ORIGINAL ARTICLE Five Year Survival Update From KEYNOTE-010: Pembrolizumab Versus Docetaxel for Previously Treated, Programmed Death-Ligand 1–Positive Advanced NSCLC Roy S. Herbst, MD, PhD,a,* Edward B. Garon, MD,b Dong-Wan Kim, MD,c Byoung Chul Cho, MD,d Radj Gervais, MD,e Jose L. Perez-Gracia, MD, PhD,f Ji-Youn Han, MD,g Margarita Majem, MD, PhD,h Martin D. Forster, M.B.B.S., PhD,i Isabelle Monnet, MD,j Silvia Novello, MD, PhD,k Matthew A. Gubens, MD,l Michael Boyer, M.B.B.S.,m Wu-Chou Su, MD,n Ayman Samkari, MD,o Erin H. Jensen, MS,o Julie Kobie, PhD,o Bilal Piperdi, MD,o Paul Baas, MDp *Corresponding author research grant from Bristol-Myers Squibb; having advisory role and receiving honoraria from AstraZeneca, Roche, Eli Lilly, Bristol-Myers Disclosure: Prof. Herbst reports having commercial research grants Squibb, Merck Sharp & Dohme, a subsidiary of Merck & Co., Inc., from AstraZeneca, Eli Lilly, and Merck; serving as consultant/advisory Kenilworth, NJ, USA, Pfizer, Boehringer Ingelheim, Novartis, Helsinn, board member for AbbVie, AstraZeneca, Biodesix, Bristol-Myers Pierre Fabre, Kyowa Kyrin, and Takeda; and receiving travel support Squibb, Eli Lilly, EMD Serono, Genentech/Roche, Heat Biologics, In- from Boehringer Ingelheim and Eli Lilly. Dr. Forster reports having finity Pharmaceuticals, Loxo Oncology, Merck, Nektar Therapeutics, research grants from AstraZeneca, Boehringer Ingelheim, Merck Sharp Neon Therapeutics, NextCure, Novartis, Pfizer, Sanofi, Seattle Ge- & Dohme, Bristol-Myers Squibb, Daiichi Sankyo, and Merck Sharp & netics, Shire, Spectrum Pharmaceuticals, Symphogen, Tesaro, ARMO Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA; BioSciences, Genmab, and Tocagen; and serving as board member having advisory roles and receiving honoraria from Achilles, Astra- (nonexecutive/independent) for Junshi Biosciences and Immunocore Zeneca, Bayer, Bristol-Myers Squibb, Celgene, Eli Lilly, Guardant, (nonexecutive director). Dr. Garon reports having research grants Merck, Merck Sharp & Dohme, Nanobiotix, Novartis, Pfizer, Pharma- from Novartis, Merck, EMD Serono, Bristol-Myers Squibb, Eli Lilly, Mar, Roche, and Takeda; and receiving support from the UCL/UCLH Genentech, AstraZeneca, Dynavax Technologies, Iovance Bio- NIHR Biomedical Research Centre. Dr. Monnet reports receiving travel therapeutics, and Mirati Therapeutics; receiving honoraria/fees from support from Roche, AstraZeneca, Pfizer, Merck Sharp & Dohme, and ABL-Bio, Xilio, Shionogi, Sanofi, Novartis, Natera, Merck, Glax- Bristol-Myers Squibb. Dr. Novello reports being on speakers bureau/ oSmithKline, Eisai, EMD Serono, Dracen Pharmaceuticals, Boehringer advisor for AMG, AstraZeneca, AbbVie, Boehringer Ingelheim, Eli Lilly, Ingelheim, and Bristol-Myers Squibb. Dr. Kim reports receiving Merck Sharp & Dohme, Celgene, Bristol-Myers Squibb, Takeda, Pfizer, research funding to institution from Alpha Biopharma, Amgen, Astra- Sanofi, Beigene, Janssen, Novartis, and Roche. Dr. Gubens reports Zeneca/Medimmune, Boehringer Ingelheim, Daiichi Sankyo, Hanmi, serving as consultant for AstraZeneca, BeyondSpring, Boehringer Janssen, Merus, Mirati Therapeutics, Merck Sharp & Dohme, Novartis, Ingelheim, Bristol-Myers Squibb, Genentech/Roche, Heron Therapeu- Ono Pharmaceuticals, Pfizer, Roche/Genentech, Takeda, TP Thera- tics, and Takeda; and receiving research support to institution from peutics, Xcovery, and Yuhan; and receiving travel and accommodation Celgene, Merck, Novartis, OncoMed, and Roche. Dr. Boyer reports support for advisory board meeting attendance from Amgen and receiving grants and nonfinancial support from Merck Sharp & Dohme Daiichi Sankyo. Dr. Cho reports receiving research grants and support during the conduct of this study; grants and nonfinancial support from from Novartis, Bayer, AstraZeneca, Mogam Biotechnology Institute, AstraZeneca, Bristol-Myers Squibb, Janssen, and Genentech/Roche; Dong-A ST, Champions Oncology, Janssen, Yuhan, Ono Pharmaceuti- and grants from Amgen, Pfizer, Eli Lilly, and Novartis. Dr. Jensen re- cals, Dizal Pharma, Merck Sharp & Dohme, AbbVie, Medpacto, GI ports being an employee of Merck Sharp & Dohme Corp., a subsidiary Innovation, Eli Lilly, Blueprint Medicines, and Interpark Bio Conver- of Merck & Co., Inc., Kenilworth, NJ, USA. Drs. Samkari, Kobie, and gence Corp; being on consulting or advisory boards for Novartis, Piperdi report being employees of Merck Sharp & Dohme Corp., a AstraZeneca, Boehringer Ingelheim, Roche, Bristol-Myers Squibb, Ono subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA, and owning stock Pharmaceuticals, Yuhan, Pfizer, Eli Lilly, Janssen, Takeda, Merck in Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Sharp & Dohme, Medpacto, Blueprint Medicines, KANAPH Therapeutic Kenilworth, NJ, USA. Dr. Baas reports receiving research grants from Inc., Brigebio Therapeutics, Cyrus Therapeutics, Guardant Health, and serving as advisory board member for Bristol-Myers Squibb and and Oscotec Inc.; having stock ownership for TheraCanVac Inc., Gen- Merck Sharp & Dohme; and serving as advisory board member for curix Inc., BridgeBio Inc., KANAPH Therapeutic Inc., Cyrus Therapeu- Boehringer Ingelheim, Beigene, AstraZeneca, and Pfizer. Dr. Su de- tics, and Interpark Bio Convergence Corp; serving on board of clares no conflict of interest. directors of Gencurix Inc. and Interpark Bio Convergence Corp; having royalty from Champions Oncology; and being a founder of DAAN Bio- A portion of these results were previously presented at the 2020 World therapeutics. Dr. Gervais reports receiving support from Merck Sharp Conference on Lung Cancer, January 28–31, 2021, Worldwide Virtual & Dohme, Bristol-Myers Squibb, Roche, and AstraZeneca for meetings Event. and advisory boards. Dr. Perez-Gracia reports receiving research grants and support from Roche, Bristol-Myers Squibb, Merck Sharp & Author for correspondence: Roy S. Herbst, MD, PhD, Department of Dohme, Ipsen, Eisai, Incyte, Janssen, and Seattle Genetics; being on Medical Oncology, Comprehensive Cancer Center, Yale University speakers bureau and advisory boards for Roche, Bristol-Myers Squibb, School of Medicine, 333 Cedar Street, WWW221, New Haven, CT Ipsen, Eisai, and Merck Sharp & Dohme; and receiving travel support 06520-8028. E-mail: roy.herbst@yale.edu from Roche, Merck Sharp & Dohme, and Bristol-Myers Squibb. Dr. Han ª 2021 Published by Elsevier Inc. on behalf of International reports receiving honoraria from Roche, AstraZeneca, Bristol-Myers Association for the Study of Lung Cancer. Squibb, Merck Sharp & Dohme, and Takeda; having advisory role for AstraZeneca, Bristol-Myers Squibb, Merck Sharp & Dohme, Eli Lilly, ISSN: 1556-0864 Novartis, Pfizer, and Takeda; and receiving research grants from https://doi.org/10.1016/j.jtho.2021.05.001 Roche, Pfizer, and Ono Pharmaceuticals. Dr. Majem reports receiving Journal of Thoracic Oncology Vol. 16 No. 10: 1718–1732 October 2021 5-Year Outcomes in KEYNOTE-010 1719 a Section of Medical Oncology, Yale Comprehensive Cancer Center, Yale University School of Medicine, New Haven, Connecticut b David Geffen School of Medicine, University of California Los Angeles, Los Angeles, California c Seoul National University Hospital, Seoul National University College of Medicine, Seoul, South Korea d Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, South Korea e Centre François Baclesse, Caen, France f Clinica Universidad de Navarra, Pamplona, Spain