Questions Remain on Optimal mCRC Treatment

Leonard Saltz, MD

Metastatic colorectal cancer (mCRC) is an area comprised of several unanswered questions, whether it be optimal chemotherapy regimens, use of frontline EGFR-targeted therapy, or the role of tumor sidedness, explains Leonard Saltz, MD.

Moreover, while the FDA approvals of pembrolizumab (Keytruda) and nivolumab (Opdivo) for patients whose tumors are microsatellite instability (MSI)-high or mismatch repair (MMR)-deficient were practice changing, the subset of patients for whom these checkpoint inhibitors benefit is a minority of overall patients with mCRC.

OncLive: Please provide an overview on your presentation on mCRC.

In an interview during the 2017 OncLive^® State of the Science Summit^TM on Gastrointestinal Cancers, Saltz, a medical oncologist at Memorial Sloan Kettering Cancer Center, highlighted the current go-to therapies and overarching questions for mCRC treatment.Saltz: The question of chemotherapy for patients with metastatic CRC remains a complicated one, simply because we have a number of options, all of which have some advantages and disadvantages, so there really isn’t one right answer. It is one of the reasons that, if someone looks at the NCCN guidelines, it really looks like a “smorgasbord” of choices.

The first decision point, when treating a patient with mCRC, is the cytotoxic backbone. Typically, we are choosing between FOLFOX and FOLFIRI. This is nothing new; we have been choosing between these regimens for quite some time now. It would be nice if we could make enough progress that we move beyond that.

This is basically, as I think everybody is familiar with, an every-other-week schedule using a 48-hour infusion of 5-fluorouracil with either oxaliplatin or irinotecan. Now, when we look at the efficacy data of these 2 regimens together, they are quite comparable. The side effect profile is the same in terms of intensity, but different in terms of character. Oxaliplatin is of course more likely to cause neuropathy, and that is more likely to be the grade-limiting step in terms of intolerance. Whereas irinotecan, contrary to what most people think, is not more likely to cause diarrhea but is more likely to cause alopecia and that can be a considerable issue for some patients.

In discussing [these options with] patients—who they are and what matters to them—the degree in which neuropathy would be problematic or the degree of which loss of privacy and self-esteem with hair loss would matter would come into play [to make the] decision.

There are data supporting the use of throwing the whole “kitchen sink” in there and call it FOLFOXIRI or FOLFIRINOX, depending on which way you want to say it. There are data suggesting that this might be better; those data are based on treating patients with excellent performance status and the [studies] are mostly performed in ex-United States sites. With the use of fluorouracil, I would be concerned about American patients tolerating it. I don’t use the FOLFOXIRI regimen regularly; I use it when I have reason to believe that I need a rapid response in terms of tumor regression and the patient is otherwise in good medical shape.

The more controversial issue becomes inclusion of the monoclonal antibodies bevacizumab (Avastin) versus either cetuximab (Erbitux) or panitumumab (Vectibix), the anti-EGFR agents. The most relevant data there come from the CALGB/SWOG 80405 study, which directly compared cetuximab with bevacizumab. In this study, patients and doctors chose FOLFOX or FOLFIRI because we didn’t care; we said, “it’s fine either way.” Then, they were randomized to receive either cetuximab or bevacizumab. For all practical purposes, all of these patients were RAS wild-type. The outcome of the study showed no difference in terms of efficacy in pretty much any parameter one wants to look at.

That said, my personal feeling is that I don’t see a role for frontline EGFR therapy. First, it is a very difficult toxicity for patients to tolerate. The skin rash that we expect patients to get with an EGFR-targeted agent is a debilitating rash and is one of the harder things that we ask people to tolerate. If I can delay that to later in the treatment [course], I prefer it. The other issue is the price of either cetuximab or panitumumab is virtually twice the price of the 5-mg/kg dose of bevacizumab that we would use. This is less expensive and has a better toxicity profile in most patients.

How have you seen the wave of PD-1 inhibitors impact patients with MSI-high patients?

The other important data that came out of this study which corroborated other data but put us on the radar screen, is that the left- and right-sided colons are not the same. The tumors that arise from the left side of the colon have a far better prognosis and they are the ones that may potentially benefit from frontline EGFR-targeted therapy. Whereas, tumors on the right side have a worse prognosis and there was no evidence that the EGFR-targeted therapies offered anything there. My personal style is that I don’t use the EGFR-targeted therapies upfront [unless] one wishes to consider it. It is only in RAS wild-type, left-sided tumors. The issue of MSI is a somewhat complicated one; about 15% of all patients who develop CRC will have MSI-high tumors, and we can look for that either by doing the PCR for MSI or immunohistochemistry for MMR deficiency. However, we know that that is a relatively good prognosis for stage II and III patients, and they have a lower [chance] to go stage IV disease. Whereas, only somewhere between 2% and 4% with mCRC will have MMR deficiency and that is the subset for whom the checkpoint inhibitors appear to be useful.

TRK fusions can be found in patients with CRC. If entrectinib gets approved by the FDA, how might this affect clinical practice?

Specifically, the PD-1 inhibitors, which are far easier to tolerate than the CTLA-4 inhibitors, are the ones that are now approved for treatment. Therefore, it has been appropriate to do genetic screening to screen all patients with MMR deficiency. The patients with metastatic disease who have MMR deficiency have a high likelihood of benefitting from either pembrolizumab or nivolumab. For patients who are microsatellite stable (MSS), we don’t have evidence that any of the commercially available checkpoint inhibitors are useful at this time. There is an active investigation underway to try to figure out how to change that, to see what other immunomodulatory agents might be useful in MSS CRC. However, outside of a clinical trial, we don’t have anything right now. That is going to be a complicated question because TRK fusions occur in 1 in 5000 patients, so we are going to get into a very complicated question: how do we go about justifying screening everybody for TRK? Is that the right thing to do? That is a question that society is going to have to wrestle with.

There are emerging data suggesting that TRK fusions are virtually nonexistent in the presence of RAS mutations, so there may be a subset of patients for whom it is a more rational thing to look for. This is investigational right now; we don’t have TRK fusion—targeting drugs available commercially.

What ongoing clinical trials in mCRC are you excited about?

It creates a real challenge in terms of determining how are we going to figure out which patients have these specific rare mutations that may benefit from precision-type applications of medicine. It is an example of where precision medicine does have a substantial impact, as long as we can find the right patients. It is hard to say what is exciting, because a lot of [studies] are in progress and they are very widely spread out across the world; no one site is developing a tremendous amount of experience. We are hopeful that some of the preliminary data that were presented with the combination of MEK inhibitors and PD-1/PD-L1 inhibitors will turn out to be useful. There were very preliminary data 2 years ago, and there hasn’t been much that we have heard from since, so we are going to have to wait on that.