advanced non-small-cell lung cancer (CheckMate 568): outcomes by programmed death ligand 1 and tumor mutational burden as biomarkers.  J Clin Oncol. 2019;37(12):992-1000. doi:10.1200/JCO.18.01042PubMedGoogle ScholarCrossref
11.
Hellmann  MD, Ciuleanu  TE, Pluzanski  A,  et al.  Nivolumab plus ipilimumab in lung cancer with a high tumor mutational burden.  N Engl J Med. 2018;378(22):2093-2104. doi:10.1056/NEJMoa1801946PubMedGoogle ScholarCrossref
12.
Mouw  KW, Goldberg  MS, Konstantinopoulos  PA, D’Andrea  AD.  DNA damage and repair biomarkers of immunotherapy response.  Cancer Discov. 2017;7(7):675-693. doi:10.1158/2159-8290.CD-17-0226PubMedGoogle ScholarCrossref
13.
Chen  J, Crutchley  J, Zhang  D, Owzar  K, Kastan  MB.  Identification of a DNA damage-induced alternative splicing pathway that regulates p53 and cellular senescence markers.  Cancer Discov. 2017;7(7):766-781. doi:10.1158/2159-8290.CD-16-0908PubMedGoogle ScholarCrossref
14.
Tommiska  J, Bartkova  J, Heinonen  M,  et al.  The DNA damage signalling kinase ATM is aberrantly reduced or lost in BRCA1/BRCA2-deficient and ER/PR/ERBB2-triple-negative breast cancer.  Oncogene. 2008;27(17):2501-2506. doi:10.1038/sj.onc.1210885PubMedGoogle ScholarCrossref
15.
Sanchez-Vega  F, Mina  M, Armenia  J,  et al; Cancer Genome Atlas Research Network.  Oncogenic signaling pathways in the Cancer Genome Atlas.  Cell. 2018;173(2):321-337.e10. doi:10.1016/j.cell.2018.03.035PubMedGoogle ScholarCrossref
16.
Kotler  E, Shani  O, Goldfeld  G,  et al.  A systematic p53 mutation library links differential functional impact to cancer mutation pattern and evolutionary conservation.  Mol Cell. 2018;71(5):873. doi:10.1016/j.molcel.2018.08.013PubMedGoogle ScholarCrossref
17.
Mitui  M, Nahas  SA, Du  LT,  et al.  Functional and computational assessment of missense variants in the ataxia-telangiectasia mutated (ATM) gene: mutations with increased cancer risk.  Hum Mutat. 2009;30(1):12-21. doi:10.1002/humu.20805PubMedGoogle ScholarCrossref
18.
Kumar  V, Alt  FW, Oksenych  V.  Functional overlaps between XLF and the ATM-dependent DNA double strand break response.  DNA Repair (Amst). 2014;16:11-22. doi:10.1016/j.dnarep.2014.01.010PubMedGoogle ScholarCrossref
19.
Cerami  E, Gao  J, Dogrusoz  U,  et al.  The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data.  Cancer Discov. 2012;2(5):401-404. doi:10.1158/2159-8290.CD-12-0095PubMedGoogle ScholarCrossref
20.
Rittmeyer  A, Barlesi  F, Waterkamp  D,  et al; OAK Study Group.  Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial.  Lancet. 2017;389(10066):255-265. doi:10.1016/S0140-6736(16)32517-XPubMedGoogle ScholarCrossref
21.
World Medical Association.  World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects.  JAMA. 2013;310(20):2191-2194. doi:10.1001/jama.2013.281053PubMedGoogle ScholarCrossref
22.
Broad Institute.  Genomic Data Analysis Center. https://gdac.broadinstitute.org/. Accessed August 12, 2019.
23.
Samstein  RM, Lee  CH, Shoushtari  AN,  et al.  Tumor mutational load predicts survival after immunotherapy across multiple cancer types.  Nat Genet. 2019;51(2):202-206. doi:10.1038/s41588-018-0312-8PubMedGoogle ScholarCrossref
24.
Bindea  G, Mlecnik  B, Tosolini  M,  et al.  Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer.  Immunity. 2013;39(4):782-795. doi:10.1016/j.immuni.2013.10.003PubMedGoogle ScholarCrossref
25.
Dong  ZY, Zhang  JT, Liu  SY,  et al.  EGFR mutation correlates with uninflamed phenotype and weak immunogenicity, causing impaired response to PD-1 blockade in non-small cell lung cancer.  Oncoimmunology. 2017;6(11):e1356145. doi:10.1080/2162402X.2017.1356145PubMedGoogle ScholarCrossref
26.
Gainor  JF, Shaw  AT, Sequist  LV,  et al.  EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.  Clin Cancer Res. 2016;22(18):4585-4593. doi:10.1158/1078-0432.CCR-15-3101PubMedGoogle ScholarCrossref
27.
Kastenhuber  ER, Lowe  SW.  Putting p53 in context.  Cell. 2017;170(6):1062-1078. doi:10.1016/j.cell.2017.08.028PubMedGoogle ScholarCrossref
28.
Williams  AB, Schumacher  B.  p53 in the DNA-damage-repair process.  Cold Spring Harb Perspect Med. 2016;6(5):a026070. doi:10.1101/cshperspect.a026070PubMedGoogle ScholarCrossref
29.
Ciriello  G, Miller  ML, Aksoy  BA, Senbabaoglu  Y, Schultz  N, Sander  C.  Emerging landscape of oncogenic signatures across human cancers.  Nat Genet. 2013;45(10):1127-1133. doi:10.1038/ng.2762PubMedGoogle ScholarCrossref
30.
Skowronska  A, Parker  A, Ahmed  G,  et al.  Biallelic ATM inactivation significantly reduces survival in patients treated on the United Kingdom Leukemia Research Fund Chronic Lymphocytic Leukemia 4 trial.  J Clin Oncol. 2012;30(36):4524-4532. doi:10.1200/JCO.2011.41.0852PubMedGoogle ScholarCrossref
31.
Ding  L, Getz  G, Wheeler  DA,  et al.  Somatic mutations affect key pathways in lung adenocarcinoma.  Nature. 2008;455(7216):1069-1075. doi:10.1038/nature07423PubMedGoogle ScholarCrossref
32.
Jiang  H, Reinhardt  HC, Bartkova  J,  et al.  The combined status of ATM and p53 link tumor development with therapeutic response.  Genes Dev. 2009;23(16):1895-1909. doi:10.1101/gad.1815309PubMedGoogle ScholarCrossref
33.
Beucher  A, Birraux  J, Tchouandong  L,  et al.  ATM and Artemis promote homologous recombination of radiation-induced DNA double-strand breaks in G2.  EMBO J. 2009;28(21):3413-3427. doi:10.1038/emboj.2009.276PubMedGoogle ScholarCrossref
34.
Álvarez-Quilón  A, Serrano-Benítez  A, Lieberman  JA,  et al.  ATM specifically mediates repair of double-strand breaks with blocked DNA ends.  Nat Commun. 2014;5(1):3347. doi:10.1038/ncomms4347PubMedGoogle ScholarCrossref
35.
Blackford  AN, Jackson  SP.  ATM, ATR, and DNA-PK: the trinity at the heart of the DNA damage response.  Mol Cell. 2017;66(6):801-817. doi:10.1016/j.molcel.2017.05.015PubMedGoogle ScholarCrossref
36.
Mao  Z, Bozzella  M, Seluanov  A, Gorbunova  V.  Comparison of nonhomologous end joining and homologous recombination in human cells.  DNA Repair (Amst). 2008;7(10):1765-1771. doi:10.1016/j.dnarep.2008.06.018PubMedGoogle ScholarCrossref
37.
Hsu  FM, Zhang  S, Chen  BP.  Role of DNA-dependent protein kinase catalytic subunit in cancer development and treatment.  Transl Cancer Res. 2012;1(1):22-34.PubMedGoogle Scholar
38.
Kent  LN, Leone  G.  The broken cycle: E2F dysfunction in cancer.  Nat Rev Cancer. 2019;19(6):326-338. doi:10.1038/s41568-019-0143-7PubMedGoogle ScholarCrossref
39.
Sabò  A, Kress  TR, Pelizzola  M,  et al.  Selective transcriptional regulation by Myc in cellular growth control and lymphomagenesis.  Nature. 2014;511(7510):488-492. doi:10.1038/nature13537PubMedGoogle ScholarCrossref
40.
Goel  HL, Mercurio  AM.  VEGF targets the tumour cell.  Nat Rev Cancer. 2013;13(12):871-882. doi:10.1038/nrc3627PubMedGoogle ScholarCrossref
41.
Gettinger  S, Politi  K.  PD-1 axis inhibitors in EGFR- and ALK-driven lung cancer: lost cause?  Clin Cancer Res. 2016;22(18):4539-4541. doi:10.1158/1078-0432.CCR-16-1401PubMedGoogle ScholarCrossref
42.
Nowicki  TS, Hu-Lieskovan  S, Ribas  A.  Mechanisms of resistance to PD-1 and PD-L1 blockade.  Cancer J. 2018;24(1):47-53. doi:10.1097/PPO.0000000000000303PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Oncology
    September 20, 2019