g Center for Lung Cancer, National Cancer Center, Goyang, South Korea h Hospital de la Santa Creu i Sant Pau, Barcelona, Spain i UCL Cancer Institute/University College London Hospitals, London, United Kingdom j Centre Hospitalier Intercommunal de Créteil, Créteil, France k Department of Oncology, University of Turin, Azienda Ospedaliero Universitaria San Luigi, Turin, Italy l University of California, San Francisco, San Francisco, California m Chris O’Brien Lifehouse, Camperdown, Australia n National Cheng Kung University Hospital, Tainan, Taiwan o Merck & Co., Inc., Kenilworth, New Jersey p The Netherlands Cancer Institute, Amsterdam, the Netherlands Received 4 March 2021; revised 19 April 2021; accepted 11 May 2021 Available online - 26 May 2021 ABSTRACT Our findings confirm pembrolizumab as a standard-of-care treatment in the second-line or later setting. Introduction: In the KEYNOTE-010 study, pembrolizumab improved overall survival (OS) versus docetaxel in patients with  2021 Published by Elsevier Inc. on behalf of International previously treated, advanced NSCLC with programmed death- Association for the Study of Lung Cancer. ligand 1 (PD-L1) tumor proportion score (TPS) 50% and 1%. We report 5-year efficacy and safety follow-up for the Keywords: Pembrolizumab; Non‒small-cell lung cancer; KEYNOTE-010 study. Chemotherapy; PD-L1 Methods: Patients were randomized to pembrolizumab 2 mg/kg or 10 mg/kg once every 3 weeks or docetaxel Introduction 75 mg/m2 once every 3 weeks for up to 35 cycles (2 y). Pembrolizumab, a humanized monoclonal antibody Patients who completed pembrolizumab treatment and against programmed death 1 (PD-1), promotes T cell- subsequently had recurrence could receive second-course mediated antitumor activity by inhibiting the interac- pembrolizumab for up to 17 cycles (1 y). Pembrolizumab tion between PD-1 and its ligands, programmed death- doses were pooled in this analysis. ligand 1 (PD-L1) and programmed death-ligand 2 Results: A total of 1034 patients were randomized (pem- (PD-L2).1 Pembrolizumab has been found to improve brolizumab, n ¼ 691; docetaxel, n ¼ 343). Median study overall survival (OS) compared with standard chemo- follow-up was 67.4 months (range: 60.0‒77.9). The hazard therapy in the first- and second-line or later settings ratio (95% confidence interval) for OS was 0.55 (0.44‒ among patients with advanced or metastatic NSCLC with 0.69) for patients with PD-L1 TPS 50% and 0.70 (0.61‒ a PD-L1 tumor proportion score (TPS) 1% and to 0.80) with PD-L1 TPS 1%. The 5-year OS rates for improve OS when combined with platinum-based pembrolizumab versus docetaxel were 25.0% versus 8.2% chemotherapy in patients with metastatic NSCLC in patients with PD-L1 TPS 50% and 15.6% versus 6.5% regardless of PD-L1 TPS in the first-line setting.2-7 with PD-L1 TPS 1%. Among 79 patients who completed The primary analysis of the phase 2/3 KEYNOTE-010 35 cycles/2 years of pembrolizumab, the OS rate 3 years study (data cutoff, September 30, 2015; median follow- after completion (w5 y from randomization) was 83.0%. up, 13.1 mo) showed significantly improved OS with A total of 21 patients received second-course pem- pembrolizumab monotherapy (2 mg/kg or 10 mg/kg brolizumab; 11 (52.4%) had an objective response after once every 3 weeks) versus docetaxel once every 3 starting the second course and 15 (71.4%) were alive at weeks in patients with previously treated advanced data cutoff. Exploratory biomarker analysis revealed that NSCLC with PD-L1 TPS 50% and PD-L1 TPS 1%.2 The higher tissue tumor mutational burden (175 mutations hazard ratios (HRs) for OS were 0.54 (95% confidence per exome) was associated with improved outcomes with interval [CI]: 0.38‒0.77) with pembrolizumab 2 mg/kg pembrolizumab. and 0.50 (95% CI: 0.36‒0.70) with pembrolizumab 10 Conclusions: Pembrolizumab continued to provide long- mg/kg in patients with PD-L1 TPS 50% and 0.71 (95% term benefit than docetaxel in patients with previously CI: 0.58‒0.88) and 0.61 (95% CI: 0.49‒0.75), respec- treated advanced NSCLC with PD-L1 TPS 50% and 1%. tively, in patients with PD-L1 TPS 1%.2 Because OS was 1720 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 comparable for the two pembrolizumab doses, data were allocated to docetaxel continued treatment for the pooled for later analyses. In previously reported updated maximum duration allowed by local regulations or until analyses with median follow-up of 31.0 and 42.6 months, disease progression, unacceptable toxicity, investigator pembrolizumab continued to show improvement in OS decision, withdrawal of patient consent, intercurrent over docetaxel in patients with PD-L1 TPS 50% and illness preventing continued treatment, noncompliance TPS 1%.8,9 with study treatment or procedures, or loss to follow-up. We report an updated analysis of efficacy and safety Patients who achieved investigator-confirmed complete outcomes for the intent-to-treat population in KEYNOTE- response per immune-related response criteria (irRC) 010 with approximately 5 years of follow-up from after treatment with pembrolizumab for 6 months or randomization to data cutoff, an additional 2 years of longer, and with an additional 2 or more cycles of follow-up since previous analysis.9 In addition, outcomes pembrolizumab beyond the initial date of response, are reported for the 79 patients who completed 35 cy- could discontinue treatment. Patients who discontinued cles/2 years of pembrolizumab treatment (as specified pembrolizumab after achieving complete response, or in the study protocol and consistent with the pem- after 35 cycles/2 years of pembrolizumab, but who brolizumab prescribing information10) and for the 21 experienced disease progression per irRC (determined patients who received second-course pembrolizumab. by investigator) were eligible for up to 17 cycles (1 y) of For the first time, we report the findings from an pembrolizumab retreatment (i.e., second course) if they exploratory biomarker analysis of the prevalence and had received no other anticancer therapy since the last association with outcomes of tissue tumor mutational dose of pembrolizumab. After the KEYNOTE-010 study burden (tTMB) among patients in KEYNOTE-010. met its primary objective, a protocol amendment allowed patients in the docetaxel group who had disease Materials and Methods progression on the study or who had started subsequent anticancer therapy after study participation and then Study Design and Patients experienced disease progression could crossover to KEYNOTE-010 was a multicenter, international trial pembrolizumab 200 mg every 3 weeks for up to 35 cy- that enrolled patients from 202 academic medical centers cles/2 years or until discontinuation criteria were met. in 24 countries (ClinicalTrials.gov, NCT01905657). Eligible patients were aged more 18 years or older with histolog- ically or cytologically confirmed stage IIIB/IV NSCLC with Assessments 1 or more measurable lesion per Response Evaluation Patients were evaluated by computed tomography Criteria in Solid Tumors (RECIST) version 1.1 (RECIST every 9 weeks until week 54 and at 12-week intervals v1.1) by investigator review and PD-L1 TPS 1%, Eastern thereafter, or more frequently if clinically indicated. Cooperative Oncology Group (ECOG) performance status of Response to treatment was evaluated according to 0 or 1, and investigator-determined disease progression RECIST v.1.1 by independent