Panelists consider the potential role for circulating tumor DNA to inform the selection of therapy for patients with metastatic colorectal cancer.
Kristen K. Ciombor, MD: Dr Raghav, let’s think about utilizing circulating tumor DNA [ctDNA] to guide treatment decisions. For instance, we’ve known for a while that anti-EGFR rechallenge can be guided by ctDNA. Do you think that’s a valid method to use? What do you think about some of the data that have arisen out of that?
Kanwal P. Raghav, MD: The [difficulty] isn’t the lack of platforms or identifying the disease. It’s the steps going forward. How do we act on the information we get? Even in molecular residual disease, there’s a lot of sensitivity in these assays. We’re going to identify all these patients. The question is what do we do with the patients apart from a prognosis? With ctDNA, 1 application that’s ready for clinical use is to rechallenge with anti-EGFR. This is based on the precept that whenever you treat patients with RAS wild type with anti-EGFR therapies, which is the mainstay of treatment for colorectal cancer, they all subsequently develop some resistance. These acquired resistances are because of the acquisition of KRAS and NRAS mutations and others. They’re usually of low clonality. With the passage of time, if you remove that treatment pressure, they’re going to go away. That’s why EGFR rechallenge works, unlike many other therapies, such as chemotherapy rechallenge, which is less effective.
Do they go away? Either you can make a clinical judgment and treat patients after 6 or 8 months. In that case, a majority of those would have gone away. But ctDNA gives us a powerful tool to make a subjective assessment into an objective category. For example, 1 of the pivotal trials in this was the CHRONOS study, which took a group of patients who had prior exposure to anti-EGFR and looked at the ctDNA profiling at an average time of about a year from their last exposure. [The investigators] found that about 30% of patients retained those mutations, but the remaining did not. Then they challenged those patients with an anti-EGFR agent. They found a response rate of about 30%, which compares very favorably with what you would have done for standard of care. It was a smaller study, so you need larger validations for it. That was a good proof of concept of how this technology can be used in clinical practice. In cases where they progressed after that anti-EGFR challenge, those patients acquired the same mutation. Now you have this situation where you can cycle patients through an anti-EGFR pulse depending on how they’re doing on their ctDNA dynamics.
Joel R. Hecht, MD: We looked at this a long time ago with Luis Diaz in 1 of the panitumumab trials. The Italian [study] did a wonderful job. But I’m going to go back to my question—the assay they used was not a standard assay. If you’re going to do this clinically, which assay would you use or look for? Are there any specific commercially available assays for a patient in clinic?
Kanwal P. Raghav, MD: Broadly speaking, most of these assays have achieved a sensitivity profile that’s sufficient to look at. In this case, you’re working on finding an alteration. If you don’t find an alteration, it’s a reasonable thing to rechallenge patients. The issue is that there are evolving literature—our group has contributed to some of that—and even lines of therapy can be different. You acquire a lot of these RAS mutations and then treat anti-EGFR in the second or third line of treatment, but when you take the same agent and treat it in combination with chemotherapy in the first-line setting, you don’t acquire a lot of mutations. That’s probably because there are other extra-genomic reasons why patients are developing resistance to these treatments. Now, if you do a ctDNA after first-line progression you don’t see these mutations, do you continue anti-EGFR? No, because there are other mechanisms. That’s where those platforms need to be defined by looking at epigenetics, methylomic patterns, and other mechanisms of resistance.
Pashtoon M. Kasi, MD: Regarding platforms, if we’re talking about serial NGS [next-generation sequencing] testing—we’re getting a better understanding of what these mutations are, but from a cost perspective, you may not need to do the broad panel every time. As we think about the global implication and cost to society, there’s also an issue or a utility of targeted panels that are less costly. [They might] focus on the MAP kinase pathway. where you’re only looking at these specific clones, and you don’t necessarily need data on large infusions or the tumor mutational burden. Those are also aspects that some platforms are looking into from a class standpoint as we get a better understanding of resistance to EGFR blockage, the BEACON regimens, and other stuff.
Kristen K. Ciombor, MD: It’s going to be increasingly important as we identify more mechanisms of resistance.
Transcript edited for clarity.