    Association of Tumor Protein p53 and Ataxia-Telangiectasia Mutated Comutation With Response to Immune Checkpoint Inhibitors and Mortality in Patients With Non–Small Cell Lung Cancer

    Author Affiliations
    • 1Fujian Provincial Key Laboratory of Translational Cancer Medicine, Fuzhou, China
    • 2Cancer Bio-immunotherapy Center, Fujian Medical University Cancer Hospital and Fujian Cancer Hospital, Fuzhou, China
    • 3Department of Medical Oncology, Fujian Medical University Cancer Hospital and Fujian Cancer Hospital, Fuzhou, China
    • 4Department of Pathology, Fujian Medical University Cancer Hospital and Fujian Cancer Hospital, Fuzhou, China
    • 5Geneplus-Beijing Institute, Beijing, China
    • 6Fujian Medical University Cancer Hospital, Fuzhou, China
    • 7China Certification and Inspection Group, Kuok Kim Medical Center III, Macao, China
    • 8Hui Xian Medical Center, Macao, China
    • 9Department of Radiation Oncology, Fujian Medical University Cancer Hospital and Fujian Cancer Hospital, Fuzhou, China
    JAMA Netw Open. 2019;2(9):e1911895. doi:10.1001/jamanetworkopen.2019.11895
    Key Points español 中文 (chinese)

    Question  What are the prevalence and association of tumor protein p53 (TP53) and ataxia-telangiectasia mutated (ATM) comutation with response to immune checkpoint inhibitors in patients with non–small cell lung cancer (NSCLC)?

    Findings  In this multiple-cohort study, TP53 and ATM comutation sites were scattered throughout the genes analyzed. Comutation in TP53 and ATM was associated with a higher tumor mutation burden and better overall survival compared with sole mutations and no mutation.

    Meaning  Patients with TP53 and ATM comutation compose a subgroup of patients with NSCLC associated with an increased tumor mutation burden and better response to immune checkpoint inhibitors; TP53 and ATM may be a clinically relevant biomarker in guiding immunotherapy treatment of NSCLC.

    Abstract

    Importance  Immune checkpoint inhibitors (ICIs) can elicit durable antitumor responses in patients with non–small cell lung cancer (NSCLC), but only 20% to 25% of patients respond to treatment. As important genes in the DNA damage response pathway, comutation in the tumor protein p53 (TP53) and ataxia-telangiectasia mutated (ATM) genes may be associated with genomic instability and hypermutation. However, the prevalence of TP53 and ATM comutation and its association with response to ICIs are not fully understood.

    Objective  To examine the prevalence of the TP53 and ATM comutation, the potential mechanism, and its association with response to ICIs among patients with NSCLC.

    Design, Setting, and Participants  This multiple-cohort study included patients with NSCLC from the Geneplus Institute, the Cancer Genome Atlas (TCGA), and the Memorial Sloan Kettering Cancer Center (MSKCC) databases and from the POPLAR and OAK randomized controlled trials. Samples in the Geneplus cohort were collected and analyzed from April 30, 2015, through February 28, 2019. Data from TCGA, the MSKCC, and the POPLAR and OAK cohorts were obtained on January 1, 2019, and analyzed from January 1 to April 10, 2019. Next-generation sequencing assays were performed on tumor samples by the Geneplus Institute. Genomic, transcriptomic, and clinical data were obtained from TCGA and MSKCC databases.