central review and treat- after 2 or more cycles of platinum-based chemotherapy, ment decisions made on the basis of irRC per investi- and an appropriate tyrosine kinase inhibitor for those with gator. After completion of study treatment or an EGFR or ALK alteration. Full inclusion and exclusion discontinuation for reasons other than disease progres- criteria have been published previously.2 sion, disease status was assessed until disease progres- The study protocol and amendments were approved sion, start of an alternative cancer therapy, death, by an investigational review board or ethics committee withdrawal of consent, or loss to follow-up. Disease at each study site. Patients provided written informed status assessments continued with the same schedule consent before participation. during second-course treatment. Patients were con- tacted every 2 months to assess survival status. Treatment Allocation Adverse events (AEs) were monitored to 30 days Patients were randomly assigned (1:1:1) to open- after the end of treatment (90 d for serious AEs) using label pembrolizumab 2 mg/kg once every 3 weeks, the National Cancer Institute Common Terminology pembrolizumab 10 mg/kg once every 3 weeks, or Criteria for Adverse Events (NCI-CTCAE) version 4.0 to docetaxel 75 mg/m2 once every 3 weeks. Randomization grade severity. was stratified according to ECOG performance status (0 PD-L1 expression was evaluated in formalin-fixed versus 1), geographic region (east Asia versus non-east tissue samples from a nonirradiated tumor lesion Asia), and PD-L1 TPS (50% versus 1%‒49%) and (44% of samples were archival, 56% were newly was managed centrally using an interactive voice/web collected)8 at a central laboratory with an immunohis- response system. Patients allocated to pembrolizumab tochemistry assay (Agilent Technologies, Carpinteria, received up to 35 cycles/2 years of treatment; patients CA) with the murine 22C3 antihuman PD-L1 antibody. October 2021 5-Year Outcomes in KEYNOTE-010 1721 Table 1. Baseline Demographic and Disease Characteristics ITT Population Patients Who Received Pembrolizumab, Docetaxel, Completed 35 Cycles of Second-Course Pembrolizumab, Characteristic, n (%) n ¼ 690 n ¼ 343 Pembrolizumab, n ¼ 79 n ¼ 21 Age group <65 y 395 (57.2) 209 (60.9) 55 (69.6) 15 (71.4) 65 y 295 (42.8) 134 (39.1) 24 (30.4) 6 (28.6) Men 425 (61.6) 209 (60.9) 53 (67.1) 16 (76.2) Race White 496 (71.9) 251 (73.2) 56 (70.9) 14 (66.7) Asian 145 (21.0) 72 (21.0) 17 (21.5) 6 (28.6) Black or African American 21 (3.0) 7 (2.0) 5 (6.3) 1 (4.8) Other 10 (1.4) 2 (0.6) 0 0 Missing 18 (2.6) 11 (3.2) 1 (1.3) 0 Geographic region East Asian 128 (18.6) 62 (18.1) 17 (21.5) 6 (28.6) Non-east Asian 562 (81.4) 281 (81.9) 62 (78.5) 15 (71.4) ECOG performance status 0 231 (33.5) 116 (33.8) 25 (31.6) 5 (23.8) 1 455 (65.9) 224 (65.3) 54 (68.4) 16 (76.2) 2 4 (0.6) 2 (0.6) 0 0 Missing 0 1 (0.3) 0 0 Smoking history Current or former 565 (81.9) 269 (78.4) 72 (91.1) 16 (76.2) Never 123 (17.8) 67 (19.5) 7 (8.9) 5 (23.8) Missing 2 (0.3) 7 (2.0) 0 0 Histologic type Squamous 156 (22.6) 66 (19.2) 21 (26.6) 6 (28.6) Nonsquamous 486 (70.4) 240 (70.0) 53 (67.1) 14 (66.7) Mixed histology 6 (0.9) 4 (1.2) 0 0 Other 9 (1.3) 6 (1.7) 1 (1.3) 0 Unknown 33 (4.8) 27 (7.9) 4 (5.1) 1 (4.8) Brain metastasis 104 (15.1) 48 (14.0) 12 (15.2) 3 (14.3) PD-L1 TPS 50% 290 (42.0) 152 (44.3) 58 (73.4) 12 (57.1) 1%‒49% 400 (58.0) 191 (55.7) 21 (26.6) 9 (42.9) EGFR mutation status Mutant 61 (8.8) 26 (7.6) 1 (1.3) 0 Wild type 581 (84.2) 293 (85.4) 68 (86.1) 21 (100.0) Undetermined/missing 48 (7.0) 24 (7.0) 10 (12.7) 0 ALK translocation present Yes 6 (0.9) 2 (0.6) 0 0 No 612 (88.7) 309 (90.1) 70 (88.6) 21 (100.0) Undetermined/missing 72 (10.4) 32 (9.3) 9 (11.4) 0 Prior lines of systemic therapya 1 477 (69.1) 236 (68.8) 63 (79.7) 18 (85.7) 2 198 (28.7) 104 (30.3) 15 (19.0) 3 (14.3) a Excludes adjuvant and neoadjuvant therapies. ECOG, Eastern Cooperative Oncology Group; ITT, intent-to-treat; PD-L1, programmed death-ligand 1; TPS, tumor proportion score. Tumor samples with TPS 1% were considered PD-L1 identified as the biologically optimal threshold across multiple positive. tumor types in pembrolizumab studies using WES.11,12 Tumor TMB status was assessed using whole-exome sequencing (WES) of tumor tissue and matched normal Study Outcomes DNA as previously described.11 A prespecified cutpoint of 175 Primary end points were OS (time from randomiza- mutations per exome (mut/exome) was used to define sub- tion to death from any cause) and progression-free groups with high tTMB (175 mut/exome) versus low tTMB survival (PFS; time from randomization to first docu- (<175 mut/exome). The TMB cutpoint was previously mented disease progression per RECIST v.1.1 by 1722 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 independent central review or death from any cause, excluded from the efficacy analysis population because it whichever occurred first). Safety and overall response was not possible to adequately assess tumor response; rate (ORR) were assessed as secondary end points. however, this patient was included in the safety analysis Evaluation of tTMB and its association with outcomes population. Baseline demographics and disease charac- was an exploratory analysis. teristics were similar between treatment groups in the intent-to-treat population (Table 1). Statistical Analysis Median (range) time from randomization to the data Statistical analysis methods for this trial have been pre- cutoff date of April 8, 2020, was 67.4 (60.0–77.9) months viously reported.2,9 Efficacy analyses were performed ac- for the intent-to-treat population (N ¼ 1033). All pa- cording to the treatment assigned (i.e., intent-to-treat); safety tients had discontinued their initially assigned treatment analyses were conducted among patients who received as of the data cutoff date. Median (range) duration of treatment analyzed according to treatment received (i.e., all treatment for first course was 3.5 months (1 d to 31.7 patients as-treated). The primary end points of OS and PFS mo) for the two pembrolizumab treatment groups in the intent-to-treat population were estimated using the (pooled) and 2.0 months (1 d to 26.4 mo) for docetaxel. Kaplan-Meier method. HRs and 95% CIs were calculated Among patients randomized to the docetaxel group, using a stratified Cox proportional hazards model with eight patients (2.3%) crossed over to pembrolizumab Efron’s tie handling method; randomization stratification on-study, and an additional 68 patients (19.8%) crossed factors were applied to the analyses. Consistent with previ- over to anti‒PD-(L)1 immunotherapy off-study, for an ous analyses, pembrolizumab dose groups were pooled for effective crossover rate of 22.2%. this analysis. No alpha was assigned to these analyses. The tTMB-evaluable population comprised all patients Long-Term Outcomes in the Intent-to-Treat with PD-L1 TPS 1% who received either pembrolizumab Population or docetaxel (all patients as-treated population) and had At the time of analysis, 893 of 1033 patients (86.4%) evaluable samples for tTMB using WES. A statistical anal- in the intent-to-treat population had died. The HR (95% ysis plan for tTMB analysis was prespecified before CI) for OS was 0.55 (0.44–0.69) for patients with PD-L1 merging the clinical and biomarker data sets. Relationships TPS 50% (Fig. 1A) and 0.70 (0.61–0.80) for patients between tTMB and ORR were assessed