GU Cancers 2026 | KEYNOTE-427: ctDNA analysis in pembrolizumab-treated ccRCC

So, ctDNA, blood-based biomarkers, is a super hot area in oncology in general. Not so much in kidney cancer. This is, in fact, probably the first look at whether ctDNA analysis might predict for outcomes with PD-1 blockade. In this case, with the Keynote 427 trial, which looked at single agent pembrolizumab in clear cell kidney cancer and non-clear cell kidney cancer. You might ask yourself, well, why isn’t there more work being done on kidney cancer with ctDNA? Part of it is because ctDNA is harder to detect in kidney cancer than it is, say, in bladder cancer. And for a large part, because kidney cancer doesn’t seem to shed as much DNA in the circulation as other tumor types where this approach is being used as complementing therapies. So what did we see in this study? Well, one, we looked at the clear cell patients with the older Natera test, so their 65-gene test, which is one of their standard tests. What we found is that in the patients for which we had tumor in blood, which was admittedly only 58 of the total patients of the 110 who went on the clear cell arm of the trial, of those 58, about a third of them, we couldn’t detect ctDNA. And you might ask why? You know, that part, that could be the test. You know, it’s an older technology. The Natera test platforms have gotten better since this work was done. But it also could be the biology that I mentioned before. It’s like hard to detect ctDNA in kidney cancer. So this was not a very sensitive test. It also wasn’t super predictive. It didn’t seem to be associated, being positive or negative, didn’t seem to be associated with response or PFS. It did have a strong signal for overall survival, which is interesting, and with tumor burden. So maybe not surprisingly, the higher your ctDNA levels, the more likely you are to have a worse outcome. Where it was more interesting, though, was the change in ctDNA over time. So we were able to check ctDNA at baseline, and then at cycle two, day one, and at some other time points. And essentially, if you stayed negative, started negative and stayed negative, or went from positive to negative, you had much better outcomes. And in this case, you had better outcomes for response and PFS and overall survival. So as a tool that might help guide patients when they’re on study, on treatment, this seemed maybe more helpful. But obviously, very relatively small numbers needs to be done with much larger data sets, with newer technologies, and potentially combining DNA analysis with protein-based analysis. And there’s some work from MD Anderson and our group looking at combining CTC DNA and KIM1 protein analysis at this meeting. So it’s an exciting new direction for the future of kidney cancer, where we might be able to answer questions like, who should get adjuvant treatment? Who can avoid it? How patients are doing on metastatic treatment before their scans to give us a sense of how we should guide therapy. Potentially, in the case of KIM-1, even early detection of kidney cancer. So it’s a very promising direction that many of our colleagues are going in, and this was the first step in that direction with ctDNA and PD-1-based therapy.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.