    Exposures  Comprehensive genetic profiling was performed to determine the prevalence of TP53 and ATM comutation and its association with prognosis and response to ICIs.

    Main Outcomes and Measures  The main outcomes were TP53 and ATM comutation frequency, overall survival (OS), progression-free survival, gene set enrichment analysis, and immune profile in NSCLC.

    Results  Patients with NSCLC analyzed in this study included 2020 patients in the Geneplus cohort (mean [SD] age, 59.5 [10.5] years; 1168 [57.8%] men), 1031 patients in TCGA cohort (mean [SD] age, 66.2 [9.5] years; 579 [56.2%] men), 1527 patients in the MSKCC cohort (662 [43.4%] men), 350 patients in the MSKCC cohort who were treated with ICIs (mean [SD] age, 61.4 [13.8] years; 170 [48.6%] men), and 853 patients in the POPLAR and OAK cohort (mean [SD] age, 63.0 [9.1] years; 527 [61.8%] men). Sites of TP53 and ATM comutation were found scattered throughout the genes, and no significant difference was observed in the frequency of TP53 and ATM comutation within the histologic subtypes and driver genes. In 5 independent cohorts of patients with NSCLC, TP53 and ATM comutation was associated with a significantly higher tumor mutation burden compared with the sole mutation and with no mutation (TCGA, MSKCC, Geneplus, and POPLAR and OAK cohort). Among patients treated with ICIs in the MSKCC cohort, TP53 and ATM comutation was associated with better OS than a single mutation and no mutation among patients with any cancer (median OS: TP53 and ATM comutation, not reached; TP53 mutation alone, 14.0 months; ATM mutation alone, 40.0 months; no mutation, 22.0 months; P = .001; NSCLC median OS: TP53 and ATM comutation, not reached; TP53 mutation alone, 11.0 months; ATM mutation alone, 16.0 months; no mutation, 14.0 months; P = .24). Similar results were found in the POPLAR and OAK cohort in which the disease control benefit rate, progression-free survival, and OS were all greater in patients with the TP53 and ATM comutation compared with the other 3 groups (median progression-free survival: TP53 and ATM comutation, 10.4 months; TP53 mutation, 1.6 months; ATM mutation, 3.5 months; no mutation, 2.8 months; P = .01; median OS: TP53 and ATM comutation, 22.1 months; TP53 mutation, 8.3 months; ATM mutation, 15.8 months; no mutation, 15.3 months; P = .002).

    Conclusions and Relevance  This study’s findings suggest that the TP53 and ATM comutation occurs in a subgroup of patients with NSCLC and is associated with an increased tumor mutation burden and response to ICIs. This suggests that TP53 and ATM comutation may have implications as a biomarker for guiding ICI treatment.

    Introduction

    Recent developments in immune checkpoint inhibitors (ICIs) have improved the survival in a multitude of advanced malignant neoplasms, including non–small cell lung cancer (NSCLC).1-6 However, most patients receiving ICIs do not derive a benefit. An important aspect of immunotherapy is how to identify and develop predictive biomarkers of ICI response.7,8 The commonly used clinically applicable predictive biomarker has been programmed cell death 1 ligand 1 (PD-L1), also known as cluster of differentiation 274 or CD274 (OMIM 605402) determined with immunohistochemistry. The Keynote-024 study1 found that the presence of tumor PD-L1 expression more than 50% was associated with the efficacy of pembrolizumab in first-line therapy. However, the sensitivity and specificity of PD-L1 expression are modest,9 which has prompted the search for additional tools.10,11 Mutation of mismatch repair (MMR) genes is known contribute to damage to the DNA damage response (DDR) pathway, which is associated with an increase in the tumor mutation burden (TMB),12 including catalytic subunit of DNA polymerase epsilon (POLE) (OMIM 174762) gene; DNA polymerase δ 1, catalytic subunit (POLD1) (OMIM 174761); breast cancer susceptibility gene 1 (BRCA1) (OMIM 113705); and breast cancer susceptibility gene 2 (BRCA2) (OMIM 600185), which are associated with the efficacy of ICIs in treating lung cancer, but the frequency of occurrence among patients with lung cancer is low.13-15

    The tumor protein p53 (TP53) (OMIM 191170) tumor suppressor gene encodes the p53 transcription factor and is the most commonly mutated gene in human cancers. Under various cellular stress conditions, p53 is activated to inhibit transformation by inducing cell cycle arrest, DNA damage repair, senescence, or apoptosis.16 Loss of ataxia-telangiectasia mutated (ATM) (OMIM 607585) function is associated with the autosomal recessive disease ataxia-telangiectasia, cerebellar degeneration, hypersensitivity to ionizing radiation, cancer susceptibility, immunodeficiency, and genomic instability.17 Human tumors deficient of ATM frequently display chemotherapy resistance and poor survival.18 The aim of this study was to describe an integrative analysis of TP53 and ATM comutation in the Cancer Genome Atlas (TCGA) database,15 Geneplus database, Memorial Sloan Kettering Cancer Center (MSKCC) database,19 and the POPLAR4 and OAK20 cohorts to highlight the importance of validation of the TP53 and ATM comutation for the delivery of precision immunotherapy.

    Method
    Patients and Samples

    From April 30, 2015, through February 28, 2019, 17 814 patients, including 2020 patients with NSCLC, underwent a next-generation sequencing (NGS) assay in the Geneplus-Beijing Institute, Beijing, China. All procedures were conducted in accordance with the Declaration of Helsinki21 and with approval from the ethics committee of Fujian Provincial Cancer Hospital. Written informed consent was obtained from all participants. The study was conducted using the Strengthening the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

    Sequencing and Analysis

    Comprehensive genomic profiling for the Chinese cohort was performed using customized panels of 59 genes or 1021 genes (eTable 1 in the Supplement). Details of sample processing, DNA extraction, library preparation, target capture, NGS, and analysis are described in eMethods 1, 2, and 3 in the Supplement.

    NSCLC Data

    Somatic mutation data for 11 097 solid tumor samples, including 1031 NSCLC samples, in the TCGA database15 and 1527 NSCLC samples in the MSKCC database were downloaded from cBioPortal.19 Gene expression data in fragments per kilobase of transcripts per million mapped reads (FPKM) for 969 NSCLC samples in the TCGA database15 were obtained from the Broad Institute Genomic Data Analysis Center.22

    A total of 1662 patients, including 350 patients with NSCLC, treated at MSKCC23 received at least 1 dose of ICIs with overall survival (OS) defined from the date of first infusion of any ICI. More characteristics of the patients treated with ICIs in the MSKCC database23 are presented in eTable 2 in the Supplement.