using logistic with PD-L1 TPS 1% (Fig. 1B). Median OS (95% CI) was regression analysis adjusted for ECOG performance status 16.9 (12.3‒21.4) months versus 8.2 (6.4‒9.8) months in and a receiver-operating characteristic curve analysis. Re- the PD-L1 TPS 50% group and 11.8 (10.4‒13.1) lationships between tTMB and OS and PFS were evaluated months versus 8.4 (7.6‒9.5) months in the PD-L1 TPS using Cox proportional hazards models (adjusted for ECOG 1% group. Kaplan-Meier estimates of the 5-year OS performance status). For the association of tTMB with rate for pembrolizumab versus docetaxel were 25.0% outcomes, tTMB was assessed as a continuous log10- versus 8.2% in the PD-L1 TPS 50% group and 15.6% transformed variable. The Wald test was used to calculate versus 6.5% in the PD-L1 TPS 1% group. Among pa- one-sided p values for pembrolizumab because the a priori tients with PD-L1 TPS 1%–49%, the HR (95% CI) for OS hypothesis was that higher tTMB was positively associated was 0.79 (0.65–0.94). Figure 1C shows HRs for survival with improved outcomes with pembrolizumab. For in key patient subgroups. chemotherapy, two-sided p values were calculated because The HR (95% CI) for PFS (per RECIST v.1.1 by there was no a priori hypothesis regarding the direction of independent central review) was 0.57 (0.46–0.71) for the association between tTMB and outcomes with patients with PD-L1 TPS 50% and 0.84 (0.73–0.96) chemotherapy. Statistical significance was determined at for patients with PD-L1 TPS 1%. Median (95% CI) the 0.05 level; there was no adjustment for multiplicity, PFS was 5.3 (4.2‒6.5) months versus 4.2 (3.8‒5.1) and no alpha was assigned. months in the PD-L1 TPS 50% group and 4.0 (3.1‒ 4.1) months versus 4.1 (3.8‒4.5) months in the PD-L1 Results TPS 1% group. In the PD-L1 TPS 50% group, the Patients PFS rate at 5 years was 18.2% with pembrolizumab; A total of 1034 patients were randomized in all in the docetaxel group had disease progression or KEYNOTE-010 between August 28, 2013, and February were censored before 5 years. The PFS rates at 5 27, 2015, of whom 691 were randomized to pem- years were 9.4% versus 0.7% for the PD-L1 TPS 1% brolizumab (pembrolizumab 2 mg/kg, n ¼ 345; pem- group, respectively (Fig. 2A and B). brolizumab 10 mg/kg, n ¼ 346) and 343 to docetaxel The ORR (95% CI, per RECIST v.1.1 by independent (Supplementary Fig. 1).2 As previously reported, one central review) was 33.1% (27.7–38.8) with pem- patient in the pembrolizumab 2 mg/kg group was brolizumab versus 9.2% (5.1–15.0) with docetaxel for October 2021 5-Year Outcomes in KEYNOTE-010 1723 A PD-L1 TPS ≥50% Events, n HR n (%) (95% CI) Pembrolizumab 290 220 (75.9) 100 0.55 (0.44–0.69) Docetaxel 152 132 (86.8) 90 34.5% 27.7% 25.0% 80 12.6% 10.5% 8.2% 70 60 Median (95% CI) OS, % 16.9 mo (12.3–21.4 mo) 50 8.2 mo (6.4–9.8 mo) 40 30 20 10 0 0 6 12 18 24 30 36 42 48 54 60 66 72 Time, months No. at risk Pembrolizumab 290 217 162 139 118 101 93 78 72 69 63 33 5 Docetaxel 152 88 52 35 25 21 18 14 14 11 9 5 2 B PD-L1 TPS ≥1% Events, n HR n (%) (95% CI) Pembrolizumab 690 584 (84.6) 0.70 (0.61–0.80) 100 Docetaxel 343 309 (90.1) 90 22.9% 17.5% 15.6% 80 11.0% 8.8% 6.5% 70 60 Median (95% CI) OS, % 11.8 mo (10.4–13.1 mo) 50 8.4 mo (7.6–9.5 mo) 40 30 20 10 0 0 6 12 18 24 30 36 42 48 54 60 66 72 Time, months No. at risk Pembrolizumab 690 494 333 267 217 171 148 121 110 101 90 51 11 Docetaxel 343 208 114 73 46 40 35 29 27 20 18 7 3 Figure 1. Kaplan-Meier analysis of OS in patients with (A) PD-L1 TPS 50% and (B) PD-L1 TPS 1% and treatment differences in OS across patient subgroups among patients with (C) PD-L1 TPS 1%. CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; OS, overall survival; PD-L1, programmed death-ligand 1; TPS, tumor proportion score. patients in the PD-L1 TPS 50% group and 21.2% docetaxel in the PD-L1 TPS 50% group and 68.4 (2.0þ (18.2–24.4) versus 9.6% (6.7–13.2) for patients in the to 71.7þ) versus 7.5 (1.4þ to 16.8) months in the PD-L1 PD-L1 TPS 1% group. Median (range) duration of TPS 1% group (“þ” indicates that there was no pro- response (DOR) was 68.4 (2.0þ to 71.7þ) months with gressive disease by the time of last disease assessment). pembrolizumab versus 8.5 (2.6–16.8) months with At data cutoff, 41 patients (43.0%) who received 1724 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 C Subgroup N#/No. of Events HR for OS (95% CI) Overall 1033/893 0.70 (0.61‒0.80) Sex Male 634/551 0.71 (0.60‒0.86) Female 399/342 0.66 (0.53‒0.84) Age <65 604/517 0.62 (0.52‒0.75) ≥65 429/376 0.80 (0.64‒1.01) Race White 747/647 0.66 (0.56‒0.78) Non-White 257/219 0.83 (0.61‒1.13) Region East Asian 190/161 0.81 (0.58‒1.14) Non-East Asian 843/732 0.68 (0.58‒0.79) Baseline ECOG 0 347/287 0.81 (0.63‒1.05) 1 679/600 0.64 (0.54‒0.76) PD-L1 TPS Status TPS ≥50% 442/352 0.55 (0.44‒0.69) TPS 1%–49% 591/541 0.79 (0.65‒0.94) Smoker Current/Exsmoker 834/722 0.69 (0.59‒0.81) Never Smoker 190/165 0.67 (0.47‒0.95) Non-small Cell Histology Squamous 222/197 0.85 (0.62‒1.18) Adenocarcinoma 710/609 0.69 (0.58‒0.81) EGFR Mutation Mutant 87/80 0.90 (0.52‒1.57) Wild type 874/754 0.70 (0.60‒0.82) 0.1 1 10 Estimated Hazard Ratio (HR) Figure 1. (continued). pembrolizumab in the PD-L1 TPS 50% group and 51 fatigue (15.8%), decreased appetite (12.8%), and rash (35.0%) in the PD-L1 TPS 1% group (of whom 10 had (12.2%), whereas the most frequently occurring PD-L1 TPS 1%–49%) had an ongoing response; no pa- treatment-related AEs in the docetaxel group were alo- tient who received docetaxel in either PD-L1 TPS group pecia (34.0%), fatigue (24.9%), and diarrhea (19.1%). had an ongoing response. Treatment-related AEs leading to death occurred in 0.7% Incidence of treatment-related AEs (any grade), grade of patients in the pembrolizumab group and 1.6% in the 3 to 5 AEs, and treatment-related AEs leading to docetaxel group: there were no new treatment-related discontinuation or death was lower in patients treated deaths since the previous analysis.9 Serious treatment- with pembrolizumab than in patients treated with related AEs were reported by similar proportions of docetaxel (Table 2). The most frequently occurring patients in the two treatment groups (pembrolizumab, treatment-related AEs in the pembrolizumab group were 11.3%; docetaxel, 14.2%). October 2021 5-Year Outcomes in KEYNOTE-010 1725 PD-L1 TPS ≥50% A Events, n HR B PD-L1 TPS ≥1% n Events, n HR (%) (95% CI) n Pembrolizumab 290 235 (81.0) (%) (95% CI) 0.57 (0.46–0.71) Pembrolizumab 690 617 (89.4) 100 Docetaxel 152 141 (92.8) 0.84 (0.73–0.96) 100 Progression-Free Survival, % Docetaxel 343 321 (93.6) Progression-Free Survival, % 90 23.5% 19.9% 18.2% 90 80 1.5% 0.8% NR 13.8% 10.7% 9.4% 80 70 2.5% 1.5% 0.7% 70 60 Median (95% CI) 60 Median (95% CI) 5.3 mo (4.2–6.5 mo) 50 4.0 mo (3.1–4.1 mo) 4.2 mo (3.8–5.1 mo) 50 40 4.1 mo (3.8–4.5 mo) 40 30 30 20 20 10 10 0 0 6 12 18 24 30 36 42 48 54 60 66 72 0 Time, months 0 6 12 18 24 30 36 42 48 54 60 66 72 No. at risk Time, months Pembrolizumab 290 137 97 85 74 59 54 44 40 36 24 13 2 No. at risk Docetaxel 152 54 19 13 6 3 2 1 1 0 0 0 0 Pembrolizumab 690 264 163 130 109 89 78 64 55 46 32 19 3 Docetaxel 343 125 48 31 16 10 7 5 4 2 2 0 0 Figure 2. Kaplan-Meier analysis of progression-free survival per RECIST version 1.1 by independent central review in patients with (A) PD-L1 TPS 50% and (B) PD-L1 TPS 1%. CI, confidence interval; HR, hazard ratio; NR, not reached; PD-L1, pro- grammed death-ligand 1; RECIST, Response Evaluation