Recent Advances in Cervical Cancer Treatment - Episode 9

Metastatic Cervical Cancer: Interpreting Data From KEYNOTE-826

, , , ,

Experts take a closer look at the data points from KEYNOTE-826 to consider pembrolizumab’s potential role in metastatic cervical cancer management.


Bradley Monk, MD, FACOG, FACS: Dr Tewari, please comment on the stage IVB population, for which there is controversy. Is that overinterpretation, because in the IVB population bevacizumab didn’t work as well, or is that false discovery?

Krishnansu S. Tewari, MD:In this case, I’ll take the other side of the argument. It works well in the stage IVB population and it should, as those patients are healthier to begin with. We saw an unplanned analysis in the GOG [Gynecologic Oncology Group] 240 trial regimen, the stage IVBs did very well. They’re a healthier cohort, and PEMBRO [pembrolizumab] is great for those patients as well. I am still unsure about BEV [bevacizumab] because, it’s not that it was unplanned, it was a stratification factor.

Bradley Monk, MD, FACOG, FACS: It was.

Krishnansu S. Tewari, MD:I know there’s a lot of maneuvering in the narrative of the paper.

Bradley Monk, MD, FACOG, FACS: There is a lot of maneuvering, that’s a good word.

Warner K. Huh, MD, FACOG, FACS:I want to make one point for any trainees who are listening to this, I think it’s important. When you look at these trials, pay very close attention to the control arm on the survival curves. People look at the difference of the delta for what it is. I’ve learned from people like Dr Monk, Dr Tewari, and Rob Coleman, [MD, FACOG, FACS,] to look at the control curve. You might be surprised what you discover there, what the truth is, just as a teaching point in how to interpret these manuscripts and these publications.

Bradley Monk, MD, FACOG, FACS: That’s the first place I went, was this underperformance as a control arm? If you look at the response rate, and I’m speaking from memory, for the GOG 240 trial it was 48%. The response rate in the control arm of the KEYNOTE-826 trail was 50%, it’s the same. But when you add pembrolizumab, the response rate goes from 50% to 66%. There’s an interval increase, and the magnitude of the increase is greater than what bevacizumab is. Dr Tewari, you said the magnitude of the benefit was 3.7 months for the GOG 240 trial. This is still early, it’s a first interim analysis. The preliminary overall survival benefit is about 8 months. Now the median is closer to 2 years. Returning to the other controversy, the hazard ratio for PD-L1 was close to 1, like 1.0.

Krishnansu S. Tewari, MD:Right.

Bradley Monk, MD, FACOG, FACS: Do we do PD-L1 testing? Or do we say it works in everyone, and it’s such a small subgroup, and the confidence intervals are wide?

Premal H. Thaker, MD, MS:Yes, as an oncologist we’d like to say, “Let’s give it to all,” but I don’t know what the FDA’s doing to decide, and they often rain on our parade when it comes to these options. I do agree that there will be benefit, even if it’s to a lesser extent.

Bradley Monk, MD, FACOG, FACS: Dr Pothuri, you said the existing pembrolizumab approval is a companion diagnostic. If it becomes a companion diagnostic here, you’ll have the same challenge. I say it’s our job as clinical trialists and educators to share with people all the information in a transparent way, and then let them make their own decision. Do you agree, or do you say it doesn’t work in PD-L1–negative patients, with a hazard ratio of 1.0?

Bhavana Pothuri, MD:I agree, you should present the information. We will be tied by the indication that we get. We should present the data and let people make the decision, but it’s not necessarily always up to us.

Bradley Monk, MD, FACOG, FACS: There’s a study called BEATcc, another GOG study, which requires bevacizumab. That’s not my style, I believe if you can give it and it’s reimbursed, give it. However, because the manufacturer of the checkpoint inhibitor atezolizumab also makes innovative bevacizumab, they require it. I always say it takes 2 trials to convince anyone of anything. That trial is in the public domain, it’s completed enrollment. Ana Oaknin, [MD, PhD,] is the PI [principal investigator] from Spain, and we’ll see if that adds further support. Any closing comments on the first-line treatment?

Bhavana Pothuri, MD:I think KEYNOTE-826 is exciting, and I look forward to the approval.

Bradley Monk, MD, FACOG, FACS: Thank you.

Krishnansu S. Tewari, MD:I can’t wait.

Bradley Monk, MD, FACOG, FACS: We have a lot of experience, and we want to teach people how to use these medicines. We have 7 checkpoint inhibitors approved now, dostarlimab was No. 7 in dMMR [deficient mismatch repair] second-line endometrial cancer. Where we are today, with the comfort and familiarity with checkpoint inhibitors, even in cervical cancer, is pretty far down the road. With the tolerability factor, monitoring, and adverse event management, we’ll be OK.

Transcript edited for clarity.