    For the POPLAR4 and OAK20 cohort, clinical and somatic mutation data were obtained from a previous study. The POPLAR4 and OAK20 cohort included data from 853 patients with NSCLC. A total of 429 patients received atezolizumab, while 424 patients received docetaxel. More characteristics of patients who received atezolizumab are presented in eTable 3 in the Supplement. Details about data sources are presented in eFigure 1 in the Supplement.

    Assessment of TMB

    The TMB was defined as the number of somatic nonsynonymous variations, insertions, and deletions in examined coding regions detected in tumor tissues by whole-exon sequencing in TCGA15 and NGS in the MSKCC19 and Geneplus databases. In the MSKCC cohort19 of patients with NSCLC, 341 samples underwent targeted NGS (ie, integrated mutation profiling of actionable cancer targets) with a customized panel of 341 genes, while 1186 samples were analyzed with a panel of 410 genes. The TMB was compared among patients who received integrated mutation profiling of actionable cancer targets testing with the same designed panel. In the POPLAR4 and OAK20 cohort, the TMB was evaluated as somatic nonsynonymous variations, insertions, and deletions detected in blood samples using a companion diagnostic assay (FoundationOne).

    Gene Set Enrichment Analysis

    The expression value was transformed by log2(FPKM + 1) for further analysis. Based on the hallmark gene sets, Gene Set Enrichment Analysis software version 3.0 (Broad Institute) was used to identify significantly altered gene sets (false discovery rate ≤ 0.10) among the groups. For significantly enriched pathways in the comutated group, single-sample gene set enrichment analysis was used to calculate the enrichment score in individual samples. The rank sum test was performed to evaluate the statistical difference. To measure the relative levels of tumor infiltrating lymphocytes subsets, published signature gene sets were assessed by single-sample gene set enrichment analysis.24

    Statistical Analysis

    Pearson χ2 test or Fisher exact test were used to assess categorical variables. Differences between the 2 groups were examined with 2-tailed unpaired t test for normally distributed variables or with the Mann-Whitney test for nonnormally distributed variables. Kaplan-Meier survival and multivariate Cox regression analyses were used to analyze associations between mutation type and survival. Statistical analyses were performed using SPSS statistical software version 23.0 (SPSS) and Prism analysis and graphic software version 8.0.1 (GraphPad). A 2-sided P value of less than .05 was considered statistically significant. Data were analyzed from January 1, 2019, to April 10, 2019.

    Results

    Our study included 17 814 patients in the Geneplus cohort, including 2020 patients with NSCLC (mean [SD] age, 59.5 [10.5] years; 1168 [57.8%] men). Our study also found 4 cohorts in the literature for analyses, including 1031 patients with NSCLC in the TCGA cohort15 (mean [SD] age, 66.2 [9.5] years; 579 [56.2%] men, 398 [38.6%] women, and 54 [5.2%] unknown sex), 1527 patients with NSCLC in the MSKCC cohort19 (662 [43.4%] men), 1662 patients in the MSKCC cohort who were treated with ICIs23 (mean [SD] age, 61.4 [13.8] years; 1034 [62.2%] men), including with 350 patients with NSCLC (170 [48.6%] men), and 853 patients in the POPLAR4 and OAK20 cohort (mean [SD] age, 63.0 [9.1] years; 527 [61.8%] men).

    Distribution and Clinical Implications of the TP53 and ATM Comutation Profile Landscape

    We found the TP53 and ATM comutation in 37 of 1031 patients with NSCLC (3.6%) in the TCGA database15 and 52 of 2020 patients with NSCLC (2.6%) in the Geneplus database (eFigure 2 in the Supplement). Subsequently, we surveyed mutation sites in TP53 and ATM between comutated and singularly mutated samples. The TP53 and ATM mutations were found scattered throughout the genes in comutated samples (eFigure 3 in the Supplement). Moreover, 532 of 1031 patients with NSCLC (54.5%) in the TCGA database15 and 1238 of 1527 patients with NSCLC (81.1%) in the MSKCC database19 had lung adenocarcinoma. We did not observe significant differences in the TP53 and ATM comutation frequency within the histologic subtypes (Figure 1A). Next, we investigated the mutation pattern of driver genes among patients with NSCLC who had the TP53 and ATM comutation. Similarly, 10.8% of patients in the TCGA cohort,15 16.0% of patients in the MSKCC cohort,19 and 36.5% of patients in the Geneplus cohort who had the TP53 and ATM comutation also had epidermal growth factor receptor (EGFR) (OMIM 131550) mutations (Figure 1B). Among patients with NSCLC and the TP53 and ATM comutation, 8.1% of patients in the TCGA cohort15 and 7.7% of patients in the Geneplus cohort also had anaplastic lymphoma kinase (ALK) tyrosine kinase receptor (OMIM 105590) fusion (Figure 1B). Neither concurrent nor exclusive mutation patterns were identified between driver genes and the TP53 and ATM comutation.

