The Hunt for Predictive Markers in Metastatic Colorectal Cancer - Episode 2

Recommendations for Molecular Testing in Metastatic CRC


John L. Marshall, MD: All patients with colon cancer need some kind of gene testing. Metastatic patients need a panel, in my opinion. You need to know about MSI [microsatellite instability]. You need to know whether the patient has RAS or BRAF. You need to know HER2 status. You want to know that right from the beginning because it will affect your overall decision making. How many cards do you have in your hand? How do you want to play with drug A versus drug B? You have to know the results of the molecular profiling early on.

Richard Kim, MD: For patients with newly diagnosed metastatic colorectal cancer, there are certain molecular profiles that we should order at the beginning. This helps us decide how to manage them. One of the predictive markers that we’ve tested for is RAS status. We know that patients who are RAS mutant do not respond to EGFR [epidermal growth factor receptor] therapy, while patients who are RAS wild type tend to respond to EGFR therapy. There are some questions about whether or not we need to test for RAS in a right-sided tumor. In the first-line setting, based on the CALGB/SWOG 80405 study and the FIRE-3 study and the PEAK study, there is clearly no benefit to adding an EGFR drug in a patient who is RAS wild type with a right-sided tumor. However, in the second-line or third-line setting, it’s quite unclear. If you look at the data in the second-line setting, adding an EGFR drug for a right-sided tumor may have no benefit. However, there was a trial that randomized to cetuximab versus best supportive care in the third-line setting.

They went back and looked at the sidedness—right-sidedness versus left-sidedness. In that study, they found that for right-sided tumors, there was no benefit in progression-free survival, in terms of adding the EGFR therapy. However, there was a slight improvement in overall survival. Therefore, based on these conclusions, I would say that you should still test for RAS status in the right-sided tumor and possibly use the drug in the third-line or fourth-line setting if you have no other options left.

In colon cancer, BRAF testing should be done at the beginning as well. There are a couple of implications for BRAF. If you have BRAF-mutant disease, you tend to have a much poorer prognosis and be resistant to EGFR therapy. Therefore, if you know that your patient has BRAF-mutant disease at the beginning, you want to be more aggressive because the median survival is only 18 to 20 months. In those cases, you would want to start with a regimen called FOLFIRINOX [folinic acid, fluorouracil, irinotecan, and oxaliplatin]. The second thing is that there are now trials targeting BRAF-mutant tumors. Therefore, if you know the patient has a BRAF-mutant tumor, you can look out for a trial and enroll them in a study.

In terms of HER2 amplification, it is known that about 2% to 3% of patients with advanced colorectal cancer are HER2 amplified. If you look at patients who are RAS wild type, it’s about 5%. The significance of HER2 amplification is 2-fold. Patients who have HER2 amplification and are RAS wild type are also resistant to EGFR therapy. Therefore, those patients should not get an EGFR drug despite having RAS wild-type disease. Secondly, there are now drugs that target HER2-driven colon cancer. For example, there’s a trial combining trastuzumab and lapatinib that shows a response rate of 30%. Therefore, if you know the patient is HER2 positive, you probably should not use EGFR therapy. You should probably look for a trial that targets HER2/neu or possibly get a drug off label for use in the patient.

Other than RAS, BRAF, and HER2/neu, there are biomarkers for immunotherapy that are well known. The MMR [mismatch repair] or MSI status must be tested at the beginning because those will predict who will respond to immunotherapy checkpoint inhibitors such as pembrolizumab and nivolumab. The incidence of MMR deficiency or MSI-high status in colon cancer is about 5%, yet we test all patients. We should test all patients with MMR because those patients will be good candidates for checkpoint inhibitors. They may get a durable response from this therapy.

Other than MMR, there are other biomarkers out there that we could also test for, such as TRK fusions. TRK fusions occur only about 1% of the time, but this is very important because there are drugs that target TRK fusions. Data show that if you have a TRK fusion, the response rate could be up to 75% if you use certain drugs. Therefore, even though they are very rare, TRK fusions are potentially life-changing events for patients, in terms of treatment.

There are others, such as the RET mutation and ALK fusions. Trials are a bit preliminary—small studies—but once again, similar to TRK fusions, if you have those fusions, because there are drugs that target those areas, you can get very durable responses out of therapy if you can find a trial or possibly get the therapy off label.

Transcript Edited for Clarity