During 2018, exciting results from several pivotal immunotherapy trials led to impactful changes in the standard of care for all patients with locally advanced and metastatic NSCLC. Fueled by the momentum of KEYNOTE-024 (N Engl J Med 2016;375:1823-1833), which led to the FDA approval of pembrolizumab for NSCLC patients with PD-L1 TPS ≥50 percent, trials reported in 2018 with checkpoint inhibitors alone and in combination with chemotherapy have afforded an immunotherapy option for all metastatic lung cancer patients in the first-line setting.
KEYNOTE-042 was a phase III trial that compared pembrolizumab to platinum-based chemotherapy in patients with PD-L1 expression ≥1 percent (J Clin Oncol 2015;33:TPS8105-TPS8105, J Clin Oncol 2018;36:LBA4-LBA4). Pembrolizumab significantly improved overall survival (OS) in the PD-L1 TPS ≥50 percent, ≥20 percent, and ≥1 percent subgroups. KEYNOTE-042 met its primary endpoint of OS in all PD-L1 subgroups. However, the greatest benefit was in patients with higher PD-L1 expression levels (see Table 1), consistent with earlier studies with pembrolizumab. Pembrolizumab provided a modest progression-free survival (PFS) in the PD-L1 TPS ≥50 percent subgroup (HR 0.81; 0.67-0.99); however, there was no PFS advantage in the ≥20 percent and ≥1 percent PD-L1 subgroups.
Of note, PFS was a co-primary endpoint in KEYNOTE-024 trial and was significantly improved with pembrolizumab compared with chemotherapy (10.3 months vs. 6.0 months, HR 0.50; 0.37-0.68) to a much greater extent than observed in KEYNOTE-042; however, the OS benefits were similar in those with PD-L1 TPS ≥50 percent. Compared with KEYNOTE-024, KEYNOTE-042 enrolled higher percentages of never smokers (22% vs. 3%) and squamous patients (38% vs. 19%), and fewer patients who received prior therapy (i.e., neoadjuvant, adjuvant, or radiotherapy) (15% vs. 30%). Regardless of these differences in patient populations evaluated, KEYNOTE-042 supports the use of pembrolizumab monotherapy in the first-line setting for both non-squamous and squamous NSCLC patients, especially for patients who are unwilling or unable to receive chemotherapy.
An analogous phase III trial evaluating the PD-1 inhibitor nivolumab versus platinum-based chemotherapy, CheckMate 026, yielded a curiously different result—PFS (among patients with a PD-L1 expression level of ≥5%) was equivalent between the arms, with no benefit in terms of secondary endpoint OS nor benefit for higher levels of PD-L1 expression.
However, the retrospective analysis of CheckMate 026 demonstrated tumor mutational burden (TMB) to be an independent predictor of response and PFS to PD-1 checkpoint inhibition (N Engl J Med 2017;376:2415-2426). Likewise, TMB has been associated with increased efficacy to dual immunotherapy agents: anti-PD-1 nivolumab and anti-CTLA-4 ipilimumab (Cancer Cell 2018;33:853-861.e854). CheckMate 568 determined the cutoff of TMB ≥10 mut/Mb was associated with an increased likelihood of response to first-line nivolumab and ipilimumab (Cancer Res 2018;78:CT078-CT078).
CheckMate 227 evaluated nivolumab plus ipilimumab, nivolumab plus chemotherapy, or chemotherapy alone in NSCLC patients stratified by PD-L1 status (N Engl J Med 2018;378:2093-2104). The co-primary endpoints were PFS with nivolumab plus ipilimumab versus chemotherapy in a TMB-selected population and OS with nivolumab plus ipilimumab versus chemotherapy in a PD-L1 selected population. The results demonstrated first-line treatment with nivolumab plus ipilimumab significantly extended PFS compared to chemotherapy in patients with TMB ≥10 mut/Mb and PD-L1 TPS < 1 percent with a trend towards improvement in OS.
Based on a request by the FDA, an exploratory OS analysis was performed that included both the TMB ≥10 mut/Mb and < 10 mut/Mb subgroups. Disappointingly, the updated OS data did not support the use of TMB as a predictive biomarker as a survival benefit was observed in patients receiving nivolumab plus ipilimumb, irrespective of TMB status. Subsequently, the manufacturer withdrew its supplemental biologics license application for nivolumab plus ipilimumab in the frontline treatment of NSCLC patients with TMB ≥10 mut/Mb.
The phase III MYSTIC trial randomized stage IV NSCLC patients (N=1,118), unselected for PD-L1 status, to durvalumab monotherapy, durvalumab plus tremelimumab, or platinum-based chemotherapy (ESMO Immuno-Oncology Congress 2018, Abstract LBA6). The co-primary endpoints were PFS for durvalumab plus tremelimumab versus chemotherapy; OS in durvalumab alone or in combination with tremelimumab versus chemotherapy. Only patients with PD-L1 TC ≥25 percent were included in the primary survival analyses (N=488). Patients treated with durvalumab monotherapy did have a numeric improvement in OS compared to chemotherapy (16.3 months vs. 12.9 months); however, the results were not statistically significant. MYSTIC did not meet its other primary endpoints of improving OS and PFS with durvalumab plus tremelimumab in the PD-L1 TC ≥25 percent population.
An interesting exploratory analysis examined TMB in both blood and tissue. In contrast to CheckMate 227, a blood TMB (bTMB) ≥16 mut/Mb was associated with an improved OS with durvalumab and tremelimumab compared to chemotherapy, suggesting bTMB may have predictive value with this immunotherapy combination.