    Association of TP53 and ATM Comutation With Increasing TMB

    To determine whether the TP53 and ATM comutation had a significant association with an increased TMB, we compared the mutation load of samples among patients who had the comutation, the TP53 mutation alone, the ATM mutation alone, or no mutation in 5 independent NSCLC cohorts: TCGA,15 MSKCC 341 NGS panel genes, MSKCC 410 NGS panel genes,19 Geneplus, and POPLAR4 and OAK.20 All comparisons indicated that the TP53 and ATM comutation was associated with a significantly higher TMB compared with the other 3 groups in all cohorts. In the TCGA cohort,15 the median (interquartile range [IQR]) TMB was 414.0 (207.5-766.0) mutations among patients with TP53 and ATM comutation, 251.5 (162.0-412.3) mutations among patients with TP53 mutation alone (P = .002), 205.0 (129.3-341.0) mutations among patients with ATM mutation alone (P = .003), and 122.0 (54.0-245.3) mutations among patients with no mutation (P < .001). Among patients in the MSKCC cohort19 who underwent the 341 panel NGS, median (IQR) TMB was 21.0 (10.0-26.5) mutations among patients with TP53 and ATM comutation, 7.0 (4.0-12.0) mutations among patients with TP53 mutation alone (P < .001), 9.0 (7.8-13.8) mutations among patients with ATM mutation alone (P = .04), and 4.0 (2.0-7.0) mutations among patients with no mutations (P < .001). Among patients in the MSKCC cohort19 who underwent the 410 panel NGS, median (IQR) TMB was 14.0 (9.0-21.0) mutations among patients with TP53 and ATM comutation, 7.0 (4.0-12.0) mutations among patients with TP53 mutation alone (P < .001), 8.0 (5.0-11.0) mutations among patients with ATM mutation alone (P < .001), and 4.0 (2.0-7.0) mutations among patients with no mutation (P < .001). In the Geneplus cohort, the median (IQR) TMB was 13.5 (6.3-23.8) mutations among patients with TP53 and ATM comutation, 6.0 (4.0-10.) mutations among patients with TP53 mutation alone (P < .001), 5.0 (4.0-10.3) mutations among patients with ATM mutation alone (P < .001), and 3.0 (2.0-6.0) mutations among patients with no mutation (P < .001). In the POPLAR4 and OAK20 cohort, the median (IQR) TMB was 19.0 (12.8-31.5) mutations among patients with TP53 and ATM comutation, 11.0 (6.0-20.0) mutations among patients with TP53 mutation alone (P < .001), 6.5 (3.0-14.8) mutations among patients with ATM mutation alone (P < .001), and 5.0 (3.0-9.0) mutations among patients with no mutation (P < .001) (Figure 2). We found a similar association of the degree of the TMB with TP53 and ATM comutation and MMR genes, POLE/D1, and BRCA1/2 mutation (eFigure 4A and B in the Supplement). In addition, driver genes, such as the EGFR mutation, were associated with a decreased TMB and impaired response to ICIs in patients with NSCLC.25,26 However, analysis of the TMB among patients with the EGFR mutation, EGFR wild type, or TP53 and ATM comutation with or without EGFR mutation found that when an EGFR mutation occurred with the TP53 and ATM comutation, patients still exhibited a high TMB level (eFigure 4C in the Supplement).

    Association of TP53 and ATM Comutation With Response to ICIs

    Non–small cell lung cancer tumors among patients with the TP53 and ATM comutation had a significantly increased TMB, so we used publicly available trial data to investigate whether these patients could benefit from ICIs. In the MSKCC cohort,23 there were 1662 patients with any cancer and 350 patients with NSCLCs who had undergone NGS and had received at least 1 dose of ICI therapy (eTable 2 in the Supplement). A total of 41 patients with any cancer and 8 patients with NSCLC specifically were found to have the TP53 and ATM comutation. We found that a TP53 and ATM comutation was associated with better OS than a TP53 mutation alone, an ATM mutation alone, and no mutation among patients with any cancer (NSCLC median OS: TP53 and ATM comutation, not reached; TP53 mutation alone, 11.0 months; ATM mutation alone, 16.0 months; no mutation 14.0 months, P = .24; any cancer median OS: TP53 and ATM comutation; TP53 mutation alone, 14.0 months; ATM mutation alone, 40.0 months; no mutation, 22.0 months; P < .001) (Figure 3; eFigure 5, eTable 4, and eTable 5 in the Supplement).

    In the POPLAR4 and OAK20 cohort, a total of 429 patients received atezolizumab, and 17 patients were identified with a TP53 and ATM comutation (eTable 3 in the Supplement). Disease control rate, progression-free survival, and OS were all greater in patients with the TP53 and ATM comutation compared with the other 3 groups (progression-free survival: TP53 and ATM comutation, 10.4 months; TP53 mutation alone, 1.6 months; ATM mutation alone, 3.5 months; no mutation, 2.8 months, P = .01; median OS: TP53 and ATM, 22.1 months; TP53 mutation alone, 8.3 months; ATM mutation alone, 15.8 months; no mutation, 15.3 months; P = .002) (Figure 4A and B; eFigure 6 and eTable 6 in the Supplement). Increased progression-free survival remained statistically significant with adjustment for sex, age, Eastern Cooperative Oncology Group performance status, histologic examination results, TMB, MMR genes, POLE/D1, and BRCA1/2 (hazard ratio, 0.48 [95% CI, 0.28-0.84]; P = .001) (eTable 7 in the Supplement).

    Gene Signatures and Pathways Associated With TP53 and ATM Comutation

    To further explore the distinct phenotypic and immunologic states associated with the TP53 and ATM comutation, we expanded the analysis to TCGA15 RNA-sequence data sets using 969 samples that had paired whole-exon sequencing information. Checkpoint ligand expression of PD-L1 was significantly higher among the comutated group (FPKM, 87.1) compared with the no mutation group (FPKM, 54.7) (eFigure 7 in the Supplement). Gene set enrichment analysis based on hallmark gene sets further identified several signaling pathways that were significantly altered (false discovery rate ≤ 0.10), including higher activation of E2F targets, MYC targets, G2M checkpoint, MTORC1 signaling, DNA repair, unfolded protein response, spermatogenesis, KRAS signaling, mitotic spindle, and hypoxia pathways in people with comutated TP53 and ATM (eFigure 8 in the Supplement). Interestingly, we found that the gene set associated with angiogenesis was significantly downregulated among people with the TP53 and ATM comutation compared with people with only TP53 or ATM mutation and people with no mutation. However, by comparing the immune landscape among the groups, we did not observe a significant difference within any immune cell subpopulation (eFigure 7 in the Supplement).

    Discussion

    In this study, we found that comutation of the DDR-related genes TP53 and ATM, regardless of the EGFR mutation and status of other DDR-related genes, was associated with a higher TMB and an improved response to ICIs in patients with NSCLC. The p53 protein promotes either the elimination or repair of damaged cells after DNA damage and stimulates DNA repair by activating target genes that encode components of the DNA repair machinery.27,28 A TP53 mutation can correlate with patterns of single-nucleotide variants and specific comutated genes.29 An ATM deficiency is likely a selected genomic aberration in multiple malignant tumors because of its protection from p53-driven apoptosis.30,31 Beyond mediating apoptosis, ATM also plays a role with TP53 in DNA double-strand break repair32 and is required for efficient repair of double-strand breaks induced in heterochromatin33 or with blocked DNA ends.34 Nonhomologous end joining and homologous recombination are the 2 major pathways for the repair of double-strand breaks.35,36 The TP53 and ATM comutation in cancer cells may lead to an homologous recombination deficiency that results in a greater dependency on nonhomologous end joining pathways. Moreover, nonhomologous end joining modifies the broken DNA ends and ligates them together with no regard for homology, generating deletions or insertions.35-37 Theoretically, TP53 and ATM comutation may cause cancer cell resistance to apoptosis and thus accumulate mutations over time.

