Management of Metastatic CRC Rooted in Molecular Subsets

John L. Marshall, MD

Treating patients with newly diagnosed metastatic colorectal cancer (mCRC) requires great foresight, explained John L. Marshall, MD.

“The oncologist is the quarterback of the team and makes the decisions, so the oncologist has to know all the plays in the playbook and know when to call the right play,” said Marshall.

Some of those “plays” include the use of molecular profiling at diagnosis to establish the tumor origin and the presence or absence of microsatellite instability (MSI) and alterations in RAS, BRAF, HER2, and NTRK.

Based on these results, physicians can then cultivate an appropriate treatment plan for their patients. For example, if a patient has left-sided RAS wild-type/BRAF wild-type CRC, physicians should opt for an EGFR-directed therapy as it has shown a clear survival benefit in that subset of patients. Moreover, if a patient has a MSI-high (MSI-H) tumor, he/she may benefit from immunotherapy.

Although HER2 amplification has not been as potent of a target as some of the other established alterations, preliminary data with trastuzumab (Herceptin) and lapatinib (Tykerb) have shown potential, and, as such, warrant its incorporation into broad molecular panels.

The oncologist may lead the team, but multidisciplinary management will ensure that the best outcomes are delivered to patients, according to Marshall.

“New discoveries in other diseases and in gastrointestinal cancers will help push the bar forward, as well as the increased recognition that the optimum management of colon cancer—particularly stage IV colon cancer—is a multidisciplinary team sport,” he said.

OncLive: What is the importance of molecular profiling in patients with newly diagnosed mCRC?

Should HER2 amplification be accounted for on molecular profiling panels?

In an interview during the 2019 OncLive^® State of the Science Summit™ on Gastrointestinal Cancers, Marshall, chief, Division of Hematology/Oncology, Medstar Georgetown University Hospital director, Otto J. Ruesch Center for the Cure of Gastrointestinal Cancer, Georgetown-Lombardi Comprehensive Cancer Center, discussed the current and future scope of mCRC.Marshall: Molecular profiling is critical in frontline metastatic colon cancer. You need to know that right from the beginning. You have to know MSI, RAS, BRAF, HER2, and NTRK. Most physicians are doing broad molecular profiling. [In my presentation tonight], I also made the point of [differentiating between] left- versus right-sided tumors. In right-sided colon cancer, EGFR therapies don't seem to work as well. In left-sided colon cancers, it’s really important to know all of the molecular testing. If your patient is in that left-sided RAS wild-type, BRAF wild-type group, there are retrospective data looking at EGFR-targeted therapy. In that 20% of patients, the data suggest that there may be a benefit to using those drugs in that space. In the rest of patients, it probably doesn't matter. You can give everyone FOLFOX and bevacizumab (Avastin), but in that group you need to think twice about EGFR therapy.HER2 is a critical thing to look for in CRC, and how one measures it probably matters. Certainly, the higher the gene amplification, the more responsive [the patient will be to therapy]. Most physicians use immunohistochemistry. Those patients with high [HER2] expression seem to be the ones who benefit.

What has been the impact of immunotherapy in patients with MSI-H as well as microsatellite stable tumors?

Let's face it: this is not as potent of a target as it is in breast cancer, where it totally transforms the disease. It helps. It's generally a refractory question, not one you would bring into first- or second-line settings—at least in today's world. [We’re] trying to take these subsets of patients and [add more data to the mix]. Some of the data with trastuzumab and lapatinib look promising. If you don't look, you won't find [these alterations]. Most people would say it's somewhere around 3% to 5% of patients with colon cancer who have HER2 expression. You have to look for it.You have to know the MSI status of every patient with colon cancer, regardless of stage. Most of us test this using immunohistochemistry first, but you can do this with next-generation testing. Tumor mutational burden helps confirm that. You need that signal in colon cancer for immunotherapy to even be in play. It's not common, so you have to keep looking to find it. Most of your patients will not have that.

What have been the biggest challenges in the field?

When you find it, what do you do with it? The temptation for all of us is to incorporate it as early as possible into treatment. The indication [for immunotherapy] is in refractory disease. There's always that trade-off of, is it worth giving it a try now or holding onto it for later? The other question is, “What drug do you use?” You have 3 different regimens: 2 single-agent regimens and a combination regimen. The combination shows a little bit of a higher response rate, a little bit higher toxicity, and a much steeper cost. Whether you choose 1 drug versus 2 depends on the setting and the patient you're looking at.When you first meet a patient with metastatic colon cancer, you're thinking, “Am I going to cure this patient or not? What's the role of chemotherapy?” The truth is, you’re not going to cure most patients. Surgery is not going to be an option. In essence, you've got a long game to play. I like to think of it as a chess game. Many doctors tend to be heavy-handed with chemotherapy. Some patient’s cancers require being more aggressive [with treatment] or it will progress. However, in the vast majority of colon cancers, you can back off after some initial induction therapy and use maintenance therapy. You can reduce the intensity of treatment.

Are there any studies that support giving maintenance strategies?

Moving forward, what can clinicians anticipate in this space?

What does the future of CRC hold?

It shifts from not how much chemotherapy you can give, but how little chemotherapy you can give and still control the cancer. The fundamental principle is that you're not going to cure patients with chemotherapy; you're controlling their disease. Don't be too heavy-handed, and don't go so far as to create bad toxicities that are never going to go away. Use the medicines and then back off. Use surgery, use liver-directed therapy, radiation—use all the tools that you have to manage patients as best you can, but don't be too heavy-handed with your chemotherapy.Many studies support giving maintenance chemotherapy. In fact, even if you start looking in the third-line setting, where you have the new oral agents, the studies are suggesting that they may have better play if you use them earlier. They don't work at all if you save them until it’s too late. They won't rescue a “falling star.” You need to play them earlier in the course of disease before you're recycling chemotherapy.We hope the future of colon cancer holds the discovery of new therapies. We've been a little stagnant for a while, although we've made some progress. Molecular profiling will help. You also need interventional radiology, surgery, radiology, and pathology. It is one of those diseases that increasingly requires the entire team [to effectively treat] the patient.What we really need to do is link our world. Right now, whether you are sitting at home typing in your electronic medical record for your patient, or collecting data sets, we need to link them in some way. Many of us, myself included, are spending a lot of time trying to build the alliances, the infrastructure, that's needed for us to do more rapid, less expensive drug discovery. I'm hopeful that that wall never goes up, so that we can connect ourselves medically more so than we already are, so that we can, indeed, find the cures for these diseases.