Michael J. Overman, MD, discusses updated findings from the single-agent nivolumab CheckMate-142 cohort and combination data for nivolumab/ipilimumab from the pivotal trial.
Michael J. Overman, MD
Updated findings from the CheckMate-142 study presented at the 2018 Gastrointestinal Cancers Symposium continued to support the use of nivolumab (Opdivo) alone or in combination with ipilimumab (Yervoy) for previously treated patients with DNA mismatch repair-deficient (dMMR)/microsatellite instability-high (MSI-H) metastatic colorectal cancer (mCRC).
An update of patients receiving nivolumab monotherapy in CheckMate-142 presented by Michael J. Overman, MD, showed that nivolumab continued to provide clinically meaningful and durable responses in the 74 patients treated, with a deepening of responses with longer follow-up.1 At 21 months’ follow-up, the overall response rate (ORR) was 34%, including a complete response (CR) rate of 9% (n = 7), and a partial response rate of 24% (n = 18). The median overall survival (OS) was not yet reached.
A separate abstract presented at the symposium reported data for the nivolumab/ipilimumab cohort of CheckMate-142.2 The findings showed a durable clinical benefit for the combination over a median follow-up of 13 months, with an ORR of 55% and a median duration of response that was not reached.
Based on previously reported results for the 74-patient nivolumab monotherapy group from CheckMate-142, the FDA approved single-agent nivolumab in August 2017 for the treatment of adult and pediatric patients with MSI-H/dMMR mCRC that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan.
In an interview with OncLive at the symposium, Overman, associate professor in the Department of Gastrointestinal Medical Oncology, The University of Texas MD Anderson Cancer Center, discussed both the updated findings he presented at the meeting for the single-agent CheckMate-142 cohort, as well as the latest combination data from the pivotal trial.Overman: This an updated analysis of the CheckMate-142 study. The main arm in that was single-agent nivolumab in a 2-stage design, where we evaluated the idea of using anti—PD-1 therapy in this population, which is dMMR/MSI-H mCRC. Therefore, dMMR patients have a deficiency in DNA repair, which leads to a lot of mutations. There was a hypothesis that this group would be responsive to immune-based therapy, and that is how this study initiated.
We have previously presented and published our results in Lancet Oncology, so this is a further update. The prior data was of a 13-month median follow-up, and this is a 21-month median follow-up for that 74-patient nivolumab monotherapy arm. Basically, what we are showing is that, as previously expected, this therapy works in a very high percentage of patients, and we see that continuing. We do see some interesting issues in regard to depth of response; we previously reported a 3% complete CR rate, and this is now up to a 9% CR rate.
The thing that is the most exciting and interesting is that previously, on our PFS and OS curves, there was some flattening of the curve on the tail end, so now we have further follow-up and we see that the curve stays completely flat. If you look at the PFS curve at 12 months, our PFS is 44% and at 18 months it remains at 44%. Survival is very similar; survival is 72% at 12 months and then 67% at 18 months. There is this kind of stabilization—when this therapy works in those patients, it appears to be durable and long lasting. That is the exciting point of this data; our update supports the idea that this is a durable therapy and it enters that idea of possibly curing patients with mCRC. The data seem to suggest that is the probably the case, although it is a little bit early to make that strong of a statement.It has had a pretty big impact, but the issue is that this therapy is a very active therapy. That is what the previous data have said, and that is what is [also found] in this update. This means identifying these patients is really critical. We have a therapy now that works extremely well and potentially cures a fraction of patients. It has really moved into this idea of universal testing in CRC; this means that all patients with CRC should have MSI testing. That is the push that has been put out, as well as that testing can be by immunohistochemical staining, pathologic complete response, or next-generation sequencing. It doesn't matter how to find these patients, it is just that we should be identifying them. Identifying that group is a critical take-home message.
Further questions include, “Could we do better in this group?” Then, how do we take what we have learned here and apply it to the majority of patients with CRC? That is the 1 frustration; this therapy works really well, but it works really well in about 5% of patients with mCRC. How do we make headway in the 95% of patients who are not MSI-H? There is a lot of ongoing work to kind of identify how to move forward, and there are some other early signs that we might have hints of an approach. However, there is still a lot of work to be done.
In regard to the subset I am talking about [in CheckMate-142], the next step for that subset could be combination therapy. There was a presentation at [the 2018 Gastrointestinal Cancers Symposium] about the first report of combination therapy looking at nivolumab with the CTLA-4 inhibitor ipilimumab. It showed that when you put 2 therapies together, you have a higher response rate and PFS curve.
It appears that this is an immune-responsive cohort; if you actually do further engagement in regard to activating the immune system, then you could have a higher benefit of activity. Right now, the FDA label is for monotherapy in this subset. Further data will be needed to make a definitive statement on combinations. Going forward in the MSI population, there are combinations of immunotherapy. We will get to a point where immunotherapies will be a big player. As I said, there are some hints that combinations have activity, but the challenge is that unlike what we saw with MSI-H CRC— where it works extremely well—when you go to microsatellite stable CRC, it is not intrinsically immune responsive. We are trying to develop a way to generate immune response, so we are seeing lower thresholds of activity. That is what we are going to see moving forward.
Do I think we are going to have immunotherapy in the future? We will, but it could very well be in a minority of patients. How durable is that? That is something we need to work on. In the long run, there is no doubt that all of us feel immunotherapy is an approach that makes the best sense to what we are trying to get at, which is not just controlling cancer, but eradicating it.
It is what people want and we understand that, and immunotherapy is giving us the tools to get there. Now, it is just a matter of getting more of those tools. We don't just use 1 type of chemotherapy everywhere, and that is going to be the immunotherapy situation. In the 5- to 10-year range, it is definitely a possibility. The take-home message is that we should be doing universal dMMR and MSI testing. That is an NCCN guideline recommendation, so any patient who has a diagnosis of colon or rectal cancer should be tested—period. There are Lynch syndrome—related indications and this indication in the metastatic space. However, with universal testing, you can test your stage III and stage II patients and, if they relapse, you already know. That is one big thing that should be mentioned.
Then, in the metastatic setting, this therapy works extremely well. It is probably the best therapy that we have. The outcomes are better than liver metastasectomy. We are potentially curing a higher fraction than you would with a surgical approach, which we frequently do in patients with CRC. Identifying the group is critical, and then letting them get this therapy is critical. Whether the therapy is nivolumab or pembrolizumab (Keytruda), based on the FDA labels, they are very similar.The future is hopeful. There is a lot of excitement, and a lot of ongoing work. We in the community are very excited about the tools we get to use, and in time we are going to see more new combinations coming into our armamentarium over the next 1 to 2 years, which is going to be very exciting.