Today I’m presenting the CREST study, subset analysis for CREST study. The CREST study is a study that looks at individuals with non-muscle-invasive bladder cancer, that’s very early bladder cancer. Non-muscle-invasive bladder cancer has a habit of coming back locally and can spread and cause lots of problems, and indeed can progress and kill patients with bladder cancer.
Intervention with BCG is the current standard of care. What we did in this trial is we gave patients with high-risk non-muscle-invasive bladder cancer standard BCG and in the study arm we added a subcut PD-1 inhibitor called sasanlimab. That’s a drug which hasn’t been licensed in urothelial cancer yet, but there are lots of other drugs with similar mechanism of action, but not subcutaneously, that have.
The prime analysis was actually presented a couple of weeks ago at AUA, and it showed that sasanlimab added to BCG reduces the relapse rate and with a hazard ratio of 0.68. What we did was we looked at key subgroups. We looked at a CIS – carcinoma in situ – subgroup, and we also looked at the relationship with PD-L1 expression, so the T1 subgroup too.
Essentially what we showed, the overall result, I said before, was 0.68. In the CIS subgroup, there’s a little bit of enrichment, a hazard ratio of 0.53. We then tried to link that in with the PD-L1 biology, and we showed that CIS patients had lower PD-L1 expression, which I wasn’t expecting. I’m not a great fan of a PD-L1 biomarker, it’s failed more times, in fact, it’s never really succeeded in urothelial cancer.
We then looked at the predictive value of that PD-L1 biomarker in the ITT population and then the CIS population and some other subgroups too. Unfortunately the PD-L1 biomarker didn’t work again. There was a little bit of enrichment, but it doesn’t tell the whole biomarker story.
Sasanlimab is associated with subcutaneous but it doesn’t mean it hasn’t got side effects, it has the same side effects as other immune checkpoint inhibition, with a chance of life-changing toxicity, and that’s an important concept in some individuals who are not immediately in harm’s way.
There was another component of the trial that we looked at. One of the arms we didn’t give the maintenance BCG; we gave sasanlimab without the maintenance BCG. We stopped the BCG a bit early, and we showed that BCG is really important, maintenance BCG is important too. So we shouldn’t be playing around with the control arm. The addition of sasanlimab reduces the risk of relapse of disease overall. There’s some enrichment in that CIS subgroup but overall it seems to work across the board. We need to do more work with biomarkers to really identify those in harm’s way.
What are the clinical implications of these findings?
The clinical implications of the findings are quite important in my opinion because what we’ve done in this story, the immune checkpoint inhibition story, started 10-15 years ago in urothelial cancer, with heavily pre-treated patients with a few, long-term, durable remissions. More recently we moved it into the frontline metastatic space, where it didn’t work as a monotherapy that well. Then into the muscle-invasive space, adjuvant and neoadjuvant, where we show quite a strong signal, and now in the non-muscle, even earlier, we can show this important signal too.
There’s a second study called POTOMAC, which looks at durvalumab not sasanlimab. It’s a press release, we don’t know the results, but we know that’s also positive. This is underlying and underpinning that principle of early immune checkpoint inhibition changing the biology of the disease. My take on this at the moment is that we need to identify the patients who need immune checkpoint inhibition better, but it’s also true that waiting later – second, third line therapy – is too long, and the earlier you seem to give it, the better it might be.