Criteria in Solid Tumors; TPS, tumor proportion score. Immune-mediated AEs and infusion reactions (irre- treatment had died. The OS rate 3 years after completion spective of attribution to treatment by the investigator) of pembrolizumab treatment (w5 years from randomi- occurred in 23.0% of patients in the pembrolizumab zation) was 83.0%. A total of 38 of 79 patients (48.1%) group and 10.0% of patients in the docetaxel group. No were alive without disease progression. additional patients experienced immune-mediated AEs Treatment-related AEs occurred in 66 of 79 patients and infusion reactions since the previous analysis.9 Grade (83.5%), including 14 (17.7%) with grade 3 to 4 3 to 5 immune-mediated AEs and infusion reactions treatment-related AEs (Table 2). One patient received occurred in 6.3% of patients in the pembrolizumab group pembrolizumab for more than 2 years (but did not com- and 1.6% of patients in the docetaxel group. The most plete 35 cycles of pembrolizumab). Immune-mediated frequently occurring immune-mediated AEs in the pem- AEs occurred in 31 of 79 patients (39.2%), most often brolizumab group were hypothyroidism (8.8%), pneu- hypothyroidism (25.3%), hyperthyroidism (8.9%), and monitis (5.9%), and hyperthyroidism (4.8%). pneumonitis (8.9%). Five patients (6.3%) had grade 3/4 immune-mediated AEs (pneumonitis, n ¼ 2; hypophysitis, Long-Term Outcomes in Patients Who Completed n ¼ 2; adrenal insufficiency, n ¼ 1; pancreatitis, n ¼ 1). 35 Cycles/2 Years of Pembrolizumab There were no fatal immune-mediated AEs. At data cutoff, 79 patients had completed 35 cycles/2 years of pembrolizumab. Median (range) time from Patients Who Received Second-Course randomization to data cutoff was 68.1 (60.5‒74.5) Pembrolizumab months for these patients. Baseline characteristics were At data cutoff, 21 patients had received second-course generally similar between these patients and patients pembrolizumab. Demographic and clinical characteristics allocated to pembrolizumab in the intent-to-treat popu- for these patients are summarized in Table 1. Of 21 pa- lation, although a higher percentage of patients were less tients, 11 (52.3%) experienced an objective response per than 65 years of age (69.6% versus 57.2%) and had PD- RECIST by independent central review (complete L1 TPS 50% (73.4% versus 42.0%) in the patients who response, n ¼ 1; partial response, n ¼ 10) after starting completed 35 cycles/2 years of pembrolizumab, whereas second-course pembrolizumab. A further six patients had fewer had received two or more prior lines of systemic stable disease, for an overall disease control rate of therapy (19.0% versus 28.7%) or had mutant EGFR 81.0%. In addition, three patients had progressive dis- status (1.3% versus 8.8%); presence of brain metastases ease per RECIST by independent central review and one at baseline was similar (15.2% versus 15.1%; Table 1). patient was unevaluable. Eight patients experienced ORR (per independent central review per RECIST subsequent disease progression per irRC by investigator v.1.1) for patients who completed 35 cycles/2 years of assessment, including three patients who had achieved a pembrolizumab was 98.7%: 15 of 79 patients (19.0%) partial response and five who had stable disease. At data achieved complete response, 63 (79.7%) achieved par- cutoff, six patients (28.6%) who received second-course tial response; and one (1.3%) further patient had stable pembrolizumab had died. Treatment duration, time to disease. Treatment duration, time to response, and DOR response, and DOR for patients who received second- for patients who completed 35 cycles/2 years of pem- course pembrolizumab are found in Figure 3B. brolizumab are shown in Figure 3A. At data cutoff, 18 of A total of 10 patients (47.6%) experienced 79 patients (22.8%) who completed 35 cycles/2 years of treatment-related AEs after initiation of second-course 1726 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 Table 2. Incidence of Treatment-Related AEs Among Treated Patients Completed 35 Cycles/ 2y of Pembrolizumab, Adverse Events Pembrolizumab, n ¼ 682 Docetaxel, n ¼ 309 n ¼ 79 Treatment-related AEs, n (%) Any 462 (67.7) 255 (82.5) 66 (83.5) Grades 3‒5 110 (16.1) 113 (36.6) 14 (17.7) Led to treatment discontinuation 40 (5.9) 37 (12.0) 1 (1.3) Led to death 5 (0.7) 5 (1.6) 0 Any Grade Grades 3‒5 Any Grade Grades 3‒5 Any Grade Grades 3‒5 Treatment-related AEs occurring in 10% of patients,a n (%) Fatigue 108 (15.8) 10 (1.5) 77 (24.9) 11 (3.6) 15 (19.0) 1 (1.3) Decreased appetite 87 (12.8) 5 (0.7) 52 (16.8) 3 (1.0) 9 (11.4) 1 (1.3) Rash 83 (12.2) 2 (0.3) 14 (4.5) 0 21 (26.6) 0 Nausea 81 (11.9) 3 (0.4) 52 (16.8) 1 (0.3) 9 (11.4) 0 Pruritus 72 (10.6) 2 (0.3) 5 (1.6) 1 (0.3) 22 (27.8) 0 Diarrhea 58 (8.5) 2 (0.3) 59 (19.1) 7 (2.3) 15 (19.0) 0 Asthenia 48 (7.0) 4 (0.6) 38 (12.3) 6 (1.9) 9 (11.4) 0 Anemia 27 (4.0) 5 (0.7) 43 (13.9) 5 (1.6) 4 (5.1) 1 (1.3) Stomatitis 22 (3.2) 1 (0.1) 44 (14.2) 3 (1.0) 4 (5.1) 0 Alopecia 7 (1.0) 0 105 (34.0) 2 (0.6) 3 (3.8) 0 Neutropenia 2 (0.3) 0 44 (14.2) 38 (12.3) 0 0 Hypothyroidism 53 (7.8) 0 1 (0.3) 0 18 (22.8) 0 Pyrexia 40 (5.9) 2 (0.3) 17 (5.5) 1 (0.3) 8 (10.1) 0 Arthralgia 38 (5.6) 2 (0.3) 18 (5.8) 0 8 (10.1) 0 Immune-mediated AEs and infusion reactions,b n (%) Hypothyroidism 60 (8.8) 0 1 (0.3) 0 20 (25.3) 0 Pneumonitis 40 (5.9) 18 (2.6) 6 (1.9) 2 (0.6) 7 (8.9) 2 (2.5) Hyperthyroidism 33 (4.8) 1 (0.1) 3 (1.0) 0 7 (8.9) 0 Infusion reactions 15 (2.2) 3 (0.4) 20 (6.5) 2 (0.6) 0 0 Severe skin reactions 11 (1.6) 7 (1.0) 1 (0.3) 1 (0.3) 2 (2.5) 0 Adrenal insufficiency 6 (0.9) 1 (0.1) 0 0 2 (2.5) 1 (1.3) Colitis 6 (0.9) 4 (0.6) 0 0 1 (1.3) 0 Thyroiditis 6 (0.9) 0 0 0 2 (2.5) 0 Pancreatitis 5 (0.7) 3 (0.4) 0 0 1 (1.3) 1 (1.3) Hypophysitis 4 (0.6) 3 (0.4) 0 0 2 (2.5) 2 (2.5) Myositis 4 (0.6) 0 1 (0.3) 0 0 0 Hepatitis 3 (0.4) 1 (0.1) 0 0 0 0 Type 1 diabetes mellitus 3 (0.4) 3 (0.4) 0 0 0 0 Nephritis 1 (0.1) 1 (0.1) 0 0 0 0 Note: AEs were monitored through 30 days after the end of treatment (90 d for serious AEs). The National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0 were used to grade severity. a Includes events that occurred in 10% of patients in either treatment group or among patients who completed 35 cycles of pembrolizumab. b Events were based on a list of terms specified at the time of analysis and were included regardless of attribution to study treatment or immune relatedness by the investigator. Related terms were included. AE, adverse events. pembrolizumab. Two had grade 3 treatment-related patient in the tTMB-evaluable population was excluded AEs: one patient with pneumonitis and one patient from the efficacy analyses because (as noted previously) with increased alanine aminotransferase, aspartate it was not possible to adequately assess this patient’s aminotransferase, and gamma-glutamyl transferase. No tumor response (Supplementary Fig. 2). Baseline char- grade 4 or 5 treatment-related AEs occurred during the acteristics were similar in the tTMB-evaluable popula- second course. All 10 of these patients had treatment- tion compared with the overall population related AEs during the first course. (Supplementary Table 1). OS, PFS, and ORR outcomes for pembrolizumab versus docetaxel were similar in the Clinical Outcomes in the tTMB-Evaluable tTMB-evaluable population and in the overall efficacy Population population (Supplementary Table 2). Of the 1034 randomized patients, 254 (24.6%) had When assessed