MSI-H Metastatic CRC: Chemo Vs I-O Debated

John L. Marshall, MD: There are a couple of things that I would like to talk about. First, we’ve been saying that chemotherapy doesn’t work in MSI [microsatellite instability]-high tumors, at least in metastatic disease. Yes, it does. There’s clearly activity of chemotherapy whether used as frontline therapy or used as a crossover into the second line. So we’ve got to be careful about our dogmas about chemotherapy working or not working.

What’s your theory on why the curves are flipped at the beginning? Marwan, why do you think chemotherapy fairs better for those few months versus pembrolizumab?

Marwan G. Fakih, MD: I think you just have a population within the MSI-high group that is fully resistant to pembrolizumab. Those are the ones who are dropping off. We’ve all seen them. We’ve seen them in the setting of pembrolizumab. We’ve even seen them in the setting of nivolumab/ipilimumab. We’ve reported that some of these patients have a bit lower TMB [tumor mutational burden] than the average patient—like a TMB of 25, for example. There are also some patients for whom you do an IHC [immunohistochemistry] and they’re missing a MLH1. But then when you do a Foundation or other certified assay, they’re MSS [microsatellite stability]. And they do not have a high TMB. Those would have been included in this study. And frankly, there are also some others with mechanisms of resistance unrelated to TMB. We’ve also seen patients with TMBs of 60, 70, with MSI-high tumors who are blowing right through immunotherapy. We’ll do a scan 2 months later and the tumor has gotten 2-, 3-times bigger. So I think those are the ones we’re selecting.

Which basket does that patient belong to, among those 3, I am not sure. And are these patients even a little bit more resistant to chemotherapy but still derive benefit from chemotherapy? We don’t know the answer to that. But I think those are the ones who are probably dropping off. It would be nice to see, specifically, in that 30% of patients, what their molecular profile is in comparison to others' profiles.

John L. Marshall, MD: Its bugs me because somebody knows the answer to that. I think that’s data that should be released. I’ve never experienced a frontline indication that is tissue agnostic in the end using a biomarker that is so messy, right? You can get in with a simple, single IHC missing, which we know is probably not as good a driver, if you will. And you can get in with a fragment analysis and next-generation sequencing. So any ticket gets you in.

Joleen, do you have any difference of opinion on that first drop? The reason I ask is we’ve also got clinical trials now combining chemotherapy in this setting. If Marwan’s right, that trial might not actually be positive because we don’t have the best biomarker. What are your thoughts on that?

Joleen M. Hubbard, MD: Yes. I agree with Marwan. There’s just something about a small percentage of patients where they blow through that initial immunotherapy. I think the SWOG trial that randomized patients to chemotherapy versus chemotherapy plus immunotherapy versus immunotherapy might help answer that question for those rapid progressors or those patients for whom we should have combined immunotherapy with chemotherapy. We may get some more information from that SWOG study.

John L. Marshall, MD: Thank you. I think it’s an interesting topic, and hopefully we’ll see more data on this to help guide us.

Transcript Edited for Clarity