A number of lung cancer specialists were anxious to comment on the positive findings of KEYNOTE-024 and were equally perplexed about the negative results of CheckMate-026. All agreed that the overall survival benefit makes pembrolizumab (Keytruda) a game-changer for the first-line treatment of advanced non–small cell lung cancer (NSCLC).
The 45% overall response rate with pembrolizumab is the best ever reported in the first-line setting—and with monotherapy. That’s unheard of!— Jean-Charles Soria, MD, PhD
Tweet this quote
Invited discussant of KEYNOTE-024, Jean-Charles Soria, MD, PhD, of the Institute Gustave Roussy and Paris University XI, where he is Professor of Medicine and Medical Oncology, commented: “The 45% overall response rate with pembrolizumab is the best ever reported in the first-line setting—and with monotherapy. That’s unheard of!”
Progression-free survival was improved by 50% and overall survival, by 40%. All subgroups benefited, except for women and never smokers (patients known to have lower mutational loads). Response rates and duration of response were also superior with pembrolizumab, Dr. Soria noted. Based on these findings, the standard-of-care platinum doublet “is no longer dominant,” he declared. “There’s a new player, probably a new standard of care.” However, he cautioned that this new standard of care only applies to 10% of front-line NSCLC. Indeed many restrictive inclusion criteria were applied: tissue availability, PD-L1 [programmed cell death ligand 1] staining > 50% in the tissue, no EGFR or ALK alterations, no untreated brain metastasis, no autoimmune disease, and no steroids for autoimmune diseases.
Additional Commentary
Roy Herbst, MD, PhD
Roy Herbst, MD, PhD, Ensign Professor of Medicine (Medical Oncology), Professor of Pharmacology, and Chief of Medical Oncology at Yale Cancer Center and Smilow Cancer Hospital, agreed. “Pembrolizumab certainly seems to work for patients with PD-L1 expression ≥ 50%. It’s historic that for this 30% of the NSCLC population—without oncogenic mutations and whether they have squamous or nonsquamous histology—treatment will be with pembrolizumab as a single agent,” he told The ASCO Post.
He added that the study confirms the utility of testing for PD-L1 expression in these patients. “A couple of years ago at the ASCO Annual Meeting, I gave a talk saying this biomarker was not ready for prime time unless it’s used in the front-line setting. Today, we have seen that biomarkers are important, and PD-L1 has been refined to the point where it is useful in lung cancer. It’s telling us that patients who are [strongly] biomarker-positive should get pembrolizumab.”
Therefore, PD-L1 expression will be used to select patients for first-line treatment, as ≥ 50% expression is still mandated for using pembrolizumab. This makes approximately one-third of patients with advanced NSCLC eligible—although as Dr. Soria pointed out, only 500 of 1,934 screened patients in KEYNOTE-024 were randomized. That is 25%, and not 33%, of patients. Indeed, in the clinic, many patients have no tissue available for extra-tissue staining. “We need to expand this pool of patients,” Dr. Soria added.
This has important implications in the clinic. “Once pembrolizumab is approved, PD-L1 assessment must be part of the standard diagnostic workup, and I argue that we do this on a reflex basis. Patients cannot wait for these results,” Dr. Langer commented.
Why Was CheckMate-026 Negative?
The other conversation at the ESMO Congress tried to dissect the surprisingly negative results for first-line nivolumab (Opdivo). The lack of benefit is “somewhat perplexing,” Dr. Herbst commented. “I’m still puzzled,” Corey Langer, MD, Director of the Thoracic Oncology Program, Abramson Cancer Center, University of Pennsylvania Health System, concurred. “I cannot really figure out why this trial was completely negative.”
“We have no clear explanation for the differences in readout between this trial and KEYNOTE-024. The results of CheckMate-026 are inconsistent with [those of] earlier phase I/II trials of nivolumab,” according to the study’s invited discussant, Naiyer Rizvi, MD, Director of Thoracic Oncology and Director of Immunotherapeutics at New York-Presbyterian/Columbia University Medical Center.
We have no clear explanation for the differences in readout between this trial and KEYNOTE-024. The results of CheckMate-026 are inconsistent with [those of] earlier phase I/II trials of nivolumab.— Naiyer Rizvi, MD
Tweet this quote
The working hypothesis is that the two study populations were different. Most important, the use of a PD-L1 cutoff of ≥ 5% expression for the primary analysis (and ≥ 1% for enrollment) may have diluted the population most likely to benefit, Dr. Rizvi noted. Also, certain imbalances between the arms (eg, more women in the chemotherapy arm and more hepatic metastases in the nivolumab arm) may have been factors. Dr. Herbst also noted the control arm of the study performed well, which hints the study populations were, indeed, “not the same.”
Despite the divergence in the study’s findings, Dr. Herbst said he does not believe the two anti–PD-1 (programmed cell death protein 1) agents are “that different…. Certainly, nivolumab has shown activity in randomized trials. I would bet it’s the biomarker,” he explained, predicting the study will be repeated in patients with higher PD-L1 expression.
Clinical trials are underway in populations with various degrees of PD-L1 expression and in combination with ipilimumab and with chemotherapy; novel checkpoint inhibitors are also in trials. According to these experts, these ongoing and future research efforts will result in treatments showing an even greater benefit than was seen with single-agent pembrolizumab. ■
Disclosure: Dr. Soria is a consultant for AstraZeneca, Astex, Clovis, GlaxoSmithKline, Gammamabs, Lilly, MSD, Mission Therapeutics, Merus, Pfizer, Pharmamar Pierre Fabre, Roche-Genentech, Sanofi, Servier, Symphogen, and Takeda. Dr. Rizvi reported no potential conflicts of interest.
