Molecular Testing for Advanced Colorectal Cancer - Episode 4
Transcript:Charles S. Fuchs, MD, MPH: The history of EGFR antibodies is really interesting, because when they were first being developed, and cetuximab was first, the hypothesis was that these antibodies work in patients whose tumors over express epidermal growth factor receptor. I have been doing this long enough to remember, most people may not, that you were supposed to test for expression of EGFR by immunohistochemistry on the colon cancer before you could decide to give them the antibody.
The irony is, we found out, after the drugs were approved, that they predict nothing, that patients who have EGFR-positive tumors respond, but so do patients with EGFR-negative tumors. So, it’s a marker for essentially nothing. After that we realized, well, there’s no biomarker that’s relevant. We should just use the clinical criteria we’d otherwise use to give these important antibodies. But what we found out, first pre-clinically, and then clinically, was that KRAS mutations predict resistance. And it makes sense, because the receptor present on the surface of the cell goes downstream through RAS. So, if RAS is mutated and constitutively active, it doesn’t matter what you do to the receptor above it. It’s going to signal positive signals no matter what you do above that. And so it makes theoretic sense and we’ve now accepted the fact that if you have KRAS mutation, you shouldn’t get these drugs.
But, what we’ve subsequently learned, and, again, these things have evolved over time, that we didn’t know it initially, is that the initial mutations in KRAS were in codons 12 and 13, a very small segment of the gene. But, we found out that mutations can arise in other parts, codon 61, 146 of KRAS. And those mutations similarly predict resistance. So, now we’ve got to do multiple spots of KRAS. We then found out that there are other RAS genes, NRAS, and then in fact, to our surprise, if you have the same mutations in NRAS, you shouldn’t give cetuximab or panitumumab; it’s resistant.
So, we’ve now evolved to all-RAS mutations. There are still controversies in this area. There was a time when we heard that if you have a BRAF-mutated colon cancer, you shouldn’t get cetuximab or panitumumab; that was resistant. We’ve since learned, no, in fact, BRAF-mutated patients appear to respond. And, some might argue the jury isn’t in there. But, ultimately, what we do know is that it’s all-RAS mutations. We know that the drug clearly has benefit in multiple lines of therapy if you have an all-RAS wild-type patient. What we still don’t know is how to best sequence this, and there’s a debate. Should it be part of first-line? Should it be part of second-line, third-line? That’s not as clear. And, hopefully, as we begin to learn more about the genomics, maybe we can make those decisions better.
Richard Kim, MD: The RAS group of proteins usually consist of RAS, the HRAS, and the NRAS. In the past, we’ve looked at KRAS codon 2, exon 12 and 13, which is about 40% of the patients with colon cancers. And the patients with those mutations were actually resistant to EGFR drugs. So that’s a fact that we’ve known for past many years. But now, they’ve looked beyond the typical KRAS codon 2. They looked at codon 3 and 4 in KRAS.
They also looked at HRAS and NRAS. And what they found out in the larger study was that about 15% of the population had extended RAS mutations other than your typical KRAS. And, in the patients with other RAS mutations than your KRAS mutations, those patients were also resistant to EGFR drugs, especially in the subset population of a larger study, such as 53, the PEAK study, and the PRIME study. They found that, that patient who had wild-type KRAS codon 2, but were mutated in other areas, were also resistant to EGFR drugs. Not only that, in some cases it was even detrimental.
Daniel G. Haller, MD, FACP, FRCP: The optimal timing for RAS testing is not universally agreed upon. When we all thought that cetuximab, for example, might be a good adjuvant therapy, then the optimal timing would have been on every patient at the time they had their surgery. Unfortunately, that turned out not to be true. Cetuximab in two large trials was not beneficial in the adjuvant setting, in part because the EGFR pathway may not simply be as important in micro metastases as they seem to be in metastatic disease.
I think with the publication of the data from the FIRE-3 trial in Europe, which showed that cetuximab was better than bevacizumab in first-line therapy in terms of overall survival, more people are doing RAS testing before first-line therapy. I think in truth, if you’re being fair with a patient and you’re saying that you’re going to explain all their options to them, you can’t discuss all their options until you know what the RAS testing is. If you don’t do it before first-line therapy, then you already have made the decision that you’re going to treat all patients with a VEGF treatment that is bevacizumab, for which there’s no predictive marker.
So, I think if you’re intelligently discussing with a patient the pros and cons of first-line EGFR therapy, or any biologic for that matter, then you need to know the RAS testing. Also, if you think of a continuum of therapy, a phrase used by Al Benson, and Axel Grothey, and others, thinking of not only what you’re going to do right now, but what you’re going to do down the road for patients, then knowing what their options are during that time, is extremely important. So, I think most people should have RAS testing before you start first-line therapy.
Transcript Edited for Clarity