Tissue tumour mutational burden (tTMB)as a marker for pembrolizumab efficacy in patients with non-small cell lung cancer (NSCLC)remained unresolved according to findings from two exploratory analyses of data from five trials reported at the ESMO Congress 2019 in Barcelona, Spain.
Roy S. Herbst, Medical Oncology, Yale University School of Medicine in New Haven, US and colleagues retrospectively investigated the association between tTMB and the clinical benefit with pembrolizumab monotherapy observed in patients with previously treated or untreated advanced NSCLC who were PD-L1-positive with a tumour proportion score (TPS) ≥1% participating in the KEYNOTE-010 (NCT01905657) or KEYNOTE-042 (NCT02220894) studies, respectively. Their exploratory analysis was performed in a subset of patients having evaluable tTMB, which was determined by whole exome sequencing of tumour and matched normal DNA. tTMB was evaluable for 253 (24%) patients in KEYNOTE-010 and 793 (62%) patients in KEYNOTE-042. The tTMB evaluable subset and total study population were similar in both trials regarding baseline characteristics and clinical outcomes.
The association of tTMB with outcomes in each treatment arm in each study, was evaluated using Cox proportional hazards models for overall survival (OS) and progression-free survival (PFS) whereas logistic regression was used for the objective response rate (ORR). Clinical utility of tTMB was assessed using a prespecified cutpoint of 175 mutations per exome.
OS, PFS and response favoured pembrolizumab monotherapy over chemotherapy in patients with tTMB ≥175 mutations per exome in the KEYNOTE-010 and KEYNOTE-042 studies
The investigators found that tTMB did not associate with TPS in either the pembrolizumab or chemotherapy arms (r < 0.18).
In both KEYNOTE trials, tTMB was associated with OS, PFS and ORR for the pembrolizumab arms in KEYNOTE-010 (1-sided p = 0.006, p = 0.001, and p = 0.009, respectively) and in KEYNOTE-042 (all 1-sided p < 0.001); however, tTMB was not associated with outcomes for chemotherapy in either study.
Improvements in OS, PFS and ORR were generally observed for patients with high tTMB ≥175 mutations per exome receiving pembrolizumab.
In KEYNOTE-010 patients with tTMB ≥175 mutations per exome, median OS was 14.1 months in 81 patients on pembrolizumab compared to 7.6 months in 51 patients on chemotherapy (hazard ratio [HR] 0.56; 95% confidence interval [CI], 0.38-0.83). In the subset of patients with tTMB <175 mutations per exome median OS was 9.3 months in 83 pembrolizumab-treated patients compared to 7.2 months in 38 patients receiving chemotherapy (HR 0.85; 95% CI, 0.56-1.30).
In KEYNOTE-010 patients with tTMB ≥175 mutations per exome, median PFS was 4.2 months versus 2.4 months in patients on pembrolizumab versus chemotherapy (HR 0.59; 95% CI, 0.40-0.87) and in patients with tTMB < 175 mutations per exome median PFS was 3.7 months versus 3.4 months in the respective treatment arms (HR 1.09; 95% CI, 0.72-1.63).
The ORR in patients with tTMB ≥175 mutations per exome was 23.5% versus 9.8%, and in patients with tTMB <175 mutations per exome the ORR was 16.9% versus 21.1% with pembrolizumab and chemotherapy, respectively.
In KEYNOTE-042, in the tTMB ≥175 mutations per exome group 189 patients received pembrolizumab and 165 patients received chemotherapy and in the tTMB <175 mutations per exome group 234 versus 214 patients received pembrolizumab versus chemotherapy respectively.
The tTMB ≥175 mutations per exome group demonstrated median OS of 21.9 versus 11.6 months with pembrolizumab versus chemotherapy (HR 0.62; 95% CI 0.48-0.80), median PFS of 6.3 versus 6.5 months (HR 0.75; 95% CI, 0.59-0.95), and ORR of 34.4% versus 30.9%, respectively.
In the tTMB <175 mutations per exome group, median OS was 12.0 versus 12.3 months with pembrolizumab versus chemotherapy (HR 1.09; 95% CI 0.88-1.36), median PFS was 4.1 versus 6.3 months (HR 1.27; 95% CI, 1.04-1.55), and ORR was 18.8% versus 22.4%, respectively.
Analysis of the KEYNOTE-021, -189, and -407 trials in metastatic squamous and non-squamous NSCLC shows no significant association between tTMB and the efficacy of pembrolizumab plus chemotherapy
Luis Paz-Ares, Medical Oncology, University Hospital 12 De Octubre in Madrid, Spain, also used data from KEYNOTE trials of first-line pembrolizumab plus chemotherapy versus chemotherapy to evaluate the relationship between tTMB and efficacy.
