Last year, the U.S. Food and Drug Administration (FDA) approved two anti-PD-1 checkpoint inhibitors, a type of immunotherapy, for treatment of non-small cell lung cancer (NSCLC) in patients whose cancer has progressed after first-line treatment with chemotherapy. Now, the manufacturers of both drugs, pembrolizumab (made by Merck) and nivolumab (made by Bristol-Myers Squibb; BMS) are intent on expanding the indications for use of their drugs. To this end, they have conducted clinical trials testing each as a first-line treatment (i.e., in previously untreated patients), comparing them to standard chemotherapy.
With results from both companies now made public, it is clear that pembrolizumab (brand name Keytruda) has “won” this phase of the checkpoint drugs war. KEYNOTE-024, the clinical trial that tested pembrolizumab, met its endpoint, improving progression-free survival (PFS) by 4.3 months (10.3 months versus 6.0 months in patients receiving chemotherapy), and improving overall survival as well. Checkmate-026, a similar trial with nivolumab (brand name Opdivo), has not achieved its primary endpoint of improving PFS over chemotherapy.
What is perhaps not well known is that the patients selected for participation in these trials were quite different from each other. The FDA previously approved pembrolizumab only in patients whose tumors showed presence of the PD-L1 protein in more than 1% of their cells, while nivolumab was approved regardless of PD-L1 status. The winning pembrolizumab trial KEYNOTE-024 really tightened the requirement of PD-L1 expression: candidates for enrollment had to have more than 50% of tumor cells expressing PD-L1. This is only observed in about 20% of lung tumors, so clearly the trial volunteers represented a highly selected group of patients versus the virtually unselected patients in the nivolumab trials.
BMS has tested nivolumab as first-line treatment in two trials. The first, Checkmate-012, had innumerable patient cohorts and eight different treatment groups. This trial reported an objective response rate (ORR) with nivolumab alone of 23% (nine of 32) in patients with any degree of tumor PD-L1 expression and 14% (two of 14) in patients with no PD-L1 expression. Four patients had a complete response. These results with a small number of patients were quite encouraging.
Checkmate-026 directly compared nivolumab to chemotherapy. It enrolled patients with PD-L1 expression in more than 1% of tumor cells. The results of nivolumab activity compared to chemotherapy (appropriate for each subtype) were disappointing. PFS was 4.2 months with nivolumab and 5.9 months with chemotherapy. Overall survival was 14.4 months for nivolumab versus 13.2 months for chemotherapy, perhaps reflecting the known durability of responses to anti-PD-1 drugs. Oncologists are still trying to understand the possible reasons for the disappointing PFS data, but the prevailing opinion points to the very low threshold for PD-L1 expression (1%) as the culprit. For now, congratulations to Merck on designing trials in which patient selection is more likely to guarantee success.
As would have been expected, the FDA (seemingly enamored with checkpoint drugs) was quick to follow submission of the results of Keynote-024 trial with an expedited approval of pembrolizumab as a first-line treatment for NSCLC patients whose tumors express PD-L1 in more than 50% cells. Well, think again about the fact that this high PD-L1 presence is only observed in about 20% of NSCLC patients, limiting use of the newly approved drug to one in five newly diagnosed patients.
One is left to wonder if BMS will try again, with a better patient selection, for a trial with nivolumab as first-line treatment in NSCLC. I would guess not. Interest already shifted a while ago to combination treatments, with new results coming in just recently.
KEYNOTE-021 asked if the addition of pembrolizumab to the standard doublet chemotherapy (treatment with two chemo drugs) improves outcomes compared to doublet chemotherapy alone. The results were published in November, showing that the trial has met its primary overall response rate (ORR) endpoint, with 55% ORR in the combination treatment group versus 29% in the chemotherapy-alone group. This trial accrued patients with different levels of PD-L1 expression, and, as might have been expected, those with PD-L1 in more than 50% of tumor cells had better responses to pembrolizumab + chemotherapy. What is interesting is that patients with high PD-L1 also had better responses to chemotherapy alone. Does this mean that very high PD-L1 expression is predictive not only of responses to anti-PD-1 drugs but to chemotherapy as well? Something to ponder in the future.
Researchers also reported additional data from the above-mentioned Checkmate-012, a complicated trial that, among its many drug combinations, also combined nivolumab with different chemotherapy regimens in different types of NSCLC. The best response rate was observed in patients who received a combination of nivolumab with carboplatin and paclitaxel: 47%. Overall survival was also significantly improved for patients who received this combination treatment. PD-L1 expression appeared to play no role in treatment responses.
With apparently high response rates in combination trials, adding checkpoint drugs to standard chemotherapy will be—and already is—further explored for newly diagnosed NSCLC. However, in the trials discussed above, more patients discontinued combination treatment due to its higher toxicities (side effects). This is obviously concerning, and the question is: what is in store for patients who cannot tolerate the combined treatment? Can they continue treatment with either chemotherapy or an anti-PD-1 drug? And if yes, does it have a chance to work?
Companies are pushing ahead with combination trials of checkpoint inhibitors and chemotherapy in a first-line setting. BMS has an ongoing phase III trial, CheckMate-227, that is investigating nivolumab alone, in combination with ipilimumab (Yervoy), or in combination with chemotherapy, all compared with platinum-doublet chemotherapy. Merck is running the phase III Keynote-407, with its familiar cleaner design of pembrolizumab plus chemotherapy versus chemotherapy alone. There are also other trials of pembrolizumab with chemotherapy in a first-line setting, such as Keynote-189 for non-squamous NSCLC.
Once more results are available, we can expect a new round in the war of checkpoints.