Dr. John Strickler:
Welcome to The ASCO Post Roundtable Series on targeted therapy for colorectal cancer. I'm Dr. John Strickler, GI Medical oncologist and Associate Professor of Medicine at Duke University Medical Center. Joining me today are two of my colleagues, Dr. Stacey Cohen and Dr. Harsh Singh. If you could both introduce yourselves.
Dr. Harsh Singh:
Hi, I'm Dr. Harsh Singh. I'm a GI medical oncologist and an instructor in medicine at Dana-Farber Cancer Institute in Boston.
Dr. Stacey Cohen:
Hi, I'm Stacey Cohen. I am a GI Medical Oncologist and Associate Professor at the Fred Hutchinson Cancer Center and University of Washington in Seattle.
Dr. Strickler:
Today we will be discussing the evolving targeted treatment landscape for colorectal cancer and integrating these new developments into three patient case studies. Our final installment will focus on the first-line treatment of metastatic colon cancer, and I'll begin with this case.
This is a 48-year-old man who presented to his primary care physician with bloody diarrhea. Colonoscopy revealed a non-obstructing sigmoid colon tumor, and pathology confirmed adenocarcinoma. CT of the chest, abdomen, and pelvis revealed a liver mass and multiple bilateral pulmonary nodules. Biopsy of the liver mass confirmed metastatic adenocarcinoma. His ECOG performance status was good at ECOG 1, and the patient was started immediately on FOLFOX chemotherapy.
At the same time, tissue next-generation sequencing profiling was ordered. Here are the results of that tissue next generation sequencing profile. Here we see the classic APC and TP53 mutations, but the patient had no mutation in KRAS, NRAS, or BRAF. Tumor mutational burden was low and the tumor was microsatellite stable.
Dr. Cohen, here's this patient in our clinic, left-sided RAS wild-type, relatively young and fit. What would you think about next for this patient?
Dr. Cohen:
This is the patient that benefits from almost too many options of what to do. And so I think the NCCN Guidelines really reflect that there is not really one correct first-line option and it needs to be customized to the patient, to the sidedness of the tumor, and to their molecular profile. For a patient such as this who has a left-sided primary and is wild-type for all of the RAS and RAF genes, we really have all of the agents available to us.
If the patient's already on FOLFOX and we find out their molecular profile is as you described, I would typically be adding an anti-EGFR agent, whereas if they had been found to be RAS-mutant, I might consider using a triplet and bevacizumab for a patient who wanted more aggressive therapy. But I typically would be holding off on any immunotherapy unless a patient is microsatellite unstable.
Dr. Strickler:
Great. And Dr. Singh, how does tumor sidedness influence your decision making in this frontline setting? What does the evidence tell us?
Dr. Singh:
I think that we've long known that tumors on the right and left side of the colon behave clinically differently. I think what's become very clear based on some retrospective analysis of tumor sidedness from large phase III studies is that tumors on the left side really seem to be very sensitive to EGFR inhibition, whereas tumors, even though they might be RAS and RAF wild-type, but on the right side, they seem to benefit from that less. There were two large randomized studies, the PEAK study and the FIRE-3 study, which randomized patients to getting doublet chemotherapy with either bevacizumab as the biologic or EGFR therapy. And I think when you break those studies down by tumor sidedness, tumors on the left side really showed an improvement in outcomes when they were treated upfront with EGFR therapy.
These results were solidified by the more recent phase III PARADIGM study, which was a study asking the same question, whether upfront chemotherapy doublet with bevacizumab or EGFR is better for patients who have RAS wild-type tumors. But really asking this with the primary endpoint being overall survival in patients who have a left-sided primary, again showing that these patients live longer when treated upfront with EGFR-directed therapies in combination with chemotherapies.
Dr. Strickler:
Great.
Dr. Cohen, looking at that PARADIGM data, how does that influence your practice? Did it change the way you approach these patients or did it reinforce the way you approach first-line management decisions, at least for a RAS wild-type colorectal cancer?
Dr. Cohen:
I'm going to say "yes" and "yes" to that one.
With the information from FIRE-3 and from PEAK, from 80405, we had this idea that treating patients with left-sided RAS/RAF wild-type would then benefit from EGFR and we expected that to be a very profound benefit based on our retrospective series. The problem with some of these studies is that they were only looking at KRAS exon 2 mutations. So at baseline, we were already biasing the patient population a little bit because typically in the clinic we would want to look at extended RAS testing, looking at exons 2, 3 and 4 in both KRAS and NRAS as well as BRAF, because BRAF-mutant patients tend not to benefit from EGFR. So if we really wanted to ask this question appropriately, we probably should have pared down the patient population a little bit.
But even then, when we look very carefully at the data, yes we do see a survival benefit from panitumumab combined with FOLFOX in the study, but what's interesting is that the curves cross and they seem to cross around the time that patients had progressed on that first-line regimen. So I will say, at the end of the day, we still need to pick a regimen for our patients and we have some retrospective studies suggesting benefit from EGFR for these left-sided patients. So I think it's reasonable to consider it, but I also don't think that we're mandated to give EGFR in the first-line because now that we're treating this prospectively, we're not necessarily seeing redemonstration of some of the benefit that we expected to see from the retrospective series.
Dr. Strickler:
That's an interesting point you brought up, that the curves cross, which means that you almost see limited benefit while the patients are on the anti-EGFR therapy relative to bevacizumab, and it seems to happen later on in the course of the disease. Do you have any ideas of why that might be the case, why that survival benefit happens long after they came off the treatment?
