Adding Immunotherapy to First-Line Chemotherapy for Esophageal Cancer: Data and Implications From KEYNOTE-590
Dr. Smyth: Hello. Welcome to PracticeUpdate. My name is Dr. Lizzy Smyth and I'm here this evening with Assistant Professor Dan Catenacci. Welcome, Dan.
Dr. Catenacci: Hey Lizzy. Pleasure to be here.
Dr. Smyth: Today we're going to talk about some of the big news that we had at ESMO 2020 around gastric and esophageal cancer. We've already touched on CheckMate-649. KEYNOTE-590 was another trial in treatment of advanced esophageal cancer. Could you tell me a little bit about this study, and what were the key differences between KEYNOTE-590 and CheckMate-649?
Dr. Catenacci: CheckMate-649 is a first-line study for gastroesophageal adenocarcinoma, and you went through that data very nicely. KEYNOTE-590 is a first-line study, but for esophageal cancers that included both squamous cell and adenocarcinomas of the esophagus and GE junction. And the makeup of the study was about 70% squamous cell. So it's a substantially different study makeup of patients compared to, say, CheckMate-649.
And it's actually built on the previous study, which was KEYNOTE-181, which was the second-line version of the same study design, randomized to pembrolizumab or to paclitaxel chemotherapy. And similarly there were squamous cells and adenocarcinomas included in that study. And in the US, at least based on KEYNOTE-181, that led to approval of pembrolizumab only in the subgroup of patients that were squamous cell and CPS PD-L1 score of 10 or higher.
Then this KEYNOTE-590 was now looking to see if that could be moved to earlier lines and was a large 700-plus, randomized study of cisplatin and 5-FU, with or without pembrolizumab or the placebo control.
Dr. Smyth: You mentioned that in CheckMate-590 there were patients with squamous and adenocarcinoma histology. Were there any significant differences in the outcomes of those patients?
Dr. Catenacci: In KEYNOTE-181, yes, there was clear difference in terms of outcome in the second line. In the first-line study, interestingly, just some makeup of the study, about 50% were Asian, about 50% were CPS 10 or higher, and about 73% were squamous cell. And so what we know about the adenocarcinoma in this particular study, interestingly, it had a hazard ratio very similar to squamous cell. Squamous cell was 0.72 and adenocarcinoma was 0.74. But there was a wide interval in range around that adenocarcinoma subgroup.
But if you look at the CPS scores, those that were above 10 clearly derived benefit and those that were CPS less than 10 were sort of right around crossing 1, essentially. And so, similar to 649, how we just discussed, I think we haven't seen all the data yet to really make an interpretation about these subgroups that are repeatedly showing the same sort of outcomes. And before I really fully commit on my comments I'd like to see those data and longer-term follow-up.
Dr. Smyth: Do you think that for CheckMate-590, it's practice-changing for what groups of patients?
Dr. Catenacci: For KEYNOTE-590, I think that the practice-changing obvious ones would be at least for squamous cell, CPS 10 or higher. That's sort of the one that we would all expect and we would all see. Then, I think it wouldn't be surprising to me if the CPS 10 or higher adenocarcinomas show a benefit there. Unfortunately, in the KEYNOTE-181 that wasn't the case, even in the adeno, CPS 10 or higher.
Dr. Smyth: Yes.
Dr. Catenacci: But I'd like to see those data to really see that and then we can make that decision. And then it would not surprise me if the CPS less than 10 patients don't quite appear to be deriving the same benefit, in both the adeno or the squam. And again, I'm only surmising because we've not seen the data.
Dr. Smyth: Again, the data is not really mature at this point. I suppose all we've seen is the abstracts. Do you consider oxaliplatin and cisplatin, then, to be interchangeable? Because I certainly have a preference towards oxaliplatin for my patients, regardless of whether they're junctional tumor, esophageal tumor or gastric tumor. Do you think that they're interchangeable with pembrolizumab and nivolumab for these patients?
Dr. Catenacci: It's a great question and its sort of one of the many possible reasons why that's being sort of rationalized why these studies might be different between, say, KEYNOTE-062 and CheckMate-649, and TRACTION-4 and KEYNOTE-590. I agree with you. I don't think I know anybody who prefers cisplatin over oxaliplatin, just for tolerability purposes. And then there are other data, as you've mentioned, and many have mentioned, that there may be some sort of better synergy with oxaliplatin with IO agents compared to cisplatin.
But the counterpoints to that are that other tumor types have shown benefit with cisplatin. This study is showing benefit, at least in the subgroup, with cisplatin. And also we see in TRACTION-4 they used oxaliplatin and there wasn't a survival benefit. So I think that of all the possible options here, I think that's the one that's least likely.
The only thing I will say, is that if IO takes some time to kick in and work, I think that more patients stay on oxaliplatin longer and that the PFS is a little bit longer, even in randomized studies of oxaliplatin and cisplatin, and maybe there it's just allowing the IO some chance to work, as a possibility. That would be my sort of concession to this oxaliplatin-cisplatin argument.
Dr. Smyth: Thanks, Dan. Any final questions on KEYNOTE-590?
Dr. Catenacci: I think that we are all excited to see positive data in the first line in many instances, including this study. I think it will change the standard of care for the obvious squamous cell, CPS of 10 or higher. And I probably will still prefer to use oxaliplatin regimen there with FOLFOX, frankly. And then it'll be 'to be determined' whether or not we're going to be able to use this in lower levels. And we'd like to see those data. Stay tuned to future meetings for us to discuss that.
Dr. Smyth: Thanks very much, Dan.
Dr. Catenacci: My pleasure
Click on any of these tags to subscribe to Topic Alerts. Once subscribed, you can get a single, daily email any time PracticeUpdate publishes content on the topics that interest you.
Visit your Preferences and Settings section to Manage All Topic Alerts