Cohorts C and G of the KEYNOTE-021 trial (NCT02039674) and the KEYNOTE-189 trial (NCT02578680) comprised patients with non-squamous NSCLC and the KEYNOTE-407 trial (NCT02775435) enrolled patients with squamous NSCLC.
Using samples from patients in these trials, tTMB was determined by whole-exome sequencing of tumour and matched normal DNA. For each study, the association of tTMB with outcomes for pembrolizumab plus chemotherapy and for chemotherapy was assessed in patients with evaluable tTMB according to the methods in the above analysis.
The primary endpoints were OS and PFS in the KEYNOTE-189 and -407 studies and ORR in KEYNOTE-021.
No statistically significant association was determined between tTMB and efficacy of pembrolizumab plus chemotherapy
tTMB was evaluable in 48% of patients in KEYNOTE-021 C and G, 48% in KEYNOTE-189, and 56% of patients in KEYNOTE-407. Baseline characteristics and efficacy were similar in the tTMB-evaluable and total populations.
This exploratory analysis demonstrated that tTMB was not significantly associated with efficacy of pembrolizumab plus chemotherapy or chemotherapy alone (across all studies; 1-sided p ≤ 0.072 for ORR, p ≤ 0.052 for PFS, and p ≤ 0.081 for OS for pembrolizumab plus chemotherapy and 2-sided p ≤ 0.086 for ORR, p ≤ 0.055 for PFS, and p ≤ 0.475 for OS for chemotherapy).
Clinical benefit was greater with pembrolizumab/chemotherapy versus chemotherapy
In the KEYNOTE-189 and 407 trials, pembrolizumab plus chemotherapy improved OS and PFS in patients with tTMB ≥175 mutations per exome and also with <175 mutations per exome.
In the tTMB ≥175 mutations per exome of the KEYNOTE-189 and KEYNOTE-407 trials, 100 and 73 patients received pembrolizumab plus chemotherapy whereas 34 and 73 patients received chemotherapy, respectively. Median OS in the respective trials was improved with pembrolizumab plus chemotherapy over chemotherapy (HR 0.64; 95% CI, 0.38-1.07 in KEYNOTE-189 and HR 0.74; 95% CI, 0.50-1.08 in KEYNOTE-407), as was PFS (HR 0.32; 95% CI, 0.21-0.51 and HR 0.57; 95% CI, 0.41-0.81, respectively).
Patients with tTMB <175 mutations per exome in KEYNOTE-189 and -407 trials also showed improved benefit with pembrolizumab plus chemotherapy (N = 107 and 70, respectively) versus chemotherapy (N = 52 and 80, respectively) for OS (HR 0.64; 95% CI, 42-0.97 in KEYNOTE-189 and HR 0.86; 95% CI, 0.57-1.28 in KEYNOTE-407) as well as PFS (HR 0.51; 95% CI, 0.35-0.74 and HR 0.68; 95% CI, 0.48-0.96, respectively).
In cohort G of KEYNOTE-021, ORR was higher with pembrolizumab plus chemotherapy versus chemotherapy in the 31 patients with tTMB ≥175 mutations per exome (71.4% vs 30.0%) and in the 39 patients with tTMB <175 mutations per exome (60.8% vs 43.8%).
Conclusions
The authors concluded that the exploratory analysis of data from the KEYNOTE-010 and -042 studies demonstrated associations between higher tTMB levels and improved clinical outcome with pembrolizumab monotherapy in patients with PD-L1-positive NSCLC. While the results suggest that tTMB may provide additional information regarding the clinical benefit of pembrolizumab monotherapy in these patients’ tumours, the analyses and effect estimates were limited to subsets of patients with available tTMB data in the randomised groups.
In the exploratory analysis done by Professor Paz-Ares and colleagues, tTMB was not significantly associated with the efficacy of first-line pembrolizumab plus platinum-based chemotherapy or with chemotherapy alone in metastatic NSCLC, regardless of histology. Pembrolizumab plus chemotherapy showed survival benefit in the tTMB-high and -low subgroups for both squamous and non-squamous NSCLC.
Professor Paz-Ares declared that some content had already been presented at 2019 World Conference on Lung Cancer. However, the abstract presented at ESMO 2019 Congress include never presented tTMB data from KEYNOTE-407, which helps to understand the role of tTMB in patients receiving pembrolizumab plus chemotherapy for NSCLC into full context.
Disclosure
The KEYNOTE studies were funded by Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ USA.
References
LBA79 – Herbst RS, Lopes G, Kowalski DM, et al. Association between tissue TMB (tTMB) and clinical outcomes with pembrolizumab monotherapy (pembro) in PD-L1-positive advanced NSCLC in the KEYNOTE-010 and -042 trials.
LBA80 – Paz-Ares L, Langer CJ, Novello S, et al. Pembrolizumab (pembro) plus platinum-based chemotherapy (chemo) for metastatic NSCLC: tissue TMB (tTMB) and outcomes in KEYNOTE-021, 189, and 407.