Rationale for I-O in mCRC

Transcript: John L. Marshall, MD: Chris, in every other disease we would be spending lots of time around immunotherapy [I-O]. But in colorectal cancer it’s still sort of subplot. So maybe talk a little bit about where we are—the big picture around immunotherapy in colorectal cancer, some of the testing, some of the approvals—and then let’s look at some of the new evidence.

Christopher Lieu, MD: You’re absolutely right. Immunotherapy is the hot topic, but for patients with microsatellite stable [MSS] colorectal cancer, the options have been literally 0. The response rates have also been 0. So microsatellite instability—about 4% of our patient population with metastatic colorectal cancer [mCRC] will be MSI-high. We know that these patients have a high tumor mutational burden [TMB]. There’s now a tissue agnostic, the first-ever FDA approval for pembrolizumab in this area for patients with MSI-high cancers, and that’s all cancers but obviously it includes colorectal cancer.

For nivolumab, another anti—PD-1 [anti-programmed cell death protein 1] drug, there’s an approval in MSI-high colorectal cancer. So this is part of the importance of knowing the MSI status either by IHC [immunohistochemistry] or by next-generation sequencing or PCR [polymerase chain reaction]. That’s the importance of getting this testing early, to know what my options are. Even though it’s a vast minority of the patients, it’s still a critical portion of patients that could benefit from immunotherapy.

We also know now that there’s an FDA approval for nivolumab and ipilimumab. That combination to buy you a higher response rate compared to single-agent therapy, alone. And so, that combination is approved and is something that practitioners can use as well.

Howard S. Hochster, MD, FACP: I wanted to remind everybody that probably more than half, maybe 60% of the MSI-high colon cancer patients will be in Lynch [syndrome] families. These are people who have inherited deficiencies of mismatch repair [MMR] enzymes, and this can be very important for other cancers in their family members, like if somebody has endometrial cancer or something that’s also Lynch-related. So it is important to explore the genetics of this as well in these patients.

John L. Marshall, MD: And the germline.

Howard S. Hochster, MD, FACP: Yes, germline.

Tanios S. Bekaii-Saab, MD: So somatic and germline respond equally well, just to be clear about that. Now, I want to go back to the nivolumab and ipilimumab. Be careful to use the word “compare,” since they were really not compared.

Howard S. Hochster, MD, FACP: That’s right.

John L. Marshall, MD: Right.

Tanios S. Bekaii-Saab, MD: We don’t know. I mean there’s always the bias. You’re going to choose the better patients for the more aggressive regimen. So there could be a bias that was introduced to the study. What we know, even when you look at those waterfall plots, is that for many patients you really don’t need both. Actually I’d probably think for most patients you don’t need both. There’s added toxicity and added cost, which certainly is a problem with ipilimumab. In fact, I was a bit disappointed when I looked at the publication.

They actually allowed the curves of the 2 to be put on 1 plot where there was no randomization, no direct comparison, which is very disappointing because it essentially reflects poorly on the true value of the results. The value of the results essentially are, and we know from other diseases, that a CTLA-4 [cytotoxic T-lymphocyte—associated protein 4] inhibitor may add value. But for most patients, it doesn’t. So in my practice, personally, I haven’t adopted the dual.

John L. Marshall, MD: A long way of saying you’ll stick with a single agent for right now.

Tanios S. Bekaii-Saab, MD: But I could introduce the CTLA-4 in those patients who may have a little bit of a response and then start progressing a little bit. I may actually go with that if I think they’re eligible. And that’s not based on anything specific except some case reports here and there. But mostly I stick with the single agent.

John L. Marshall, MD: Dr Hubbard, what are your thoughts?

Joleen M. Hubbard, MD: I agree with Tony in this situation. I think we have to think about 2 things. We need to think about overall survival, but we also need to think about response rate. So if I have an MSI-high patient that definitely needs a response, that may be someone where I would consider the dual immunotherapy approach. But if I’m just looking at achieving a good overall survival result, we don’t have clear data that doing the combination with added toxicity, that of financial toxicity, is going to improve overall survival. So if I don’t see a strong rationale to do the combination, I will just do the single agent.

Howard S. Hochster, MD, FACP: Could I disagree?

John L. Marshall, MD: Howard, let me set you up on that too. So to me, we’ve got a biomarker. We’ve got a bunch of different ways to measure it. You alluded to that earlier. There’s IHC, gene testing, tumor mutational burden. And even when we have this positive biomarker and patients know about it, they’re excited about it, we still are in the 50%, 60%, 70% response rate. So clearly, it’s not working for everybody. So expand a little bit on where you are with all of this and maybe a little bit about the gene testing and tumor mutational burden.

Howard S. Hochster, MD, FACP: Well, for the mismatch repair enzymes on deficiency, that’s normally tested initially by immunohistochemistry staining, and many people confirm it by a PCR test that actually looks at fragment length. Because when you have the microsatellites, they’re smaller pieces of DNA, and you run a PCR test looking at fragment length, and you can find that they actually have microsatellitosis of their DNA.

So those are the kinds of tests that are out there. Now I think people should really test for this first-line. And I support the people suggesting that patients go on the COMMIT trial, the NCTN [National Clinical Trials Network] trial for first-line therapy of MSI-high patients looking at single-agent, anti—PD-1, or in combination with chemotherapy, or chemotherapy alone. Because the chemotherapy works very well sometimes in people who have MMR deficiency, and we don’t really know the right sequence or if you should just give the antibody in the first-line instead of chemotherapy, and we don’t really have another way to get that answer. So people should do that.

But when it comes to more refractory and the use of single-agent PD-1 versus the combination of NIVO [nivolumab] and IPI [ipilimumab], I found the CheckMate 142 study to be pretty persuasive. I know it’s not specifically randomized, but there are different cohorts that were somewhat contemporaneous. And I don’t think there’s a lot more toxicity for giving the NIVO/IPI at the doses of 3 and 1 mg per kilo, with only 4 doses of IPI, which is the schedule they use, but it does look like it has high response rate and a longer progression-free survival. So I would tend to use that for most of my patients if I’m going on the immunotherapy pathway.

I just want to say one other thing. I don’t underestimate the toxicity. There’s a lot more toxicity even for single-agent PD-1 than people appreciate. It’s not a free lunch. There are a few patients who get very significant toxicities where they land up in ICUs [intensive care units] or they’re on thyroid replacement for the rest of their life. You know, these are part of it and it’s a little bit more with the IPI, but not that much more.

John L. Marshall, MD: We all get a lot of pressure in the microsatellite stable patient or all of our patients to just try a little immune therapy. Is there any role for this? Should we just try it because maybe even in MSS you’ll get a response?

Howard S. Hochster, MD, FACP: No.

Tanios S. Bekaii-Saab, MD: Outside of a clinical trial, the answer is an absolute no. The likelihood of a response in a patient with MSS and tumor mutational burden low—we have to emphasize the TMB component—because you can have some non—MSI-high, TMB-high that could eligible for immune therapy. The likelihood of response is 0%. In fact, I’d say minus 10% because you get the added toxicity, the false hope, and the cost. And I think it’s detrimental to patients, especially when they reach a time in their life when they should actually move on to the comfort care, hospice care level. So it’s an absolute no.

Transcript Edited for Clarity