Yelena Y. Janjigian, MD: MSI gastric tumors are a known entity, and there are several ways to differentiate or diagnose a tumor being MSI. You can look on the protein level or you can look on the DNA level. And all of these are CLIA (Clinical Laboratory Improvement Amendments)—certified and validated biomarkers. Within the protein level, you could look at mismatch repair proteins, and you look at 4 proteins, similar to colon cancer. If any of them are deficient, you will see absence. Unlike any other test in this situation, it is a bad thing if it’s present. That means your patient is not MSI and wouldn’t qualify for pembrolizumab. But absence of this protein staining shows that the tumor is likely MSI. In gastric cancer it’s a rare entity, but in colon cancer or uterine cancer this could be associated with germline Lynch syndrome.

On a DNA level, you can look at this through PCR or fragment analysis, which is also a very direct and quick test. Or—this is where the field is moving, particularly with the TCGA data, and this is my preference—you could look at it in a multiplex fashion by next-generation sequencing. In other words, you are looking at it one time and you’re able to get the MSI status because of the mutational burden of the tumor, and in gastric cancer it’s usually not subtle. You see a mutation profile that’s 2 pages long or there are 30-plus mutations, and that usually signifies MSI. And at the same time, you could make HER2 status known for these patients—all these amplifications are picked up—and learn about the tumor as a whole on a deeper level.

Right now, the pembrolizumab approval for MSI tumors, regardless of organ of origin, is after chemotherapy stops working. So, it’s in a chemotherapy-refractory setting. Theoretically, in the future, we should and would do trials to get pembrolizumab to earlier lines of therapy, but right now the indication is in later lines of therapy. That being said, data from MSKCC (Memorial Sloan Kettering Cancer Center), MRC (Medical Research Council), and others have shown that MSI gastric cancer is actually chemotherapy refractory.

In fact, the adjuvant MAGIC study analysis recently published in The Journal of the American Medical Association Oncology showed patients with early localized tumors, MSI tumors, who got chemotherapy in the perioperative setting did worse than going straight to surgery. And our data from MSKCC in the metastatic setting, which was recently published in Cancer Discovery, shows that MSI metastatic gastric cancer progresses on systemic chemotherapy within the first 4 months of treatment, showing that these tumors are biologically distinct and that immunotherapy should be considered in earlier lines of therapy.

As part of the NCCN compendium and NCCN guidelines, routine testing for PD-L1 and MMR (mismatch repair), or any form of MSI testing, should be performed for metastatic gastric cancer. And really, any clearly certified PD-L1 lab would be sufficient in patients who have both PD-L1 positivity in the tumor and the stroma. That’s an important distinction that you need to communicate to your pathologists because the majority of pathologists are used to taking these tests and doing these tests for lung cancer or melanoma. PD-L1 positivity in these tumors, and the cutoffs, are quite distinct for gastric cancer. So, for the purposes of PD-L1 positivity in metastatic gastric cancer, 1% of positivity in the tumor or the stroma is sufficient to cause a tumor to be a PD-L1 positive.

The question that often comes up is, if this patient went through multiple lines of therapy and they’re still fit and very motivated to go on either a clinical trial or the next generation of therapies, but unfortunately after running 8 immunohistochemistry tests and analyses over the last 3 years that the patient has been with you, you’ve exhausted tissue, what do you do now? I can tell you that the majority of the patients who I discuss this with would not have a problem with getting a second biopsy. In a fit patient who is otherwise motivated to get more therapy, who you think is fit to receive more therapy, a biopsy is absolutely very important to determine PD-L1 and MSI status because these agents can be life changing.

Transcript Edited for Clarity