2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
The expression of PD-L1 has been at the forefront of biomarker development for PD-1/PD-L1 inhibitors, but there is much uncertainty surrounding its use and other biomarkers are needed.
Sarah B. Goldberg, MD
The expression of PD-L1 has been at the forefront of biomarker development for PD-1/PD-L1 inhibitors, but there is much uncertainty surrounding its use and other biomarkers are needed, says Sarah B. Goldberg, MD.
“There are patients who are very positive for PD-L1, and don’t benefit, and then there are patients who are negative for PD-L1, who do benefit. That’s not a great biomarker. So there are other biomarkers that we haven’t fully developed yet,” said Goldberg.
Specifically, in non—small cell lung cancer (NSCLC), the anti¬–PD-1 agents nivolumab (Opdivo) and pembrolizumab (Keytruda) quickly moved through clinical development and into the treatment paradigm. Researchers have now set their sights on the optimal use of these agents in the first-line setting, and continuing biomarker development will serve to advance this effort.
OncLive: Could you provide an overview of your talk here at the meeting?
In an interview with OncLive at the 34th Annual Chemotherapy Foundation Symposium, Goldberg, assistant professor of Medicine, Yale Cancer Center, discussed the use of PD-L1 as a biomarker and other emerging biomarkers in NSCLC.Goldberg: I spoke about PD-L1 as a biomarker and discussed whether PD-L1 should be used when selecting patients for treatment with immunotherapy. PD-L1 is a very important biomarker, specifically when thinking about whether to use immunotherapy for patients in the first-line setting.
We know from a trial of pembrolizumab versus chemotherapy in the first-line [NSCLC] setting that, in that trial, patients were selected based on a PD-L1 cutoff of 50% or greater. And that trial was very positive, with patients benefitting from the immunotherapy versus chemotherapy. So in that setting, it’s very important to get the PD-L1 biomarker, and I think that’s very much accepted now. I think that’s a very clear area in which to use that biomarker.
From there, I think it’s much more difficult, because the trials are variable. PD-L1 is being tested in a variety of different ways. There are multiple different assays, and there are different cut-points that are being used. Some of the trials have used 1%, some have used 50%, 5%, 25%—so, when you’re looking at the different trials, it’s really important to look at which assay is being used, which cut-point is being used. Is it all-comers, is it a specific cutoff being used? Because of that, the data look different when you look at different trials.
Could the variable cutoffs have been an issue, for example, with CheckMate-026, which did not show good results for nivolumab in the frontline NSCLC setting?
But overall, if you look at all the different trials, and try to come up with a comprehensive understanding of PD-L1 as a biomarker, I think generally you can say that there does seem to be an increasing benefit with immunotherapy with higher levels of PD-L1 expression. However, it’s also become clear that even patients without PD-L1 expression—with 0% positive cells for PD-L1 [expression]—they still can benefit from treatment, although there is less of a chance of benefit.I think that could have been part of it. That’s the trial that looked at nivolumab versus chemotherapy in the first-line setting. They used a lower threshold; they allowed patients with 1% positivity, and then in the end, they looked at patients with 5% or greater positivity. And they did not find a survival benefit, or even a progression-free survival (PFS) benefit with nivolumab versus chemotherapy. It’s possible that the reason was the cutoff. If they had had a higher cutoff of 50%, like the pembrolizumab trial did, that might have shown benefit for nivolumab over chemotherapy.
What are some immediate next steps to take with PD-L1 as a biomarker?
The other side to that is, if you look at the patients on that trial that did have 50% or greater positivity, they still didn’t seem to benefit. That was a subset, there were a smaller numbers of patients, and there were imbalances in the trial, so it’s very hard to know. You could tear apart that data in several different ways, yet in the end, we still don’t really know why that was a negative trial.I think it’s something that will continue to be used, for sure, because it’s already part of standard practice. So it’s going to continue to be used, but we have to do better. First of all, I think we need to do better with using PD-L1. We need to try to standardize it, not just use a different assay and cut-point for every trial. We need to try to have a more uniform way of using it.
What are some of those other potentially promising biomarkers?
We should also be thinking about incorporating other biomarkers, because it’s clear that PD-L1 is not the only biomarker that will be useful. There are patients who are very positive for PD-L1, and don’t benefit, and then there are patients who are negative for PD-L1, who do benefit. That’s not a great biomarker. So there are other biomarkers that we haven’t fully developed yet. There are some that look very promising and have been shown to have some correlation with benefit from PD-1/PD-L1 inhibitors, but are not yet part of standard practice; they’re still in development. I think part of it is making the PD-L1 biomarker better, and the other part involves incorporating other biomarkers into practice.I think one that is very exciting is looking at mutational burden. The idea there is, in patients who have lung cancer, overall, they tend to have a higher mutational burden, meaning they have more mutations in their cancers than many other types of cancers. That’s likely because of the incidence of smoking in patients who have lung cancer. And it seems that those patients who have the higher number of mutations, they’re the ones who have a higher chance of benefit, or so it seems, based on some studies that have been done.
Now we need more data to confirm that. So far, it’s been linked to response rate and PFS. We need to see if that also correlates to overall survival. We need to look at it with more drugs, and bigger studies. More work needs to be done, but it’s a really intriguing finding. It makes sense; the more mutations in the cancer, the more neoantigens there are on the tumor cells, the more recognizable it is to the immune system. It makes sense that it should work. And it looks like it’s potentially bearing out in the studies. We just have to see if it really could be a good biomarker.