Upfront Immunotherapy + Chemotherapy for NSCLC

Transcript:Roy S. Herbst, MD, PhD: I think it’s very exciting that the combination of carboplatin/pemetrexed with pembrolizumab was approved by the FDA. It’s a small randomized phase II trial. We don’t yet know the biomarker distribution—the PD-L1 for the patients who benefited or not. However, as a practicing oncologist and researcher, I think it’s nice that we have this tool available for our patients. The progression-free survival had a hazard ratio of 0.5; the updated survival, presented at the 2017 ASCO Annual Meeting, showed a hazard ratio of 0.69. This is not significant, but it is turning in that direction.

Also importantly, are the toxicities. When you compare the grade 3 and grade 4 toxicities, they’re not dramatically different when you put the 2 drugs together. I think it’s reasonable that we’re using this and providing it to patients while we figure more things out. Certainly, if we had an immune-oncology combination (I-O/I-O), meaning 2 immune drugs working together, and we could prove that, we might want to use that first. But we don’t. Or, if we had a targeted therapy and immune-oncology combination, we would want to use that as well. In this case, in the real world, many patients come in and you want to give them immunotherapy.

In my practice and in my experience, the only patients who have had long-term survival with non-driver mutated lung cancer are those who’ve gotten PD-1 or PD-L1 inhibitors, so, I want to get this into a patient as quickly as possible. If someone comes in and their PD-L1 is low, or even if it’s 0, and I can give them this drug—pembrolizumab with chemotherapy—I want to do that. Remember, however, that this trial is only for the nonsquamous disease, so it’s for a bit of a limited population. If I gave chemotherapy first, they might not do well. They might never get to the immunotherapy. Our goal should always be to help more patients. Since it doesn’t seem like we’re hurting them in these data that are released, I would definitely do this. There’s a chance of a long-term benefit.

The one caveat is that there’s a group of patients whom you know are PD-L1—high, and maybe, should get immunotherapy alone. Or perhaps chemotherapy even inhibits the long-term benefit. All that needs to be sorted out still, but I’m very in favor of this. I still want to do clinical trials and research, and I offer those to patients at Yale. But, if someone isn’t a candidate, I have been talking to them about this combination.

Corey J. Langer, MD: I had the privilege of being involved in KEYNOTE-021, specifically cohort G, which compared standard chemotherapy in non-squamous non—small cell disease—pemetrexed and carboplatin—to the combination of chemotherapy with pembrolizumab. This was based on preexisting phase I data that showed that combination was quite well tolerated and that response rates, irrespective of PD-L1 level, were north of 55% to 60%.

I had the privilege of presenting the randomized phase II data to the ESMO Congress in October of 2016. The results were quite favorable—even more favorable than I had anticipated. The response rate for the triplet therapy was 55% or 56%, and that’s been updated as of the 2017 ASCO Annual Meeting as it has inched up further. The response rate for the control group was about 28% to 30%. Progression-free survival, as we had seen in KEYNOTE-024, was significantly higher—13 months. This is the first time I’ve ever seen progression-free survival actually penetrate that 1-year barrier, and this was compared to 8.9 months for the control group, which frankly outflanks what we had seen historically. We can’t impugn a benefit, here, based on a poorly performing control group. The control group, essentially outpaced the historic control.

To date, there’s been no significant difference in overall survival. But, at the 2017 ASCO Annual Meeting, the results were updated. In fact, here, too, we are starting to see some separation in the survival curves. The 1-year survival rate for the 3-drug combination was 76% compared with 69% for the control group. The P-values dropped from 0.39 to 0.13. Now, again, it’s a relatively small randomized phase II trial. I think it’s asking a lot to actually see a survival advantage in such a relatively small patient population. But still, all of the results are quite favorable. And as of May 2017, the FDA actually issued conditional approval—accelerated approval—for the triplet combination in non-squamous non—small cell disease, independent of PD-L1 status.

Transcript Edited for Clarity