as a continuous variable, higher tTMB samples evaluable for analysis of tTMB by WES. One was significantly associated with improved OS, PFS, and October 2021 5-Year Outcomes in KEYNOTE-010 1727 A                                                  CR     PR     irPD  Received Second Course    Death    0 12 24 36 48 60 72 Time, months B      End of first course   Start of second course   Second course CR  Second S d course PR  Second course SD  Second course PD (IRC)  First course treatment not completed  Second course PD (investigator-assessed) First course PD Death 0 6 12 18 24 30 36 42 48 54 60 66 72 78 Time, months Figure 3. (A) Outcomes in patients who completed 35 cycles of pembrolizumab. Bar lengths indicate duration of treatment (dark green) and months of follow-up (light green). Follow-up was defined as date of the last known nonprogression scan or date of last investigator assessment the patient was alive. Response evaluation is per RECIST version 1.1 by independent central assessment. (B) Outcomes in patients who started second-course pembrolizumab. Bar lengths indicate duration of second-course treatment (dark green) and months of second-course follow-up (light green bar after dark green bar). Follow- up was defined as the last known nonprogression scan or date of last investigator assessment the patient was alive. Response 1728 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 ORR for patients receiving pembrolizumab (Wald test, docetaxel.2,9 Consistent with OS, HRs for PFS favored the one-sided p  0.005 for all) but not in patients who pembrolizumab group. Notably, our findings are consis- received docetaxel (Wald test, two-sided p > 0.05 for all) tent with 5-year OS outcomes from the single-arm phase (Supplementary Table 3). There was no correlation be- 1b KEYNOTE-001 study of pembrolizumab (2 or 10 mg/ tween tTMB and PD-L1 TPS in either the pembrolizumab kg) in patients with advanced NSCLC, in which the 5-year group (r ¼ 0.16) or the docetaxel group (r ¼ 0.18) OS rate was 25.0% among previously treated patients (Supplementary Fig. 3). with PD-L1 TPS 50% and 15.5% in previously treated The clinical utility of tTMB as a biomarker for pem- patients with PD-L1 TPS 1%.13 A pooled analysis of the brolizumab was assessed in 253 patients with evaluable phase 3 CheckMate 017 and CheckMate 057 studies tTMB and with available PFS and OS data. Among these revealed a 5-year OS rate of 13.4% with nivolumab versus patients, 132 (52.2%) had tTMB 175 mut/exome (the 2.6% with docetaxel.14 Finally, in an updated analysis of prespecified tTMB cutpoint; pembrolizumab, n ¼ 81; the OAK study, atezolizumab revealed a 4-year OS rate of docetaxel, n ¼ 51) and 121 patients (47.8%) had tTMB 15.5%.15 In all studies, 5-year OS was higher with anti‒ <175 mut/exome (pembrolizumab, n ¼ 83; docetaxel, PD-(L)1 therapy than the historical 5-year relative sur- n ¼ 38). The HR (95% CI) for OS was 0.54 (0.37‒0.79) vival rate of 6.9% for patients with distant metastases in among patients with tTMB 175 mut/exome and 0.87 the United States from 2010 to 2016.16 (0.58‒1.31) in patients with tTMB <175 mut/exome. Pembrolizumab monotherapy revealed long-term re- Similarly, the HR (95% CI) for PFS was 0.61 (0.42‒0.89) sponses in a subset of patients who completed 35 cycles and 1.05 (0.70‒1.56) in the tTMB 175 and tTMB <175 or 2 years of treatment. Responses were durable among mut/exome groups, respectively (Fig. 4A and B). ORR these patients, with some patients having a response was higher among patients with tTMB 175 mut/exome duration exceeding 5 years. Most of these patients who received pembrolizumab versus docetaxel (24.7% (77.2%) were alive at data cutoff. These results are versus 9.8%), whereas in patients with tTMB <175 mut/ consistent with outcomes in patients from KEYNOTE-001, exome, the ORR favored patients who received docetaxel who received 2 or more years of pembrolizumab treat- (16.9% versus 21.1%; Fig. 4C). ment with an ORR of 91% among patients in the previ- ously treated group and a 5-year OS rate of 75.8%.13 Notably, most patients who received second-course Discussion pembrolizumab had disease control (i.e., an objective In this 5-year long-term follow-up analysis of the response or stable disease) during treatment, illustrating KEYNOTE-010 study of pembrolizumab versus docetaxel the effectiveness of retreatment in initial responders at in patients with previously treated, PD-L1‒positive the time of progression after completing 2 years of advanced NSCLC, pembrolizumab continued to improve pembrolizumab treatment. Findings from the KEYNOTE- OS than docetaxel in patients with PD-L1 TPS 50% and 024 study were consistent with these findings: in that TPS 1%. Outcomes among patients in the pem- study, 4 of 12 patients (33.3%) who received second- brolizumab group represented a clinically meaningful course pembrolizumab after disease progression, with improvement over docetaxel with 5-year OS rates of an additional six patients having stable disease.17 25.0% versus 8.2% in patients with PD-L1 TPS 50% and Potentially, a meta-analysis of outcomes among pa- 15.6% versus 6.5% in patients with PD-L1 TPS 1%, tients who have received second-course pembrolizumab despite 22.2% of patients in the docetaxel group crossing treatment in clinical studies may be warranted to pro- over to either pembrolizumab on-study or other anti‒PD- vide more definitive results. An exploratory analysis (L)1 immunotherapies. Median OS among patients in the from the CheckMate 153 study evaluated outcomes pembrolizumab group was 16.9 versus 8.2 months for among patients with previously treated NSCLC who patients in the docetaxel group (HR ¼ 0.55) among those ceased treatment with the anti–PD-1 monoclonal anti- with PD-L1 TPS 50% and 11.8 versus 8.4 months (HR ¼ body nivolumab at 1 year (with an option of subsequent 0.70), respectively, among those with PD-L1 TPS greater retreatment) compared with those who received than or equal to 1%. These data extend and confirm continuous treatment after 1 year and found that median findings from previous analyses of KEYNOTE-010, in PFS was longer for patients in the continuous group which pembrolizumab was found to improve OS versus (24.7 mo versus 9.4 mo).18 assessment is per RECIST version 1.1 by independent central review; PD is per irRC by investigator review, as this was the basis of treatment decisions. *One patient received a second course of pembrolizumab but did not meet eligibility criteria for having completed 35 cycles/2 years of first-course pembrolizumab. CR, complete response; irPD, progressive disease per immune-related response criteria; irRC, immune-related response criteria; PD, progressive disease; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumors; SD, stable disease. October 2021 5-Year Outcomes in KEYNOTE-010 1729 tTMB ≥175 mut/exome tTMB <175 mut/exome A n Events, n HR n Events, n HR (%) (95% CI) (%) (95% CI) 100 66 (81.5) 100 78 (94.0) Pembrolizumab 81 0.54 Pembrolizumab 83 0.87 Docetaxel 51 49 (96.1) (0.37–0.79) Docetaxel 38 33 (86.8) (0.58–1.31) 90 90 80 80 70 70 60 60 Median (95% CI) Median (95% CI) OS, % OS, % 14.1 mo (10.0–19.2 mo) 9.3 mo (8.2–12.5 mo) 50 50 7.6 mo (5.0–10.7 mo) 7.2 mo (4.5–14.3 mo) 40 40 30 30 20 20 10 10 0 0 0 6 12 18 24 30 36 42 48 54 60 66 72 78 0 6 12 18 24 30 36 42 48 54 60 66 72 78 Time, months Time, months No. at Risk No. at Risk Pembrolizumab 81 60 46 32 28 26 22 20 19 16 15 10 2 