Evolution of Predictive Biomarkers for Metastatic CRC - Episode 12
Transcript: Richard Kim, MD: One of the drugs that’s being looked in colon cancer is a stem cell inhibitor. So napabucasin is a stem cell inhibitor that blocks the STAT3 pathway. Activation of the STAT3 pathway can actually cause many issues: it can cause oncogenesis, it can increase angiogenesis, and it can upregulate PD-L1 [programmed death-ligand 1], thus allowing tumor cells to evade immune mechanism. Therefore, based on that rationale, napabucasin was studied in colon cancer in a phase III study. This was a study in patients who were refractory, who got randomized to napabucasin versus placebo. And the primary end point was overall survival. Unfortunately, the study was negative.
However, they collected tissue and they were trying to look at any biomarker for the study, and they looked at a phospho-STAT [pSTAT] biomarker. About a quarter of the patients were positive for this pSTAT area. And in those patients there was an improvement in overall survival of about 2 months—5 months versus 3 months—telling us that this pSTAT biomarker may become a potential biomarker for a stem cell inhibitor. However, this is a post hoc analysis, it’s not ready for prime time by any means, but it is something of use.
Now, based on some of these data, at this time there is a randomized phase III study using a combination of FOLFIRI [folinic acid, fluorouracil, irinotecan] plus napabucasin versus FOLFIRI alone in the second-line setting. Now, they’re not enriching patients who are positive for pSTAT. However, they’re collecting tissue in this study to look back retrospectively to see if this is a good biomarker for the stem cell inhibitors. We’re eager to wait for the results of the study, and hopefully once everything is out we’ll find out if pSTAT is a good biomarker for stem cell inhibitors.
Another area of interest right now is that with immunotherapy [I/O], we know that only 5% of patients are candidates for a checkpoint inhibitor. So the biggest research ongoing right now is trying to see if you could convert what we call a cold tumor into a hot tumor. Last year at ASCO [the American Society of Clinical Oncology annual meeting] in summer, there was a nice presentation by the group in Japan where they combined a TKI [tyrosine kinase inhibitor], regorafenib, plus a checkpoint inhibitor, nivolumab. Now, the whole concept behind that is that we know in microsatellite stable [MSS] disease, generally speaking, the checkpoint inhibitors do not work. But by giving regorafenib at the beginning with the combination, you’re able to change the microenvironment allowing the checkpoint inhibitor to work better.
And the result was somewhat fascinating in that response was close to 30% in microsatellite stable patients, which is obviously very high. We know that as a single agent, regorafenib’s response rate is less than 5%. And the single agent, the checkpoint inhibitor’s response rate is close to 0%. So that combination is something that has drawn a lot of attention. Therefore, obviously, we need more data on it and at this time, even at our institution, we do have a phase I/phase Ib study going on that is very similar study to the Japanese group’s with the combination.
The only caveat is that because of the toxicity of regorafenib, in the group in Japan, they used a low dose. They recommend 80 mg of regorafenib versus the full dose, plus a full dose of nivolumab, because of the rash and fatigue issue. I think that’s the way we’re going right now. Obviously, we have to wait for more results. I’m hoping that the same results could be duplicated here in United States, therefore possibly offering this combination option to patients who are diagnosed with MSS colon cancer.
Scott Kopetz, MD, PhD: There’s a number of novel areas coming in research. I think the highest on a lot of people’s radar is trying to understand the potential of regorafenib and nivolumab to provide benefit for a microsatellite stable refractory colorectal cancer, given a provocative Japanese study that reported a high response rate above 30% in the MSS colorectal cancer population. And there are confirmatory studies ongoing in that space.
I think in general, there’s interest in trying to figure out how do we combine targeted and I/O treatments? For example, in the BRAF setting, there’s a study of encorafenib and cetuximab—that’s BRAF/EGFR inhibition—combined with nivolumab PD-1 [programmed cell death protein 1] inhibition in microsatellite stable BRAF-mutated colorectal cancer. And the idea is that it can provide augmented benefit and hopefully longer duration of disease control than the single agents alone.
We’re also very interested in the area of the antibody drug conjugates, which are now starting to show some activity in colorectal cancer in HER2- and potentially HER3-expressing patients. And so, this is another area of interest.
I/O combinations more broadly is an area of continued interest. We recognize that in metastatic colorectal cancer, there is a subgroup of patients that are what we’ll call an “immune desert.” There’s really no engagement with the immune system. And really, we’re thinking about how we use vaccines or T-cells or bispecific strategies to recruit an immune response. But there are others who have really more of an immune-excluded phenotype, so there’s something in the microenvironment—MDSCs [myeloid-derived suppressor cells], for example, or TGF [transforming growth factor] beta signaling—that may be playing a key role in suppressing the immune system. And studies with combination strategies that target the different populations with checkpoint inhibition are ongoing to try to understand, is there a subgroup of patients where a checkpoint inhibitor or I/O combination may be beneficial, even in the setting of a microsatellite stable tumor?
Transcript Edited for Clarity