Biomarker Testing and Precision Medicine in mCRC

Tanios S. Bekaii-Saab, MD: Moving on to novel approaches for the treatment of metastatic colorectal cancer. I'm going to start with John. ASCO [American Society of Clinical Oncology] selected molecular profiling driving progress in GI cancer as the 2021 Advance of the Year. Thinking about your patients and all of our patients, what biomarkers testing do you need to have at the minimum available, but what is the preferred approach?

John H. Strickler, MD: Just rewinding the clock it's amazing how much change there has been in the area over the last decade. Ten years ago, 2011, there was only 1 biomarker we needed to test in the metastatic setting, it was KRAS. Now we think about where we are just 10 years later and it is a radically different environment. All patients who come in the door with a new diagnosis of a metastatic disease need KRAS and NRAS testing. In addition to BRAF 3600E and microsatellite testing, MSI [microsatellite instability] testing through either EGS, PCR [polymerase chain reaction], or immunohistochemistry [IHC].

Additionally, HER2 testing is important. Particularly for our patients with RAS wild-type disease because we have a number of anti-HER2 therapies available either through clinical trial or listed in NCCN guidelines. It has been an erratically changed landscape for us, but additionally, there are other targets that are rare that are also actionable. Things like NTRK [neurotrophic tyrosine receptor kinase], tumor mutational burden, and others. We've reached a critical threshold now where we can't just order PCR individually for all these different targets. We need one assay to test everything simultaneously and next-generation sequencing is the right approach, given the sheer number of targets that we have.

Tanios S. Bekaii-Saab, MD: Raghav, you've done some nice work with HER2 amplifications as predictors of resistance to anti-EGFR [estimated glomerular filtration rate]. When do you want to see that HER2 complication tested? Is it before you see the patient? Do you want to ensure that you have that HER2 available on that? Is it after the first line? How do you approach this?

Kanwal Raghav, MD, MBBS: I agree with John. The success of HER2 is critical in this disease. Therefore, if you're doing NGS [next-generation sequencing] testing, I would like to see that being done at first-line therapy. Preferably any time before you've treated a patient with anti-EGFR for sure. Earlier the better because HER2 therapies are well tolerated and they can have very prolonged responses. If you could get these patients early on into clinical trials or early onto HER2 therapies, you're not struggling with giving patients a lot of toxicities from chemotherapy.

We had one of the patients that came to us on FOLFOX [folinic acid, fluorouracil, oxaliplatin] for 1 1/2 years, now has disabling neuropathy, and was hard to amplify it. This is a patient where even if you had a little bit of progression on maintaining therapy you could've gone to a HER2 therapy to begin with. I would like to see all of those done at diagnosis.

Tanios S. Bekaii-Saab, MD: Joleen, all of us here made sure that MSI-H [high] is universally included now as part of the pathology report. How about HER2 and?

Joleen M. Hubbard, MD: The standardization of HER2 IHC for breast cancer and gastric cancer has made it easy for us to do this also in colorectal cancer. I will order that, but a lot of our NGS panels will also have ERBB2 [Erb-B2 receptor tyrosine kinase 2], so we can look for any alterations on the NGS panels as well. It's not always necessary. I like to get the results of HER2 whether it's NGS or IHC right at the beginning. I do that because I want to be able to plan out my treatment strategies and make sure that I have this patient flagged for a HER2-directed trial.

Tanios S. Bekaii-Saab, MD: John, we have our usual implications and mutations on circulating tumor DNA. What do you want to know about MSI-H, tumor mutation burden [TMB], how do we interpret those? In what context do we use them in the clinic if we do not have access to the tissue?

John H. Strickler, MD: We are now able to detect microsatellite instability with our commercially available blood-based assays. That's beneficial particularly in patients who maybe don't have tissue available or we’re looking at substantial delays to do that testing. It's an important screen when we don't have that tissue. The second part of your question is around tumor mutational burden. That's an interesting area and a rapidly evolving treatment landscape. We do have FDA-approved immunotherapy, pembrolizumab, for patients who have a tumor mutational burden in tissue greater than 10. The appropriate cutoff for a blood-based tumor mutational burden remains unclear. I'd like to see that data developed in the years to come so that we understand what the appropriate cut-off is for that particular type of test.

It should be said that the FDA approval for patients with tumor mutational burden-high disease did not include patients with colorectal cancer. It's unclear whether the ideal cutoff for a patient with TMB-high metastatic colorectal cancer is 10. There was some data presented recently from the ASCO TAPUR Study presented at ASCO GI looking at the performance of pembrolizumab specifically in patients with TMB-high disease. It did meet its primary efficacy end point of disease control, but barely. It may be the case that if you ultra-select these patients to a higher tumor mutational burden cutoff granite we'd have fewer patients treated, but we might see a greater effect size.

Tanios S. Bekaii-Saab, MD: In other words, for the circulating tumor DNA the cutoff is not what it was established we don't understand it. We don't understand what that line draws, what 50 means versus at least 10 to 15 in the tissue, and more work needs to be done. MSI-high, it's more clear-cut and we can rely on the liquid biopsy?

John Strickler, MD: Absolutely, it appears that the tumor mutational burden also in blood evolves over time. That will add an extra layer of complexity to making those types of management decisions for now.

Tanios S. Bekaii-Saab, MD: When you check it will look different, which makes sense if your later lines have a higher TMB versus treatment naïve patients. A lot more to learn there, we're not there yet.

Transcript Edited for Clarity