Neoadjuvant Immunotherapy Combo Promising in Select Gastric/GEJ Cancers

At this year’s American Society of Clinical Oncology (ASCO) Gastrointestinal Cancers Symposium, results from the INFINITY trial were presented. The phase II study examined tremelimumab (Imjudo) and durvalumab (Imfinzi) in patients with microsatellite instability (MSI)-high gastric and gastroesophageal junction (GEJ) cancers.

In this exclusive MedPage Today video, Marc Shapiro, MD, a medical oncologist at the Cleveland Clinic in Ohio, discusses the results and potential clinical implications of the study.

Following is a transcript of his remarks:

We’re talking a lot about neoadjuvant therapies for gastrointestinal cancer. Memorial Sloan Kettering, of course, presented their data for neoadjuvant checkpoint inhibitors that got a lot of press. And I think we’re seeing some very similar data now in gastric cancer based on a phase II trial that looked at a combination of a CTLA-4 inhibitor and a PD-L1 inhibitor.

Just like in the rectal tumors, these are in resectable gastric cancers, GE [gastroesophageal] junction cancers, MSI-high tumors, and they looked at 18 patients. They did a combination of tremelimumab and durvalumab. The CTLA-4 inhibitor was given [in] one dose to induce T-cell expansion, and then durvalumab was given for 3 months.

Most of the patients went to the operating room afterwards. Of the 18 patients, three patients did not go to the operating room, mostly because they had very good responses and deferred surgery personally. But 15 patients did go to the operating room out of the 18, and 60% had a pathologic CR [complete response; pCR], which is really quite impressive for gastric cancer. Eighty percent had a major pCR rate, meaning that they had less than 10% viable cells.

And you can kind of get some hint as to who these patients are, not just endoscopically, but all the patients with the pCR had negative ctDNA [circulating tumor DNA] levels. There were some differences to try and piece out who’s most likely to get these kind of responses by T stage. So if you look at patients who had T3 tumors, 89% of them had a pathologic complete response. Whereas in the T4 group, only 17% had pathologic CR. These could be node positive, node negative, really the bulk of the tumor was driving what kind of a response they were able to get.

It was associated with TMB [tumor mutational burden], but not PD-L1 CPS [combined positive score]. So really this is a different line that we’re normally used to thinking about for gastric cancers. This is more of our MSI-high tumors.

So the natural step is a cohort two, which has now been opened, which is looking for a nonoperative approach for those with a clinical CR. They’re including only T3 tumors, and so they have dropped T4 because of such low responses that they were getting, low pathologic complete responses.

I think it really all brings us pause. We’re certainly struggling with this in our patients with MSI-high rectal tumors. We know that these tumors are really not all that chemo-responsive, and so none of us feels good about giving them chemotherapy. But I do really pause trying to make standard-of-care decisions based on 15 to 18 patients.