    In this study, we found that some cell cycle–related pathways, such as E2F targets and MYC targets, invasive-related hypoxia, and the angiogenesis pathway were overactivated. This indicated an aggressive property and poor survival of TP53 and ATM comutated tumors.38-40 We found that TP53 and ATM comutation was associated with a significantly increased TMB in large independent cohorts with no difference in POLE, POLD1, or MMR genes or BRCA1/2 mutation; PD-L1 expression was significantly upregulated in the comutation group. Furthermore, in the POPLAR4 and OAK20 cohort and the MSKCC all-cancer cohort,19 we found a clinical benefit and survival improvement associated with TP53 and ATM comutation.

    Recent studies have shown that EGFR mutations are associated with a low TMB, an uninflamed tumor microenvironment, and weak immunogenicity, which are associated with an inferior response to programmed cell death protein 1 and PD-L1 blockade in NSCLC.25,41,42 We found that patients who had EGFR mutations accompanied by TP53 and ATM comutation still had a high TMB, but whether this small subgroup of patients would benefit from ICIs needs further confirmation in randomized clinical trials. A 2017 study by Dong et al25 found that GTPase (KRAS) (OMIM 190070) mutation, a TP53 and KRAS proto-oncogene, may boost PD-L1 expression, T-cell infiltration, and augment tumor immunogenicity, resulting in a response to programmed cell death protein 1 and PD-L1 inhibitor. This indicates there may be clinical relevance for NSCLC subtyping by driver mutation genes. However, our data suggest that members of the DDR pathway, such as TP53 and ATM, potentially result in genomic instability and further lead to a high TMB. This phenomenon is independent of driver mutation status and indicated there may be a subpopulation of patients who have a better chance of benefiting from ICI treatment.

    This finding may have important implications for clinical practice, and we recommend TP53 and ATM screening for patients with NSCLC. Even in patients with EGFR and other driver genes mutations, TP53 and ATM comutation may be associated with an additional clinical benefit for ICI therapy.

    Limitations

    Our study has limitations. Despite the TP53 and ATM comutation being associated with increasing TMBs in 5 large independent NSCLC cohorts, the small number of TP53 and ATM mutation tumors and few patients who received ICIs in the MSKCC23 and POPLAR4 and OAK20 cohorts with a recorded survival advantage were not well reflected in the Kaplan-Meier survival analysis. This indicates that our results should be interpreted with caution, and further additional prospective clinical trials of checkpoint blockade in patients with TP53 and ATM comutation and NSCLC are warranted. Additionally, TP53 and ATM comutation also occurred in many other cancer types, suggesting that it may be a generalized cancer phenotype (eFigure 2 in the Supplement). The mechanisms underlying the association of TP53 and ATM comutation with a better prognosis for ICI treatment and a higher TMB in other cancer types are still unclear. The full implications of TP53 and ATM comutation remain elusive and require further study.

    Conclusions

    Our findings suggest that the TP53 and ATM comutation occurs in a subgroup of patients with NSCLC and is associated with an increased TMB and response to ICIs. Comutation of TP53 and ATM may have implications as a potential biomarker for guiding ICI immunotherapy.

    Back to top
    Article Information

    Accepted for Publication: August 4, 2019.

    Published: September 20, 2019. doi:10.1001/jamanetworkopen.2019.11895

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Chen Y et al. JAMA Network Open.

    Corresponding Author: Zeng-Qing Guo, MD, Department of Medical Oncology (guozengqing@fjmu.edu.cn), and Jian-Ji Pan, MD, Department of Radiation Oncology (panjianji1956@fjmu.edu.cn), Fujian Medical University Cancer Hospital and Fujian Cancer Hospital, Fuzhou, 350000, China.

    Author Contributions: Drs Guo and J.-J. Pan had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Y. Chen, G. Chen, J. Li and Huang contributed equally to this work.

    Concept and design: Y. Chen, G. Chen, Lin, L.-Z. Chen, L. K. Pan, Yi, C.-B. Chen, Zheng, Guo.

    Acquisition, analysis, or interpretation of data: Y. Chen, J. Li, Huang, Y. Li, Lu, Y.-Q. Wang, C.-X. Wang, Xia, Guo, J.-J. Pan.

    Drafting of the manuscript: Y. Chen, G. Chen, J. Li, Lin, L.-Z. Chen, Lu, C.-X. Wang, L. K. Pan, C.-B. Chen, Zheng, Guo.

    Critical revision of the manuscript for important intellectual content: Y. Chen, J. Li, Huang, Y. Li, Y.-Q. Wang, Xia, Yi, J.-J. Pan.

    Statistical analysis: Y. Chen, J. Li, Huang, Y. Li, C.-X. Wang.

    Administrative, technical, or material support: Y. Chen, J. Li, Huang, Y. Li, Y.-Q. Wang, L. K. Pan, Guo, J.-J. Pan.

    Supervision: Xia, Yi.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: The article was supported by the National Natural Science Foundation of China (U1705282); Natural Science Foundation of Fujian Province (2017J01259, 2018J01267); Fujian Provincial Health and Family Planning Research Talent Training Program (2018-ZQN-13, 2016-1-11); Joint Funds for the Innovation of Science and Technology, Fujian Province (2017Y9077); and the National Clinical Key Specialty Construction Program.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: David B. Solit, MD, Timothy A. Chan, MD, and Luc G. T. Morris, MD (Memorial Sloan Kettering Cancer Center [MSKCC]), recruited the participants treated with immune checkpoint inhibitors in the MSKCC cohort. Louis Fehrenbacher, MD (Kaiser Permanente Medical Center), and David R. Gandara, MD (University of California, Davis Comprehensive Cancer Center), recruited the participants in the POPLAR and OAK cohort. None of these individuals were compensated for their contributions.