0 Pembrolizumab 83 58 32 24 16 13 13 7 7 6 5 4 0 0 Docetaxel 51 27 17 11 7 7 7 6 6 3 3 0 0 0 Docetaxel 38 19 12 8 6 6 6 4 2 2 2 0 0 0 tTMB ≥175 mut/exome tTMB <175 mut/exome B n Events, n HR n Events, n HR (%) (95% CI) (%) (95% CI) 100 68 (84.0) 100 79 (95.2) Pembrolizumab 81 0.61 Pembrolizumab 83 1.05 Docetaxel 51 49 (96.1) (0.42–0.89) Docetaxel 38 35 (92.1) (0.70–1.56) 90 90 80 80 70 70 60 60 PFS, % PFS, % Median (95% CI) Median (95% CI) 4.2 mo (2.2–10.0 mo) 3.6 mo (2.1–4.5 mo) 50 50 2.4 mo (2.1–6.0 mo) 3.4 mo (2.1–7.5 mo) 40 40 30 30 20 20 10 10 0 0 0 6 12 18 24 30 36 42 48 54 60 66 72 78 0 6 12 18 24 30 36 42 48 54 60 66 72 78 Time, months Time, months No. at Risk No. at Risk Pembrolizumab 81 36 25 20 17 16 12 12 11 9 8 3 1 0 Pembrolizumab 83 27 15 11 7 5 5 2 2 2 1 1 0 0 Docetaxel 51 17 6 5 3 2 2 2 2 1 1 0 0 0 Docetaxel 38 14 7 4 4 3 3 1 0 0 0 0 0 0 C tTMB ≥175 mut/exome tTMB <175 mut/exome 50 50 21.1% 24.7% (9.6–37.3) 40 (15.8–35.5) 40 ORR, % (95% CI) ORR, % (95% CI) 16.9% (9.5–26.7) 30 9.8% 30 (3.3–21.4) 20 20 10 10 0 0 Pembrolizumab Docetaxel Pembrolizumab Docetaxel (n = 81) (n = 51) (n = 83) (n = 38) Figure 4. Outcomes in patients with tTMB 175 and 175 mut/exome. (A) OS, (B) PFS, and (C) ORR. CI, confidence interval; HR, hazard ratio; mut/exome, mutation per exome; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; tTMB, tissue tumor mutational burden. Several studies have found that higher levels of tTMB untreated advanced NSCLC with a PD-L1 TPS 1%.21 are associated with response among patients receiving The finding of an association with OS contrasts with pembrolizumab.19,20 However, there have been limited other studies evaluating tTMB as a biomarker for anti– data supporting this hypothesis derived from controlled PD-(L)1 therapies in patients with advanced NSCLC, studies. In the current exploratory analysis, tTMB 175 which have identified associations only with PFS and mut/exome was associated with improved clinical out- ORR.22-24 Because there was no strong association be- comes (OS, PFS, and ORR) for pembrolizumab versus tween tumor PD-L1 expression and tTMB, this finding is docetaxel. Our findings are consistent with those from an likely not due to increased tumor PD-L1 expression analysis of outcomes by tTMB in the KEYNOTE-042 among patients with higher tTMB. These findings study, which showed an association between tTMB suggest that tTMB may provide additional information 175 mut/exome and improved outcomes (OS, PFS, and regarding the clinical benefit of pembrolizumab mon- ORR) for pembrolizumab in patients with previously otherapy in patients with PD-L1–positive advanced 1730 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 NSCLC in the first-line and previously treated settings. CRediT Authorship Contribution The exploratory analysis of tTMB had several limita- Statement tions. tTMB ascertainment was low; therefore, only a Bilal Piperdi, Edward B. Garon, Byoung Chul Cho, small subset of the intent-to-treat population was Roy S. Herbst: Conception, design or planning of the included. In addition, because KEYNOTE-010 only study. included patients with PD-L1 TPS 1%, the study Dong-Wan Kim, Edward B. Garon, Isabelle Mon- cannot provide information on the potential predictive net, Radj Gervais, Jose L. Perez-Gracia, Byoung Chul value of tTMB in patients whose tumors do not Cho, Ji-Youn Han, Roy S. Herbst, Ayman Samkari, express PD-L1. Margarita Majem, Matthew A. Gubens, Paul Baas: The present analysis confirms the efficacy of pem- Acquisition of the data. brolizumab as second-line or later treatment for Bilal Piperdi, Byoung Chul Cho, Erin H. Jensen, Ji- advanced NSCLC. Notably, in the first-line setting, Youn Han, Roy S. Herbst, Wu-Chou Su, Julie Kobie, pembrolizumab has been reported to improve OS in Ayman Samkari: Analysis of the data. patients with NSCLC as monotherapy (in patients with Bilal Piperdi, Dong-Wan Kim, Edward B. Garon, PD-L1–expressing tumors)3,7 and when combined with Jose L. Perez-Gracia, Byoung Chul Cho, Erin H. Jensen, platinum-based chemotherapy (irrespective of tumor PD- Ji-Youn Han, Martin D. Forster, Michael Boyer, Roy S. L1 expression),4-6 suggesting first-line pembrolizumab Herbst, Julie Kobie, Ayman Samkari, Margarita options may provide greater benefit. In the real-world Majem, Paul Baas, Silvia Novello: Interpretation of the setting, approximately 50% of patients with advanced results. NSCLC do not receive second-line therapy because of Bilal Piperdi, Byoung Chul Cho, Ji-Youn Han, rapid clinical deterioration.25 Consequently, delaying Ayman Samkari: Drafting of the manuscript. pembrolizumab to second line may deprive patients of Roy S. Herbst, Edward B. Garon, Dong-Wan Kim, potential treatment benefit from pembrolizumab in the Byoung Chul Cho, Radj Gervais, Jose L. Perez- first-line setting. In the phase 3 KEYNOTE-024 study, the Gracia, Ji-Youn Han, Margarita Majem, Martin D. 5-year OS rate among patients with metastatic NSCLC Forster, Isabelle Monnet, Silvia Novello, Matthew with PD-L1 TPS 50% and without EGFR or ALK alter- A. Gubens, Michael Boyer, Wu-Chou Su, Ayman ations was 31.9% for those in the pembrolizumab group Samkari, Erin H. Jensen, Julie Kobie, Bilal Piperdi, versus 16.3% for those in the platinum-based chemo- Paul Baas: Review of the manuscript for important therapy group.17 Findings from the current study, more intellectual content, Decision to submit the manu- than any other, have revealed the predictive value of the script for publication, Access to study data and related PD-L1 IHC 22C3 pharmDx assay. analyses. In this updated analysis, no new safety signals were Dong-Wan Kim, Edward B. Garon, Perez-Gracia, Ji- identified for pembrolizumab with long-term follow-up. Youn Han, Michael Boyer, Matthew A. Gubens, Silvia AEs were manageable among patients who received Novello: Provision of study materials/patients. second-course pembrolizumab after disease progression, with no grade 4 or 5 treatment-related AEs reported during the second course. The updated safety data are Acknowledgments consistent with the long-term (5-y) safety profile Funding for this research was provided by Merck Sharp observed in the phase 1 KEYNOTE-001 study.13 Notably, & Dohme Corp., a subsidiary of Merck & Co., Inc., the rate of any treatment-related AEs, grade 3 to 5 AEs, Kenilworth, NJ, USA. Representatives of the funder and treatment-related AEs leading to discontinuation or participated in the study design; in the collection, anal- death was lower in patients who received pem- ysis, and interpretation of data; in the writing of the brolizumab versus those who received docetaxel. These report; and in the decision to submit the article for results support the long-term tolerability of pem- publication. Medical writing assistance was provided by brolizumab monotherapy. Christabel Wilson, MSc, of ICON plc (North Wales, PA, In conclusion, pembrolizumab continued to provide USA). This assistance was funded by Merck Sharp & long-term OS and PFS benefit than docetaxel in patients Dohme Corp., a subsidiary of Merck & Co., Inc., Kenil- with previously treated, PD-L1–expressing advanced worth, NJ, USA. NSCLC. Treatment benefit was observed in patients who received 35 cycles/2 years of pembrolizumab and in Supplementary Data those who received second-course pembrolizumab. Note: To access the supplementary material accompa- Exploratory analyses suggested an association between nying this article, visit