Andrew Wagner, MD: Sharesh, another drug was recently approved for GIST [gastrointestinal stromal tumors] as well, and that would be avapritinib. Could you discuss a little about its mechanism of action and the evidence that supported its FDA approval.
Shreyaskumar Patel, MD: Absolutely. Avapritinib and ripretinib are the 2 drugs that have gone through some relatively parallel development. They have some commonalities in terms of how active they were in the unique subset of the PDGFR D842V–mutant subset. Whereas we described earlier imatinib doesn’t work and there’s really very marginal limited activity of any of the other kinase inhibitors that we have been describing. The true unmet need was real, for example, in that patient population who had absolutely no standard-of-care treatment.
Avapritinib tested in a single-arm cohort of 50-plus patients had a pretty phenomenal response rate that was upward of 70%, a disease-control rate that was upward of 80% to 90%, and a median survival that was not reached at the time when the analysis was presented. The FDA looked at the drug and application. Because of the phenomenal activity in that particular subset, despite the single-arm nature, no randomization, I think—as the FDA likes to say—clinical benefit was meaningful enough that the drug got its indication in specifically that D842V-mutant subset for the PDGFR alpha group.
There has also been evidence of activity of the drug in the KIT-mutant subsets. It is selectively active in the more common secondary mutations in the KIT-mutant subset—exon 17, exon 18, for example. It’s important to note that even that as homogeneous as we think that population may be, that’s still a heterogeneous population, given the exact sites of mutation in the exon 17 and 18 and the changes in the amino acid sequences.
In the third- or fourth-line setting, in the KIT-mutant subset, avapritinib had a response rate of 20%, 22% which is actually very impressive. That’s a high RECIST response rate. Patients who responded did have a durable response, because the median duration of response was upward of 10 months. The catch is that the overall median PFS [progression-free survival] turned out to be 3.8 months. My interpretation of that data is that the drug is very active in a specific mutant subset. Where it works, it works great, and that is durable benefit. But there is enough for the other population where it has modular or limited activity, which dragged down the median PFS.
The phase 2 data were telling us these numbers. And probably an extrapolation of the data explains the VOYAGER trial, the randomized phase 3 study comparing avapritinib with regorafenib in the third-line setting. There was a press release recently, and it’s public knowledge that the primary end point of progression-free survival improvement was not reached and is not likely to be reached, and therefore the trial has been halted.
Along those same lines, efforts at trying to move the drug in earlier-line settings, in the second line, also likewise have been halted. But coming back a full circle on a positive note, avapritinib from an efficacy standpoint certainly has defined activity and therefore approval in the D842V PDGFRA subset. It does have defined activity in certain specific mutations in exon 17 and 18 that will need to be defined a little better for it to be off any use. From an adverse-effect and toxicity standpoint, some of the common GI [gastrointestinal] toxicities were relatively well managed; they’re manageable. One of the unique toxicities of avapritinib was this cognitive defect.
There were some neurological toxicities, including some intracranial hemorrhages and memory losses, that were an area of concern. There was an active investigation in terms of what is causing it. How can we predict it to try to allow for better management of those things? I think that still will be relevant in the D842V population, where the drug has dramatic activity and is a commercially available agent.
Andrew Wagner, MD: That was a great summary. I would add a few comments. One is that the cognitive toxicity that we’re seeing, I feel like we’re seeing less of it now than we did initially. That’s probably because of greater recognition of this as a potential adverse effect, earlier detection, and then either dose modification or dose interruption or some other strategy to improve it.
For people who are prescribing it for the D842V-mutant tumors, there needs to be really careful attention to cognitive changes and early management. That makes a big difference in the severity of that toxicity. Two other comments. One, I totally agree that there is activity of avapritinib in KIT-mutant GIST. A response rate of, I think, 17 % in the VOYAGER study is really impressive. But the work that needs to be done is to figure out who those patients are or what tumors are responding.
Difficult to do from archival tissue, and hopefully we can make some new set of circulating tumor DNA or other approaches to understand which patients were responding to treatment. And to see if there is still a strategy for using this medication to help control disease or even generate responses. I mean, 17% is higher than we saw certainly with sunitinib and regorafenib and even higher than what was reported for ripretinib.
Shreyaskumar Patel, MD: It’s fair to comment that while we have paid more attention to PFS all along in later-line settings, response rates are not trivial. For patients with bulky disease, we’ll see some significant quality metric reduction of the tumor. It does allow for better symptom management. It absolutely impacts quality of life and their ability to carry on a normal day-to-day routine. So it is absolutely not trivial. And I agree with your additional comments.
Andrew Wagner, MD: The median PFS is a point, 1 spot among 100% of patients, that’s the 50th percentile. If there is a tale that has a good response and if it’s durable, that is an area of activity. The other comment I’d like to make is that the benefit seen in the G842V population is noteworthy. I would make the analogy to imatinib for KIT, standard-of-care mutant GIST. Both of these situations were a disease that had no effective treatment.
Before imatinib there was no treatment for metastatic GIST. Before avapritinib, there was no treatment for D842V-mutant GIST. That is something that should be celebrated: For a rare, rare, rare patient population, there has been a drug developed and it is effective, and it can be safely administered and managed. It’s life changing for this patient population. I think that is something worthy of celebration.
Shreyaskumar Patel, MD: Absolutely. A nonmedical term that is appropriate here is that they’re both game changers in that population of patients. No question, yes.
Transcript edited for clarity.
Andrew Wagner, MD, reports the following disclosures: Consulting with Deciphera, Daiichi-Sankyo, NanoCarrier. He also reports research funding to his institution with Daiichi-Sankyo, Eli Lilly, Karyopharm, Plexxikon, Aadi Biosciences.