Transcript:John Marshall, MD: Tony, we’re doing molecular profiling in our metastatic patients. We think a lot about stage II needing to know MSI [microsatellite instability]. I even think we ought to know in stage III, although I’m not sure what to do with it. What’s your practice before making a treatment decision around MSI in the stage II and stage III patient?

Tanios Bekaii-Saab, MD: I think every patient with colon cancer should get their MSI. That’s being done at most of our institutions. MSI has 2 roles. One of the primary reasons we get it is because it may lead us down the pathway of Lynch syndrome, and identifying Lynch syndrome would be important for not just the patient, but their families. It also affects how we follow up those patients, regardless of their risk. Second, it’s a good prognostic element, not that that matters a lot in some ways, because it doesn’t change what you do. Third, it seems to negatively predicate the efficacy of 5-FU [5-fluorouracil], not for oxaliplatin plus a fluoropyrimidine, but for 5-FU alone. That may sway the decision-making process, especially in stage II, but also in stage III when you’re thinking, perhaps, about maintaining some of these patients on 5-FU after you’re done with your 3 months of oxaliplatin.

John Marshall, MD: Mike, I had this rash of patients, and one doctor actually called me to ask about an over 70-year-old in a stage III setting, where they found him to be MSI-high, an acquired non-Lynch MSI-high. They’re like, “Well, what should I give this patient?” I’m struggling between a watch and wait, for all the reasons Tony just said—oxaliplatin not really that helpful in the over-70 crowd, and we can debate that a bit—versus giving them 5-FU—based, oxaliplatin-based therapy. What’s your take on an over-70 stage III MSI? We’re finding more of those.

Michael Morse, MD: That’s true. The current thinking is that stage II and stage III are biologically different, and the MSI seems to have a different impact in stage II and stage III. I treat the stage III patients with FOLFOX [folinic acid/fluorouracil/oxaliplatin]. I agree with you that the ACCENT [Adjuvant Colon Cancer End Points] database raised a question about benefits in patients older than 70, but I think most of us feel that it has a lot to do with performance status and ability to tolerate the therapy, and less so that there’s something magic about the older-than-70 patients not necessarily benefitting from the therapy.

John Marshall, MD: I always think this MSI-high stage II means it hasn’t spread, which maybe means the immune system is actually recognizing it, whereas in a stage III MSI-high, it somehow already escaped, because it’s growing in a lymph node. Is there any rationale to that, or is that just my craziness?

Michael Morse, MD: No, I think there clearly is, and obviously, when you get into stage IV, being MSI-high does not confer a benefit to you prognostically.

John Marshall, MD: You’ve already demonstrated an escape from the immune system.

Michael Morse, MD: Yes, presumably, there’s something that’s keeping the immune system from recognizing all those neoantigens at that point.

Cathy Eng, MD: Wouldn’t you want to consider enrolling them in the MSI-high trial for stage III patients?

Michael Morse, MD: Indeed, you would.

John Marshall, MD: This is a struggle, right? That’s back to my initial question to Tony. When do you test? When do you know? A lot of times, we don’t know. Some of our pathology partners are going ahead and reflexively doing at least IHC [immunohistochemistry] on these, so we often have this when we’re making the decision, but sometimes we don’t. Do you wait?

Tanios Bekaii-Saab, MD: Well, that’s a tough one. Do you wait? Ideally, yes, but how long? We’re limited by time. Oftentimes, we see these patients about 5 to 6 weeks into their surgery, and this is a quick test, but if their surgery was outside, then to get their tissue and do all those, you may miss your window of opportunity. We know we lose benefit beyond 8 weeks. Ideally, yes, you should wait and have this as an element.

John Marshall, MD: You could do that fairly quickly at most of our laboratories.

Tanios Bekaii-Saab, MD: Absolutely, as long as you have the tissue. Actually, we have to really train all our pathologists to make sure, regardless of the setting, to have this as a part, the same way you count the lymph nodes.

John Marshall, MD: Same way they do ER/PR [estrogen receptor/progesterone receptor testing] on every breast cancer without thinking.

Tanios Bekaii-Saab, MD: Exactly. It shouldn’t even be a thought.

Transcript Edited for Clarity