Evolution of Predictive Biomarkers for Metastatic CRC - Episode 1
Transcript: Richard Kim, MD: I think leading up to 2020, we made a lot of progress in colon cancer in terms of personalized medicine. I think we are in the era of what you call a molecular revolution. In 2020, all the patients with diagnosed stage IV colorectal cancer should get NGS [next-generation sequencing] or molecular profiling done at the beginning, especially the stage IV patients. Now there is always controversy about stage II and III patients. There are certain biomarkers that have to be tested, such as MSI [microsatellite instability] status or MMR [mismatch repair] status, because those would help dictate, especially in stage II patients, whether they should get adjuvant chemotherapy. However, regarding other profiling in stage II and III patients, we don’t know exactly what to do with them. So other than a trial basis, I would recommend at minimum testing MSI status, MSS [microsatellite stable] status, MMR status, and BRAF status for Lynch syndrome.
However, in stage IV patients, it is a little bit different. We have many actionable mutations that exist in colon cancer that could help us personalize treatment for those patients. For example, the KRAS mutation, the BRAF mutation, MSI status, and HER2 status. That is some of the profiling that’s available that we can potentially use at the beginning to dictate the treatment of the patients.
Now some of the challenges that we face with this are severalfold. Number 1 is that ideally when we start somebody with chemotherapy as a first-line treatment, we would like to have this profiling available at the beginning. However, unfortunately, most of the time the tissues are out for testing. By the time we get the tissue and we send it out for NGS profiling, it takes about 4 to 6 weeks, and most of the patients do not want to wait 4 to 6 weeks to get treated. Therefore, in those settings we would start a patient on chemotherapy without the profiling available. That’s one of the challenges we have to overcome.
The second thing is that there are several actionable mutations. However, if you look at the whole profiling panel, it’s usually anywhere from a 200- to 250-gene panel. Most are not actionable, but those can pick up some of the rare fusions. The question is at the beginning, should we do what we call selective testing versus the whole panel testing? And that’s still controversial at this time.
Personally, I would like to get a whole panel because when you use a selective panel, I may run out of tissue at the end when they fail first-line therapy. Therefore, ideally, I like to get a whole panel at the beginning, and this is where I think a liquid biopsy will come into play in the future because it’s much easier to get, easily accessible, and the turnaround time with liquid biopsy is only 7 to 10 days.
In colon cancer, there are many biomarkers that are prognostic and that are predictive. Recently based on the CALGB/SWOG 80405 study, there was a full mutation analysis that was presented last year. Now when you look at those data, some of the facts we knew already. What are the prognostic markers in colon cancer? In that study the patients with BRAF-mutated tumors had overall survival of up to 12 to 13 months. So we know that BRAF is a poor prognostic marker. They also in that study found that the patients with RAS mutations also had a poor prognosis. Their overall survival is only 25 to 27 months. So the patients with the best phenotype are what we call a triple wild type: KRAS wild type, NRAS wild type, and BRAF wild type. Those are the patients who really had a very good outcome.
And in that study they also looked at tumor mutational burden, what we call TMB. This is the first study that looked at this as a prognostic marker, and what they found is that patients with high TMB had a better prognosis. This makes sense because patients with high TMB colon cancer tend to have high levels of neoantigens, which means that there’s more tumor-infiltrating lymphocytes, which means that their immune system is intact and able to control the cancer in a much better way.
Now in terms of the predictive biomarkers, we know there are many predictive biomarkers that we use, such as the KRAS status, the BRAF status, and the HER2 status, that we use for immunotherapy. We do use MSI-high and MMR status for it. And very rarely we do use TRK fusions as well because now there are drugs that are able to target those. In that study patients who were MSI-high, interestingly enough, responded better on the bevacizumab arm compared to the cetuximab arm. And that makes sense because most of the tumors that are MSI-high are the right-sided tumors, and we know the right-sided tumors tend not to respond to EGFR inhibitors.
Transcript Edited for Clarity