Tanios Bekaii-Saab, MD, discusses the emerging role of immunotherapy in patients with MSI-H colorectal cancer.
Tanios Bekaii-Saab, MD
Microsatellite instability-high (MSI-H) tumors comprise approximately 4% of patients with colorectal cancer (CRC), and those patients are likely to respond to checkpoint inhibitors, such as nivolumab (Opdivo) and pembrolizumab (Keytruda), according to Tanios Bekaii-Saab, MD.
An emerging area of research is attempting to have the remaining 96% of patients with CRC respond to immunotherapy. An ongoing phase III study is investigating that question by combining the MEK inhibitor cobimetinib (Cotellic) with the PD-L1 inhibitor atezolizumab (Tecentriq; NCT02788279). Results of this trial are expected to read out in the coming months.
“Strategies such as introducing a MEK inhibitor seem to invite tumor-infiltrating lymphocytes, which could perhaps increase the yield of response,” explained Bekaii-Saab, a professor of medicine at Mayo Clinic.
In an interview during the 2017 OncLive® State of the Science SummitTM on Gastrointestinal Malignancies, Bekaii-Saab discussed the emerging role of immunotherapy in patients with MSI-H tumors and beyond.Bekaii-Saab: My discussion focused on the role of immunotherapy, specifically PD-1 and PD-L1 inhibitors, in patients who have MSI-H tumors. Those are cancers that lack the expression of some proteins. This is a surrogate marker for tumors that are loaded with mutations, such as hypermutated tumors.
The presence of this MSI-H phenotype is dependent on 2 elements. Firstly, it could be acquired through the genetics of the patients, such as in the setting of Lynch syndrome. The other group would acquire this sporadically. This is 4% of patients in the metastatic setting.
In the earlier stages, the percentage is higher. Stages II to III are anywhere between 15% to 18% of patients. In the earlier stages, it is a good prognosis; whereas, in the later stages, it is an incredibly poor prognosis. In those patients with hypermutated or hyperinflamed tumors, there is a higher likelihood to see responses from immunotherapies. When looking across the board at single-agent PD-1 inhibitors with pembrolizumab or nivolumab, there are no responses in CRC.
On the other hand, when looking at the patient population that has this MSI-H phenotype, you do see a response rate up to 50% to 60%. More interestingly, as we look over time in recent publications from the authors of a study that looked at pembrolizumab in patients with MSI-H, many of the patients who had either stable disease or a partial response (PR) ended up with complete response (CR). However, it was an average of about 24 to 26 weeks before they had better responses. Some of these benefits can be seen earlier, but they also accumulate with time. Many of these patients may end up with a CR or a very solid PR after 6 months on the therapy.
In that study, pembrolizumab had 3 cohorts. The first cohort was the MSI-H CRC cohort. The second cohort was non-CRC with MSI-H. The third cohort was microsatellite stability (MSS). The MSS cohort had 0% response or 0% benefit. There was no control arm.
The 2 arms that seemed to benefit were the non-CRC and the CRC arms with MSI-H. Many of these responses that were seen in the CRC arm were very impressive, even more so than the non-CRC arm. The patients with pancreas cancer, endometrial cancer, and other forms of cancer, which is at a lower baseline of MSI-H than CRC, had even more pronounced responses that happened earlier.
Across the board, at least with pembrolizumab, our accumulative knowledge is that patients with MSI-H tumors will have a high likelihood of response or benefit from the application of pembrolizumab.
With nivolumab, our only experience so far is with CRC. There was a study that looked at the effects of nivolumab plus the CTLA-4 inhibitor ipilimumab (Yervoy). That led to an additional approval for nivolumab, similar to pembrolizumab, in MSI-H CRC tumors. In this study, we saw similar benefits with nivolumab in MSI-H CRC. When adding ipilimumab to nivolumab, there seems to be a trend towards improving the likelihood of response, but the toxicities were much worse.
The question as this study continues is whether there is a true benefit to adding the CTLA-4 inhibitor. It’s toxic, expensive, and the likelihood of response doesn’t appear to be significantly better. Although the trends are there—taking into perspective the toxicity and the cost—one would question whether this is going to be a viable option.
At this point, we need to wait for the study to complete before we arrive to any conclusions. What’s clear for patients with MSI-H tumors is that there are 2 options—pembrolizumab and nivolumab. For all other non-CRC tumors that are MSI-H, pembrolizumab is the only FDA-approved drug.
The other question that always come about is, “Since this in 4% of the patients, what do we do with the other 96%?” At some point, we are going to run out of options and we need to determine if immunotherapy is an option for those patients. At this point, it is unclear. We think that there is a way for us to transform cold tumors—meaning those that are not inflamed—and inflame them so they are more likely to respond to PD-1/PD-L1 inhibitors.
Strategies such as introducing a MEK inhibitor seem to invite tumor-infiltrating lymphocytes could perhaps increase the yield of response. As proof of this concept, there was a phase Ib study adding the MEK inhibitor cobimetinib to atezolizumab, which showed some baseline response rates of close to 15% to 17%. Each one of those agents individually expects a 0% response. That was a signal that was promising enough to move the combination into a phase III study.
That phase III study looked at the combination versus atezolizumab alone versus regorafenib (Stivarga), which would be considered the standard of care as the competitive arm. This study has completed and we should have some results in early 2018. That could be the first proof of whether we can take those tumors from cold to hot and apply immunotherapy.
There are also other studies looking at multikinase inhibitors that hit multiple targets, including the microenvironment, making these cells more susceptible to the effects of PD-1 inhibitors.
A lot of work is being done in this area for 96% of the patients. It is clear that, with the MSI-H tumors, we are seeing significant improvements in patient outcomes. Although the studies were small and didn’t technically have a competitor arm, the treating physician who sees those patients in clinic and who has personally treated a lot of patients with these agents are seeing some incredible responses.
The rate is 0.5% to detect an MSI-H tumor, so those patients have incredibly bad outcomes. In my own clinic, I’ve seen a patient go from no response to a CR with MSI-H pancreas cancer. There is a significant role for those patients in that setting.
The more important question is not whether there is validity for these PD-1 inhibitors in the MSI-H tumors. It is, “Where do we place these agents?” These questions are being answered in large phase III studies. A question that is still pending is, “Can we bring the benefits that we've seen with the MSI-H to the MSS?”
Additionally, there is a large study that is led by investigators from Mayo Clinic through the Alliance for Clinical Trials in Oncology, looking at the benefits of bringing the immunotherapy agent atezolizumab with the chemotherapy agent FOLFOX versus FOLFOX alone in the adjuvant setting. That’s another area being explored.Although the field of immunotherapy seems very promising, it still applies to very few patients. Overall, 10% of all patients with cancer will benefit from some form of immunotherapy. Those who have a meaningful benefit are probably less than that. That is not an insignificant number; 10% is a lot of patients and those patients who benefit do benefit well. Unfortunately, for the other 90%, it remains questionable whether you are going to see that benefit or not.
I do not believe that the world is going to shift from chemotherapy to immunotherapy. There will be a role for chemotherapy, biologics, and an emerging role for immunotherapeutic agents. Immunotherapy is going to be a part of our treatment, but it is not going to throw away the other agents we use right now. It's going to be all of these agents used in some shape or form across multiple lines of therapy. The most important thing that we started implementing at Mayo Clinic is that every patient with adenocarcinoma should be tested for MSI. It is a small group, but they will benefit tremendously from the effects of immunotherapy. Testing for MSI-H across the board makes sense. That is the main takeaway point that we need to overemphasize so we can identify those patients who are most likely to benefit.