    References
    1.
    Reck  M, Rodríguez-Abreu  D, Robinson  AG,  et al; KEYNOTE-024 Investigators.  Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer.  N Engl J Med. 2016;375(19):1823-1833. doi:10.1056/NEJMoa1606774PubMedGoogle ScholarCrossref
    2.
    Carbone  DP, Reck  M, Paz-Ares  L,  et al; CheckMate 026 Investigators.  First-line nivolumab in stage IV or recurrent non-small-cell lung cancer.  N Engl J Med. 2017;376(25):2415-2426. doi:10.1056/NEJMoa1613493PubMedGoogle ScholarCrossref
    3.
    Herbst  RS, Baas  P, Kim  DW,  et al.  Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial.  Lancet. 2016;387(10027):1540-1550. doi:10.1016/S0140-6736(15)01281-7PubMedGoogle ScholarCrossref
    4.
    Fehrenbacher  L, Spira  A, Ballinger  M,  et al; POPLAR Study Group.  Atezolizumab versus docetaxel for patients with previously treated non-small-cell lung cancer (POPLAR): a multicentre, open-label, phase 2 randomised controlled trial.  Lancet. 2016;387(10030):1837-1846. doi:10.1016/S0140-6736(16)00587-0PubMedGoogle ScholarCrossref
    5.
    Borghaei  H, Paz-Ares  L, Horn  L,  et al.  Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer.  N Engl J Med. 2015;373(17):1627-1639. doi:10.1056/NEJMoa1507643PubMedGoogle ScholarCrossref
    6.
     Nivolumab approved for lung cancer.  Cancer Discov. 2015;5(5):OF1. doi:10.1158/2159-8290.CD-NB2015-042PubMedGoogle ScholarCrossref
    7.
    Langer  CJ, Gadgeel  SM, Borghaei  H,  et al; KEYNOTE-021 Investigators.  Carboplatin and pemetrexed with or without pembrolizumab for advanced, non-squamous non-small-cell lung cancer: a randomised, phase 2 cohort of the open-label KEYNOTE-021 study.  Lancet Oncol. 2016;17(11):1497-1508. doi:10.1016/S1470-2045(16)30498-3PubMedGoogle ScholarCrossref
    8.
    Hellmann  MD, Rizvi  NA, Goldman  JW,  et al.  Nivolumab plus ipilimumab as first-line treatment for advanced non-small-cell lung cancer (CheckMate 012): results of an open-label, phase 1, multicohort study.  Lancet Oncol. 2017;18(1):31-41. doi:10.1016/S1470-2045(16)30624-6PubMedGoogle ScholarCrossref
    9.
    Sacher  AG, Gandhi  L.  Biomarkers for the clinical use of PD-1/PD-L1 inhibitors in non-small-cell lung cancer: a review.  JAMA Oncol. 2016;2(9):1217-1222. doi:10.1001/jamaoncol.2016.0639PubMedGoogle ScholarCrossref
    10.
    Ready  N, Hellmann  MD, Awad  MM,  et al.  First-line nivolumab plus ipilimumab in advanced non-small-cell lung cancer (CheckMate 568): outcomes by programmed death ligand 1 and tumor mutational burden as biomarkers.  J Clin Oncol. 2019;37(12):992-1000. doi:10.1200/JCO.18.01042PubMedGoogle ScholarCrossref
    11.
    Hellmann  MD, Ciuleanu  TE, Pluzanski  A,  et al.  Nivolumab plus ipilimumab in lung cancer with a high tumor mutational burden.  N Engl J Med. 2018;378(22):2093-2104. doi:10.1056/NEJMoa1801946PubMedGoogle ScholarCrossref
    12.
    Mouw  KW, Goldberg  MS, Konstantinopoulos  PA, D’Andrea  AD.  DNA damage and repair biomarkers of immunotherapy response.  Cancer Discov. 2017;7(7):675-693. doi:10.1158/2159-8290.CD-17-0226PubMedGoogle ScholarCrossref
    13.
    Chen  J, Crutchley  J, Zhang  D, Owzar  K, Kastan  MB.  Identification of a DNA damage-induced alternative splicing pathway that regulates p53 and cellular senescence markers.  Cancer Discov. 2017;7(7):766-781. doi:10.1158/2159-8290.CD-16-0908PubMedGoogle ScholarCrossref
    14.
    Tommiska  J, Bartkova  J, Heinonen  M,  et al.  The DNA damage signalling kinase ATM is aberrantly reduced or lost in BRCA1/BRCA2-deficient and ER/PR/ERBB2-triple-negative breast cancer.  Oncogene. 2008;27(17):2501-2506. doi:10.1038/sj.onc.1210885PubMedGoogle ScholarCrossref
    15.
    Sanchez-Vega  F, Mina  M, Armenia  J,  et al; Cancer Genome Atlas Research Network.  Oncogenic signaling pathways in the Cancer Genome Atlas.  Cell. 2018;173(2):321-337.e10. doi:10.1016/j.cell.2018.03.035PubMedGoogle ScholarCrossref
    16.
    Kotler  E, Shani  O, Goldfeld  G,  et al.  A systematic p53 mutation library links differential functional impact to cancer mutation pattern and evolutionary conservation.  Mol Cell. 2018;71(5):873. doi:10.1016/j.molcel.2018.08.013PubMedGoogle ScholarCrossref
    17.
    Mitui  M, Nahas  SA, Du  LT,  et al.  Functional and computational assessment of missense variants in the ataxia-telangiectasia mutated (ATM) gene: mutations with increased cancer risk.  Hum Mutat. 2009;30(1):12-21. doi:10.1002/humu.20805PubMedGoogle ScholarCrossref
    18.
    Kumar  V, Alt  FW, Oksenych  V.  Functional overlaps between XLF and the ATM-dependent DNA double strand break response.  DNA Repair (Amst). 2014;16:11-22. doi:10.1016/j.dnarep.2014.01.010PubMedGoogle ScholarCrossref
    19.
    Cerami  E, Gao  J, Dogrusoz  U,  et al.  The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data.  Cancer Discov. 2012;2(5):401-404. doi:10.1158/2159-8290.CD-12-0095PubMedGoogle ScholarCrossref
    20.
