Round 3: Are We Adequately Stratifying mCRC Tumors to Inform Therapy?

Return to series home.

Apama Parikh: So, finally, today we're going to be discussing whether we are adequately stratifying refractory metastatic colorectal cancer patients or tumors to inform therapy.

Nataliya Uboha: I'll be arguing that we are effectively stratifying patients. Currently, we are offering these agents to patients regardless of biomarker selection, but several other biomarkers are already part of standard practice in the management of colorectal cancer. We test all our patients for microcellular instability and for BRAF or KRAS mutations, but the presence or absence of these biomarkers doesn't affect whether we offer patients anti-angiogenic therapies or multi-kinase inhibitors. And that's because, as we discussed in the prior section, we haven't been able to find a reliable biomarker that would allow us to either exclude or include patients for the management of this disease.

So I think our therapy selection with these agents aligns with what we know about biomarkers, and they should be offered to all patients, regardless of the presence or absence of specific genetic alterations. And again, as discussed earlier, in large, randomized phase 3 trials, the activity of these agents, regorafenib or fruquintinib, in correct and relevant trials, respectively, has been shown to improve overall survival in these heavily pretreated patients regardless of the biomarker status.

Kanwal Raghav: I appreciate the point you're making, Nataliya. And I do agree that to a certain extent, we've been successful in stratifying patients. But there's another way of looking at it, that advanced colorectal cancer, it really evolves in a highly context-dependent TME, which is shaped by what your baseline genetics were, what your stromal architecture was, what your immune composition is, and how you have been treated. Granted, biomarker-selected subsets such as BRAF V600E, KRAS 12C, and HER2 amplification, as we've discussed, show clinical data demonstrating substantially higher response rates than in the unselected population. And it illustrates the value of genomic stratification, which can identify biologically distinct subgroups. But they are a small set of subgroups, and therefore, I do feel that genomic stratification alone, which is currently being used, is insufficient for refractory disease, where the tumor microenvironment has been increasingly modulated and shaped by our therapeutic stress and response.

Hypoxia, immune infiltration patterns, and angiogenic signaling signatures may better define biologically relevant subtypes. We know that consensus molecular subtyping, both in early-stage and metastatic disease, has shown differential treatment effects with standard of care, biological therapy, and chemotherapy, including differential prognostic values. And when you look at these molecular subtypes, you notice that distinct biology drives each subtype. And when you look at the genomic biomarkers, they're spread across these subtypes. So a BRAF in CMS1 setting may be slightly different from a BRAF in CMS2 setting. And I do feel that future classifications and their frameworks should integrate this multiomic strategy with both tumor-intrinsic genomic alterations and the tumor microenvironment and immune microenvironment, if we are going to move towards any kind of meaningful guided therapy.

Apama Parikh: So, great discussions today around the ability for our current therapeutic strategies to modulate the tumor microenvironment. I think we've heard two very compelling arguments from both sides. On the affirmative side, we have really seen consistently that, albeit limited benefit, I think in terms of what we would all like to see, we're not curing patients; response rates are limited. That being said, we do have multiple prospective trials with superior outcomes for patients using various angiogenesis and TKI different strategies. On the skeptic side, I think what we've heard today is that future progress in refractory colorectal cancer is really going to require us to better understand the classification of patients, not only characterizing the tumor, characterizing the tumor microenvironment, and really biologically understanding what is happening, not only to the tumor, but the microenvironment with the various agents that we're using, and really understanding what is happening on the angiogenic side, what is happening on the hypoxia side.

As we modulate this, what is happening with immune infiltration? Are these just tumors that are entirely immune-excluded? And regardless of how you modulate the microenvironment, if there are no immune cells present, we'll never be able to foster infiltration. What do we need dynamically over different therapies to understand how to modulate the microenvironment? This may change across different lines of therapy. So, the microenvironment and the tumor interactions at first line, with exposure to multiple lines of therapy, may change over time.

So in terms of, I think, where we are today, I think we all, I think, have agreed that we need to move the needle more. I think the bar is quite low, unfortunately, for metastatic colorectal cancer patients, and we really need to start thinking a little more strategically, integrating biomarkers to better understand how therapies across different lines of therapy are informing patient care. So, I guess the final discussion point, in terms of today's question, is: should refractory colorectal cancer be reclassified by biological state rather than by line of therapy? And what would we need to change in terms of trial designs to operationalize this shift?

Kanwal Raghav: This is a discussion, Apama, and you and I have had many thoughts about it. I think if we want to really shift the curve, then we have to shift our thinking. And our traditional thinking of being reductionist as to how cancers behave has only served us so far, but it probably is not going to serve us at some point in time. Even in the most biomarker-selected trials that we have, barring the exception of, say, MSI-high colorectal cancer, our responses are still limited, and resistance still occurs. So we haven't broken this barrier. And Nataliya pointed out very well that you will need paired biopsies. At some point in time, we'll have to ask the hard questions of, can we go forward by the shotgun approach of taking a treatment and throwing it at an unselected cohort, and wishing and hoping that something hits? So I think starting biomarker selection earlier in the trial process and learning about tumor evolution on that treatment are key to designing the next phase of clinical trials.

Nataliya Uboha: I absolutely agree with everything that's been said. I think we need to consider tumor biology when designing a clinical trial. However, there are practical implications for how we design studies and how we report results. We need to consider where patients are on the treatment continuum so that the data we receive are interpretable. So, I think we still need to say: do we want to enroll patients with chemotherapy-refractory disease? Rather than counting lines, maybe state that these patients have progressed on both known chemotherapies and ask whether we want to enroll patients with chemosensitive disease. And that's important if you try to combine normal agents with chemotherapy.

Looking at patients with or without treated liver metastases will also be important. We know, as you've pointed out, Dr Parikh, that these liver meds have very strong immunosuppressive effects, at least in immunotherapy. So, I think we can't just forget about lines of therapy when we design studies, but incorporating biology will also be important moving forward.

Apama Parikh: We'd like to close by first thanking my colleagues for this great discussion. Really important topics discussed today. And I think what we all agree on, and there's no debate on, is that we just need to do better by our patients with colorectal cancer. And I'm hopeful that, in our careers, we will see a paradigm shift in which MSS patients are treated with outcomes akin to those of our MSI-high patients. I think the true advent of immunotherapy was something we were all privileged to see in our careers, and I hope we'll get to see it for MSS patients as well in the coming years, if not sooner.

So, in terms of just wrapping up, I think what we heard today is that we've been able to look at the refractory colorectal cancer landscape of clinical trials, and I think what we heard from the affirmative is that the current therapies, angiogenic therapies, and RTK-based therapies are modulating the tumor vasculature and aspects of the TME to invoke some response. What we've seen is reproducible survival gains across clinical trials. And though these aren't the type of progress we would have liked to see, there is some incremental progress, and it reflects biological activity.

On the skeptical side, and on the con side, what we see consistently is that the immune exclusion phenomenon in MSS disease is, in fact, not just a myth. I think it's a reality. I think we see fewer responses in these patients. When we see responses, they are rare and short-lived. And the hypothesis is that we are temporarily modulating the biology of these tumors with our therapies, rather than truly reprogramming the tumor microenvironment, so the benefits are transient and not lasting.

So, with that, I would like to, again, thank my colleagues and friends for joining us in this discussion today, and look forward to more of these, and hopefully an update when we've been able to move the needle further for MSS patients.