Dr. Cohen:
It just makes me wonder what are the subsequent regimens that they were on? Did other patients get EGFR therapy? Did they actually go on to get irinotecan or sometimes in these trials, what we'll see is when we examine the second-line therapy they got in what we consider somewhat suboptimal therapy, maybe the patients got quite ill and they were not really able to go on what we'd consider standard therapy. It makes me question how do we interpret these results and what is the take home message if what we expected to see didn't really happen and we see that crossover later on?
Dr. Strickler:
So Dr. Cohen, how did the PARADIGM study influence your practice? And specifically, how did it influence how you think about right-sided disease?
Dr. Cohen:
So with right-sided disease, we classically have said they get less benefit from EGFR, and I'm not advocating for patients to get EGFR in the first-line. But at the same time, if you have a patient who's been on a line or two of therapy and they've never received EGFR, if they are RAS and RAF wild-type, I will still try anti-EGFR therapy in a later line because I think that we need to still better understand if those patients benefit at a subsequent time. So for right-sided patients, I will still consider EGFR, but I would not be giving it in the first-line.
Dr. Strickler:
And now for you Dr. Singh, are we done? All we need to check is KRAS, NRAS, and BRAF and if they're all coming in wild-type left-sided, go straight to FOLFOX and anti-EGFR or are there other biomarkers that might influence your decision here?
Dr. Singh:
I think that's a very pertinent question, especially with the advent of liquid ctDNA biopsies and having genomic testing, which gives us a better sense of tumor heterogeneity, and within a few days rather than a few weeks with tissue-based testing. And results, further negative hyperselection for these extended panel of other MAP kinase drivers, whether that be fusion proteins or whether that be MET amplifications or HER2 amplifications was presented a few months ago at GI ASCO this year, where they then further segmented this population in PARADIGM and looked at patients who then negatively hyperselected for truly being RAS/RAF wild-type, by excluding other MAP kinase drivers by liquid biopsy. What they showed was that the results were further strengthened in the benefit of EGFR addition.
I think that has impacted my own practice where when I see a patient I would send tumor-based testing, but then also liquid biopsies, which you can get back sooner, which may even give you a better sense of perhaps some heterogeneity in the clonal alteration which may exist or subclonal alterations which may exist. And I think if a patient truly is RAS/RAF wild-type, then I think that that makes the case that their disease is going to be very sensitive to these EGFR therapies.
Dr. Cohen:
What do you think about the fact that the curves don't separate, though, until really they're off of that first-line therapy? Because that's something I still struggle with when I look at the data.
Dr. Strickler:
Yeah.
Dr. Singh:
This is something which is really interesting, because traditionally we would expect the curves to separate early and then stay separated. What's kind of provocative is that if you go back and look at the curves from the FIRE-3 and PEAK studies, they look awfully similar where there wasn't really a difference in PFS, but the OS was in favor of EGFR therapy.
I don't know the answer. I think I can speculate. One of the things which comes to mind is perhaps we're providing more long-term benefit with these therapies. Perhaps some of these patients are making it to resection more. And that's not necessarily translating into PFS, but we're seeing it in OS. But I would be speculating.
I do agree that these curves don't look like the standard curves you would see when we expect to see an overall survival benefit.
Dr. Strickler:
I agree. It's something that was also seen in FIRE-3 and CALGB 80405, we see that the positive survival impact of the anti-EGFR therapies long after they, long after they should have completed those anti-EGFR therapies and moved on to other lines of therapies. Some have speculated that it's a deeper response with an anti-EGFR therapy. Maybe that provides a little bit more benefit for patients to have a greater depth of response, but there could be something going on at a molecular level that's very favorable for these patients.
Dr. Singh, it seems like the strongest competitor to frontline FOLFOX/anti-EGFR would be FOLFOXIRI/bevacizumab, which also has outstanding data. Why not just throw it all together and do FOLFOXIRI with anti-EGFR? Has that been looked at and what has been shown with that combination?
Dr. Singh:
Yeah, I think this was looked at in the triplet study, again looking mostly at a population of left-sided tumors focusing on RAS wild-type disease, where FOLFOXIRI with EGFR therapy was compared head-to-head with chemotherapy doublet with EGFR therapy. Strikingly, there was no difference either in responses, either in the depth of response or in outcomes with PFS or OS. And I think what that speaks to is these tumors perhaps are much more chemosensitive and much more sensitive to EGFR inhibition, where we perhaps are not getting our bang for our buck for adding the third chemotherapy regimen. So I think that the triplet study answers that question, at least personally, I would not offer a triplet chemotherapy with anti-EGFR based on that.
Dr. Strickler:
So it sounds like we all agree that in the first-line setting, tumor sidedness does influence our choice of therapy. If there's a right-sided primary, we're generally thinking about a chemotherapy plus bevacizumab combination and for a left-sided primary, if a chemotherapy doublet is used, then anti-EGFR offers superior survival to an anti-VEGF, but we also have a triplet option with bevacizumab.
Additionally, the point was made that negative hyperselection, meaning to select for patients who are RAS, BRAF, wild-type and HER2 nonamplified as well as other biomarkers may further refine the patient population most likely to benefit from anti-EGFR therapies, and I think this illustrates the fact that outcomes are optimized for our patients when we use these precision cancer medicine approaches and combine those approaches with individual patient factors.
This brings us to the end of this case. Please see the other segments for further discussion about the latest data in colorectal cancer or visit ascopost.com.