the online version of the Journal of tTMB 175 mut/exome and pembrolizumab treatment Thoracic Oncology at www.jto.org and at https://doi. effect. org/10.1016/j.jtho.2021.05.001. October 2021 5-Year Outcomes in KEYNOTE-010 1731 Data Sharing Statement 5. Gandhi L, Rodriguez-Abreu D, Gadgeel S, et al. Pem- brolizumab plus chemotherapy in metastatic non-small- Merck Sharp & Dohme Corp., a subsidiary of Merck & cell lung cancer. N Engl J Med. 2018;378:2078–2092. Co., Inc., Kenilworth, New Jersey (MSD), is committed to 6. Paz-Ares L, Luft A, Vicente D, et al. Pembrolizumab plus providing qualified scientific researchers access to ano- chemotherapy for squamous non-small-cell lung cancer. nymized data and clinical study reports from the com- N Engl J Med. 2018;379:2040–2051. pany’s clinical trials for the purpose of conducting 7. Mok TSK, Wu YL, Kudaba I, et al. Pembrolizumab versus legitimate scientific research. MSD is also obligated to chemotherapy for previously untreated, PD-L1-expressing, protect the rights and privacy of trial participants and, as locally advanced or metastatic non-small-cell lung cancer such, has a procedure in place for evaluating and fulfilling (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. Lancet. 2019;393:1819–1830. requests for sharing company clinical trial data with 8. Herbst RS, Baas P, Perez-Gracia JL, et al. Use of archival qualified external scientific researchers. The MSD data- versus newly collected tumor samples for assessing PD- sharing website (available at: http://engagezone.msd. L1 expression and overall survival: an updated analysis com/ds_documentation.php) outlines the process and of KEYNOTE-010 trial. Ann Oncol. 2019;30:281–289. requirements for submitting a data request. Applications 9. Herbst RS, Garon EB, Kim DW, et al. Long-term outcomes will be promptly assessed for completeness and policy and retreatment among patients with previously compliance. Feasible requests will be reviewed by a treated, programmed death-ligand 1‒positive, advanced non‒small-cell lung cancer in the KEYNOTE-010 study. committee of MSD subject matter experts to evaluate the J Clin Oncol. 2020;38:1580–1590. scientific validity of the request and the qualifications of 10. KEYTRUDA® (pembrolizumab) full prescribing informa- the requestors. In line with data privacy legislation, sub- tion. Whitehouse Station, NJ: Merck Sharp & Dohme mitters of approved requests must enter into a standard Corp; 2020. data-sharing agreement with MSD before data access is 11. Cristescu R, Mogg R, Ayers M, et al. Pan-tumor genomic granted. Data will be made available for request after biomarkers for PD-1 checkpoint blockade-based immu- product approval in the US and EU or after product notherapy. Science. 2018;362:eaar3593. 12. Panda A, Betigeri A, Subramanian K, et al. Identifying a development is discontinued. There are circumstances clinically applicable mutational burden threshold as a po- that may prevent MSD from sharing requested data, tential biomarker of response to immune checkpoint therapy including country- or region-specific regulations. If the in solid tumors. JCO Precis Oncol. 2017;2017:PO.17.00146. request is declined, it will be communicated to the 13. Garon EB, Hellmann MD, Rizvi NA, et al. Five-year investigator. Access to genetic or exploratory biomarker overall survival for patients with advanced non-small- data requires a detailed, hypothesis-driven statistical cell lung cancer treated with pembrolizumab: results analysis plan that is collaboratively developed by the from the phase I KEYNOTE-001 study. J Clin Oncol. requestor and MSD subject matter experts; after approval 2019;37:2518–2527. 14. Borghaei H, Gettinger S, Vokes EE, et al. Five-year out- of the statistical analysis plan and execution of a data comes from the randomized, phase III trials CheckMate sharing agreement, MSD will either perform the proposed 017 and 057: nivolumab versus docetaxel in previously analyses and share the results with the requestor or will treated non-small-cell lung cancer. J Clin Oncol. construct biomarker covariates and add them to a file 2021;39:723–733. with clinical data that is uploaded to an analysis portal so 15. Mazieres J, Rittmeyer A, Gadgeel S, et al. Atezolizumab that the requestor can perform the proposed analyses. versus docetaxel in pretreated patients with NSCLC: final results from the randomized phase 2 POPLAR and phase 3 OAK clinical trials. J Thorac Oncol. 2021;16:140–150. References 16. Howlader N, Noone AM, Krapcho M, et al. SEER cancer 1. Peters S, Kerr KM, Stahel R. PD-1 blockade in advanced statistics review, 1975–2017: Bethesda, MD: National NSCLC: a focus on pembrolizumab. Cancer Treat Rev. Cancer Institute. https://seer.cancer.gov/csr/1975_2 2018;62:39–49. 017/. Accessed July 18, 2021. 2. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus 17. Reck M, Rodríguez-Abreu D, Robinson AG, et al. Five-year docetaxel for previously treated, PD-L1-positive, advanced outcomes with pembrolizumab versus chemotherapy for non–small-cell lung cancer (KEYNOTE-010): a randomised metastatic non-small-cell lung cancer with PD-L1 tumor controlled trial. Lancet. 2016;387:1540–1550. proportion score 50%. J Clin Oncol. 2021:JCO2100174. 3. Reck M, Rodriguez-Abreu D, Robinson AG, et al. Pem- 18. Waterhouse DM, Garon EB, Chandler J, et al. Continuous brolizumab versus chemotherapy for PD-L1-positive versus 1-year fixed-duration nivolumab in previously non–small-cell lung cancer. N Engl J Med. treated advanced non-small-cell lung cancer: CheckMate 2016;375:1823–1833. 153. J Clin Oncol. 2020;38:3863–3873. 4. Langer CJ, Gadgeel SM, Borghaei H, et al. Carboplatin 19. Marabelle A, Fakih M, Lopez J, et al. Association of tumour and pemetrexed with or without pembrolizumab for mutational burden with outcomes in patients with advanced advanced, non-squamous non-small-cell lung cancer: a solid tumours treated with pembrolizumab: prospective randomised, phase 2 cohort of the open-label KEYNOTE- biomarker analysis of the multicohort, open-label, phase 2 021 study. Lancet Oncol. 2016;17:1497–1508. KEYNOTE-158 study. Lancet Oncol. 2020;21:1353–1365. 1732 Herbst et al Journal of Thoracic Oncology Vol. 16 No. 10 20. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immu- 23. Carbone DP, Reck M, Paz-Ares L, et al. First-line nivolu- nology. Mutational landscape determines sensitivity to mab in stage IV or recurrent non-small-cell lung cancer. PD-1 blockade in non-small cell lung cancer. Science. N Engl J Med. 2017;376:2415–2426. 2015;348:124–128. 24. Hellmann MD, Ciuleanu TE, Pluzanski A, et al. 21. Herbst RS, Lopes G, Kowalski DM, et al. Association be- Nivolumab plus ipilimumab in lung cancer with a high tween tissue TMB (tTMB) and clinical outcomes with tumor mutational burden. N Engl J Med. 2018;378:2093– pembrolizumab monotherapy (pembro) in PD-L1-positive 2104. advanced NSCLC in the KEYNOTE-010 and -042 trials. 25. Davies J, Patel M, Gridelli C, de Marinis F, Waterkamp D, Ann Oncol. 2019;30(suppl 5):v851–v934. McCusker ME. Real-world treatment patterns for pa- 22. Hu-Lieskovan S, Lisberg A, Zaretsky JM, et al. Tumor tients receiving second-line and third-line treatment for characteristics associated with benefit from pem- advanced non-small cell lung cancer: a systematic re- brolizumab in advanced non-small cell lung cancer. Clin view of recently published studies. PLoS One. 2017;12: Cancer Res. 2019;25:5061–5068. e0175679.