    Rittmeyer  A, Barlesi  F, Waterkamp  D,  et al; OAK Study Group.  Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial.  Lancet. 2017;389(10066):255-265. doi:10.1016/S0140-6736(16)32517-XPubMedGoogle ScholarCrossref
    21.
    World Medical Association.  World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects.  JAMA. 2013;310(20):2191-2194. doi:10.1001/jama.2013.281053PubMedGoogle ScholarCrossref
    22.
    Broad Institute.  Genomic Data Analysis Center. https://gdac.broadinstitute.org/. Accessed August 12, 2019.
    23.
    Samstein  RM, Lee  CH, Shoushtari  AN,  et al.  Tumor mutational load predicts survival after immunotherapy across multiple cancer types.  Nat Genet. 2019;51(2):202-206. doi:10.1038/s41588-018-0312-8PubMedGoogle ScholarCrossref
    24.
    Bindea  G, Mlecnik  B, Tosolini  M,  et al.  Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer.  Immunity. 2013;39(4):782-795. doi:10.1016/j.immuni.2013.10.003PubMedGoogle ScholarCrossref
    25.
    Dong  ZY, Zhang  JT, Liu  SY,  et al.  EGFR mutation correlates with uninflamed phenotype and weak immunogenicity, causing impaired response to PD-1 blockade in non-small cell lung cancer.  Oncoimmunology. 2017;6(11):e1356145. doi:10.1080/2162402X.2017.1356145PubMedGoogle ScholarCrossref
    26.
    Gainor  JF, Shaw  AT, Sequist  LV,  et al.  EGFR mutations and ALK rearrangements are associated with low response rates to PD-1 pathway blockade in non-small cell lung cancer: a retrospective analysis.  Clin Cancer Res. 2016;22(18):4585-4593. doi:10.1158/1078-0432.CCR-15-3101PubMedGoogle ScholarCrossref
    27.
    Kastenhuber  ER, Lowe  SW.  Putting p53 in context.  Cell. 2017;170(6):1062-1078. doi:10.1016/j.cell.2017.08.028PubMedGoogle ScholarCrossref
    28.
    Williams  AB, Schumacher  B.  p53 in the DNA-damage-repair process.  Cold Spring Harb Perspect Med. 2016;6(5):a026070. doi:10.1101/cshperspect.a026070PubMedGoogle ScholarCrossref
    29.
    Ciriello  G, Miller  ML, Aksoy  BA, Senbabaoglu  Y, Schultz  N, Sander  C.  Emerging landscape of oncogenic signatures across human cancers.  Nat Genet. 2013;45(10):1127-1133. doi:10.1038/ng.2762PubMedGoogle ScholarCrossref
    30.
    Skowronska  A, Parker  A, Ahmed  G,  et al.  Biallelic ATM inactivation significantly reduces survival in patients treated on the United Kingdom Leukemia Research Fund Chronic Lymphocytic Leukemia 4 trial.  J Clin Oncol. 2012;30(36):4524-4532. doi:10.1200/JCO.2011.41.0852PubMedGoogle ScholarCrossref
    31.
    Ding  L, Getz  G, Wheeler  DA,  et al.  Somatic mutations affect key pathways in lung adenocarcinoma.  Nature. 2008;455(7216):1069-1075. doi:10.1038/nature07423PubMedGoogle ScholarCrossref
    32.
    Jiang  H, Reinhardt  HC, Bartkova  J,  et al.  The combined status of ATM and p53 link tumor development with therapeutic response.  Genes Dev. 2009;23(16):1895-1909. doi:10.1101/gad.1815309PubMedGoogle ScholarCrossref
    33.
    Beucher  A, Birraux  J, Tchouandong  L,  et al.  ATM and Artemis promote homologous recombination of radiation-induced DNA double-strand breaks in G2.  EMBO J. 2009;28(21):3413-3427. doi:10.1038/emboj.2009.276PubMedGoogle ScholarCrossref
    34.
    Álvarez-Quilón  A, Serrano-Benítez  A, Lieberman  JA,  et al.  ATM specifically mediates repair of double-strand breaks with blocked DNA ends.  Nat Commun. 2014;5(1):3347. doi:10.1038/ncomms4347PubMedGoogle ScholarCrossref
    35.
    Blackford  AN, Jackson  SP.  ATM, ATR, and DNA-PK: the trinity at the heart of the DNA damage response.  Mol Cell. 2017;66(6):801-817. doi:10.1016/j.molcel.2017.05.015PubMedGoogle ScholarCrossref
    36.
    Mao  Z, Bozzella  M, Seluanov  A, Gorbunova  V.  Comparison of nonhomologous end joining and homologous recombination in human cells.  DNA Repair (Amst). 2008;7(10):1765-1771. doi:10.1016/j.dnarep.2008.06.018PubMedGoogle ScholarCrossref
    37.
    Hsu  FM, Zhang  S, Chen  BP.  Role of DNA-dependent protein kinase catalytic subunit in cancer development and treatment.  Transl Cancer Res. 2012;1(1):22-34.PubMedGoogle Scholar
    38.
    Kent  LN, Leone  G.  The broken cycle: E2F dysfunction in cancer.  Nat Rev Cancer. 2019;19(6):326-338. doi:10.1038/s41568-019-0143-7PubMedGoogle ScholarCrossref
    39.
    Sabò  A, Kress  TR, Pelizzola  M,  et al.  Selective transcriptional regulation by Myc in cellular growth control and lymphomagenesis.  Nature. 2014;511(7510):488-492. doi:10.1038/nature13537PubMedGoogle ScholarCrossref
    40.
    Goel  HL, Mercurio  AM.  VEGF targets the tumour cell.  Nat Rev Cancer. 2013;13(12):871-882. doi:10.1038/nrc3627PubMedGoogle ScholarCrossref
    41.
    Gettinger  S, Politi  K.  PD-1 axis inhibitors in EGFR- and ALK-driven lung cancer: lost cause?  Clin Cancer Res. 2016;22(18):4539-4541. doi:10.1158/1078-0432.CCR-16-1401PubMedGoogle ScholarCrossref
    42.
    Nowicki  TS, Hu-Lieskovan  S, Ribas  A.  Mechanisms of resistance to PD-1 and PD-L1 blockade.  Cancer J. 2018;24(1):47-53. doi:10.1097/PPO.0000000000000303PubMedGoogle ScholarCrossref
    ×