Extending the role of pembrolizumab in NSCLC, the results of three pivotal first-line trials, KEYNOTE-189, KEYNOTE-407, and IMpower150, led to the FDA approval of immunotherapy plus chemotherapy for newly diagnosed metastatic NSCLC patients.
KEYNOTE-189 randomized non-squamous NSCLC patients 2:1 to pembrolizumab plus carboplatin and pemetrexed (CPP) or carboplatin and pemetrexed (CP) and excluded individuals with EGFR or ALK alterations (N Engl J Med 2018;379:e18). The CPP triplet regimen demonstrated a statistically significant OS (HR 0.49; 0.38-0.64) as well as PFS improvement (HR 0.52; 0.43-0.64) in the intent-to-treat population. Although patients benefited across all PD-L1 subgroups, a more pronounced benefit was in those with higher PD-L1 expression. The KEYNOTE-189 data supported earlier data from the smaller randomized phase II KEYNOTE-021G study (Ann Oncol 2017;28(Suppl_5): mdx440.052, J Thorac Oncol 2019;14:124-129).
Complementary to KEYNOTE-189, KEYNOTE-407 randomized squamous patients to pembrolizumab or placebo with carboplatin and either paclitaxel or nab-paclitaxel (N Engl J Med 2018;379:2040-2051). The pembrolizumab plus chemotherapy combination improved OS (HR 0.64; 0.49-0.85) with a notable 44 percent reduction in the risk of disease progression (HR 0.56; 0.45-0.70). Similar to KEYNOTE-189, higher PD-L1 expression appeared to be a predictor for clinical benefit, as those with PD-L1 TPS ≥50 percent experienced the greatest PFS advantage. PD-L1 expression does not, however, appear to be as robust of an indicator of OS for squamous patients.
IMpower150 evaluated the efficacy of atezolizumab, bevacizumab, carboplatin, and paclitaxel plus bevacizumab (ABCP) verses atezolizumab, carboplatin, and paclitaxel (ACP) or bevacizumab, carboplatin, and paclitaxel (BCP) in metastatic non-squamous NSCLC (N Engl J Med 2018;378:2288-2301). The quadruplet regimen afforded an improved OS benefit compared to the BCP alone (HR 0.78; 0.64-0.96) and was observed across all PD-L1 TPS subgroups; updated PFS data also favored ABCP compared to BCP (HR 0.59; 0.50-0.70). A distinctive aspect of IMpower150 was the inclusion of patients with EGFR or ALK gene alterations, since patients with driver oncogenes were excluded from most other first-line trials. Subgroup analysis demonstrated favorable OS with ABCP in patients with EGFR and ALK alterations, as well as in patients with liver metastases, perhaps due to hypothesized synergy between immunotherapy and anti-angiogenic agents. However, this subgroup was not included in the regulatory label (Semin Cancer Biol 2018;52(Pt2):117-124).
Additional phase III trials further explored the clinical benefit of first-line atezolizumab for advanced NSCLC patients with a variety of chemotherapy scaffolds and included IMpower 130, IMpower 132, and IMpower131.
IMpower130 randomized non-squamous patients to atezolizumab, carboplatin, and nab-paclitaxel or carboplatin and nab-paclitaxel, followed by maintenance atezolizumab or pemetrexed/best supportive care (Ann Oncol 2018;29(Suppl 8)). Patients with EGFR and ALK alterations were also included. The trial reported improvement in OS, as well as PFS, seen across all PD-L1 TPS subgroups receiving the atezolizumab plus chemotherapy.
IMpower132 randomized 1:1 non-squamous stage IV NSCLC patients without EGFR and ALK alterations to atezolizumab plus carboplatin or cisplatin and pemetrexed followed by atezolizumab plus pemetrexed or chemotherapy alone with maintenance pemetrexed (J Thorac Oncol 2018;13(10):S332-S333). The atezolizumab-containing arm improved PFS (HR 0.60; 0.49-0.72) and ORR over chemotherapy alone (47% vs. 32%), with the interim OS analysis showing a trend toward improved survival (HR 0.81; 0.64-1.03).
For squamous histology, IMpower131 atezolizumab plus carboplatin and nab-paclitaxel provided PFS benefit compared to chemotherapy alone across all PD-L1 TPS subgroups (HR 0.74; 0.62-0.87); with a greater PFS advantage in those with high PD-L1 expression (HR 0.44; 0.27- 0.71) (J Clin Oncol 2018;36:LBA9000-LBA9000). The second interim OS analyses did not show a significant difference in OS (HR 0.92; 0.76-1.12) (Ann Oncol 2018;29(suppl_8)).
While significant improvements in clinical efficacy have been achieved with immunotherapy-based treatments in the first-line setting, changes in the standard therapeutic landscape have raised many new questions to answer in 2019. Such questions include the following: 1) When should we use immunotherapy monotherapy versus a combination regimen? 2) Should TMB status be realistically incorporated into decision-making prior to initiating first-line treatment? 3) How much maintenance therapy is enough or too much? 4) If both chemotherapy and immunotherapy are administered as first-line therapy, what has become the best second-line option? Do we continue PD-1 therapy in the second-line? Do we switch to a PD-L1 inhibitor? Lastly, how can the successes realized during 2018 be used to shape future clinical trials to identify true benefit from the treatments we select?
MELISSA JOHNSON, MD, is Medical Oncologist and LISA J. ZIMMERMAN, PHD, MBA, is Lung Program Manager, both at Sarah Cannon in Nashville, Tenn.
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.
More on ONCOLOGY-TIMES.com...