Recent Advances in Treatment of ROS1-Rearranged NSCLC

Ben Levy, MD: Hatim, do you want to walk us through some of the emerging ROS1 rearrangement data that we’ve seen recently?

Hatim Husain, MD: The ROS1 space has been equally exciting with more medicines coming into the picture. Historically, we had the medicine crizotinib, which in long-term data sets has showed consistent improvement in overall survival and PFS [progression-free survival], but the limitations are the CNS [central nervous system]. There are also some of the targeting resistance clones. One of the big ones to talk about in the ROS1 space is the G2032R. Like in the ALK space with the G1202R, these analogous resistance mutations are an area of further evaluation.

In the ROS1 space, entrectinib has been particularly important, partly because of the fact that it’s CNS-penetrant, which has an important value vis à vis crizotinib. When we think about ways to further address unmet needs in this space, those medicines that can target G2032R are important. Repotrectinib has been studied and preclinically has shown efficacy in this. In limited clinical evaluations, it has also shown response in the G2032R. As a whole, we’re addressing some of these unmet needs—resistance, CNS penetration—and the clinical end points are going to drive some of the value here.

To talk about some other data that are also being evaluated, some of the other medicines in this space have also been positioned in ROS1. Dr [Martin] Dietrich has spoken about taletrectinib and its TRK-related functions and positioning of clinical trials. It’s also being studied in ROS1. We’ll see in larger cohorts how that medicine shapes up. Brigatinib, used and FDA approved for the ALK space, also had some recent data in the ROS1 space with a response rate of about 26%, which seems lower than entrectinib in those data sets that are coming out with repotrectinib. Those were some interesting data from within the last year.

[We also have] data on repotrectinib showing some important overall response rates that are quite high. In fact, the response rate in the limited cohort was 91%. We’ll see how that shapes up in larger data sets. But importantly, there were CNS responses, and there have been data suggesting response in the G2032R. Let’s see how all these medicines compete. The critical question is finding these patients, and that means next-generation sequencing, fusion-detection strategies around clinical suspension—even with the negative result—and making sure patients are getting access, preferably before I/O [immuno-oncology] because of some of the toxicities that can be seen.

Ben Levy, MD: Great point. First, identifying the fusions are always a challenge. Getting the optimal platform, whether it’s liquid or tissue, is important. Hatim, how do you treat these patients off the clinical trial? Is crizotinib your go-to? Is entrectinib? Do you factor in brain metastases for these patients when you’re making a decision? Or are you using just 1 drug independent of clinical factors?

Hatim Husain, MD: Historically, before entrectinib, my practice was crizotinib. It’s important medicine and had some really good clinical benefits for many patients. With the FDA approval of entrectinib, my practice pattern has changed to adopt entrectinib because of the CNS efficacy. That was enough for me to consider. When I think about a superior medicine, now that we’re going to have perhaps several options to think through in the ROS1 space, I think about the clinical performance with response, survival, and PFS, but also CNS response, management of resistance, and tolerability. My experience is that going forward, let’s see how some of these next-generation or newer medicines pair up against crizotinib.

Transcript edited for clarity.