The optimal approach to non-small cell lung cancer (NSCLC) patients with programmed death ligand-1 (PD-L1) expression of more than 50 percent should include the checkpoint inhibitor pembrolizumab. But should chemotherapy be added in as well?
"Pembrolizumab monotherapy and pembrolizumab plus histology-specific chemotherapy are both reasonable options for the unselected NSCLC patient," said Aaron Lisberg, MD, Hematologist and Medical Oncologist the University of California, Los Angeles, at the 2022 Great Debates & Updates in Lung Cancers. "There are no high-level randomized trial data to allow a head-to-head comparison of these approaches in NSCLC patients. In the absence of high-level data, we are left with cross-trial comparisons to inform this decision. In a patient agnostic sense, that is, all we know is advanced NSCLC and PD-L1 more than 50 percent, what elements should we factor into our decision?"
Pembrolizumab Plus Chemotherapy
Lisberg leveraged efficacy and safety data from three recent trials to support his contention that chemotherapy plus pembrolizumab was the more appropriate choice. Those trials were the randomized, open-label, Phase III KEYNOTE-024 trial of pembrolizumab versus chemotherapy, and subset analyses of patients with more than 50 percent PD-L1 expression in both the Phase III KEYNOTE-189 trial (non-squamous patients only) who received pembrolizumab plus platinum doublet chemotherapy versus chemotherapy, and the randomized, placebo-controlled KEYNOTE-407 trial of pembrolizumab plus platinum doublet chemotherapy versus chemotherapy.
In terms of efficacy, Lisberg compared overall response rate (ORR) data. In KEYNOTE-024, ORR was 45 percent; KEYNOTE-189, 62.1 percent; and KEYNOTE-407, 60.3 percent. "Cross-trial comparison suggests about 15-20 percent higher response rate with addition of chemotherapy," he said.
He noted that overall survival (OS) outcomes after about 1 year from the three trials are very similar, but "OS outcomes in the first year appear to be improved when chemotherapy is added to pembrolizumab, with 70 percent of patients still alive. It appears patients are more likely to have a high chance of OS with pembrolizumab plus chemotherapy versus pembrolizumab alone."
Progression-free survival (PFS) data after about 1 year is very similar to OS data in the trials. "PFS data again show longer PFS for patients on two agents. The logical conclusion is that some patients are doing better in the long-term due to chemotherapy plus immunotherapy," said Lisberg.
As for safety, he compared treatment-related adverse events (TRAEs) of Grade 3 or higher. In KEYNOTE-024, these were 31.2 percent; KEYNOTE-189, 67.2 percent; and KEYNOTE-407, 68.2 percent. "Cross-trial comparison suggests a 35 percent higher rate of Grade 3 or higher adverse events with chemotherapy, but comparing TRAEs to all-cause adverse events confounds the results," Lisberg said.
Pembrolizumab alone is less toxic, which is what would be expected. "In KEYNOTE-024, pembrolizumab alone was very well-tolerated. In KEYNOTE-189 the combination was the more toxic regimen than pembrolizumab alone. KEYNOTE-407 showed similar toxicity results as KEYNOTE-189, and toxicity led one-quarter of patients to discontinue therapy," he noted.
In conclusion, Lisberg asked: "Which approach is more likely to benefit the patient? ORR appears to be higher with the addition of chemotherapy. For less than 1 year, the addition of chemotherapy appears to improve both PFS and OS. For more than 1 year, there is no clear PFS or OS benefit to the addition of chemotherapy."
Then he asked: "Which approach is more likely to harm the patient? We see much higher rates of therapy discontinuation and death with the addition of chemotherapy. Clinicians have to make a value judgment to reveal the 'best' approach. Development of a biomarker would help us make that decision."
Pembrolizumab Alone
In adopting the opposite side of the debate, Edward B. Garon, MD, MS, Thoracic Medical Oncologist at University of California, Los Angeles, stated, "It's very clear some of these patients have high PD-L1 expression. This is a unique population that can experience clinical benefit in ORR that also translates into a PFS and OS advantage."
In comparisons of chemotherapy to chemotherapy plus immunotherapy versus immunotherapy alone, most researchers agree the results look similar. "I would argue that CTLA-4 inhibitor data look particularly good as monotherapy," said Garon.
He noted that the non-squamous NSCLC population in the KEYNOTE-189 "is the great majority of what we see in the clinic. There was a clear improvement in adding chemotherapy to pembrolizumab out to 48 months versus placebo."
In contrast, in the Phase I KEYNOTE-001 trial of pembrolizumab alone in patients with advanced NSCLC, "the OS long-term data at 18 months show nearly 80 percent of patients alive, with more than 40 percent previously treated patients still alive," he said.
Garon concluded: "As a biomarker, PD-L1 of more than 50 percent is good. This population has good long-term OS. We have no data to indicate chemotherapy increases durable responders, and chemotherapy decreases tolerability. Patients who have high PD-L1 expression should receive monotherapy with a PD-L1 inhibitor."
Mark L. Fuerst is a contributing writer.