Genomic Testing and ctDNA Add to Personalized Treatment Approach in Metastatic CRC

The benefit observed with immunotherapy vs chemotherapy in patients with mismatch repair–deficient (dMMR)/microsatellite instability–high (MSI-H) colorectal cancer (CRC) underscores the importance of up-front genetic testing in this patient population, according to Benjamin Adam Weinberg, MD.

“Without knowing that information [on dMMR/MSI-H status], you are not [getting] an adequate perspective to choose if a patient should get immunotherapy or chemotherapy in the metastatic setting,” said Weinberg, who presented data on current treatment strategies for patients with CRC at the 4th Annual Precision Medicine Symposium: An Illustrated Tumor Board.

MSI/dMMR status is not the only marker to look for in patients with CRC. Groundwork laid by researchers in breast cancer, melanoma, and other malignancies have demonstrated the importance of testing for HER2 and BRAF, respectively, to match patients to available targeted therapies.

In an interview with OncLive^®, Weinberg discussed the expanding role of circulating tumor DNA (ctDNA) in CRC and the various testing methods, the utilization of liquid biopsies, and the need for up-front genetic testing to identify molecular alterations. Weinberg is an associate professor of medicine and an attending physician in Gastrointestinal Medical Oncology and Sarcoma at MedStar Health.

OncLive^®: How do you utilize genetic testing when treating patients with CRC?

Weinberg: CRC is being further defined genomically in early stage disease and in more advanced disease. We are now more commonly using ctDNA as a test for minimal residual disease [MRD] for patients with early-stage disease, [particularly] stage II or III CRC. We hope that in the future, [ctDNA] findings can be used to guide adjuvant treatment decisions.

We believe that the best way of [measuring ctDNA] is taking a patient's own tumor tissue and using a tumor-informed assay to make a personalized MRD test for each individual patient, using their genomic signature.

Additionally, in the more advanced setting, we have been sending tumors for broad molecular profiling to try to find novel treatment options and to better select patients based on their mutational profile. We now have more targeted therapies for unique subsets of patients if their tumor expresses a specific mutation or amplification.

In both the early and later-stage settings, we are using genomics across the board for patients with CRC.

How has recent research in CRC impacted treatment decisions?

In the more advanced setting, we need to know if a patient has a dMMR or MSI-H tumor. This [accounts] for only about 4% of patients with metastatic disease. However, if a patient is in this rarefied group, they will benefit from immunotherapy as opposed to chemotherapy. Typically, this includes PD-1 inhibitors, such as pembrolizumab [Keytruda] or nivolumab [Opdivo], [the latter of which] can also be given in combination with anti–CTLA-4 drugs like ipilimumab [Yervoy].

Those combinations or single-agent immunotherapy beat chemotherapy for patients with this immune-sensitive [dMMR or MSI-H] subtype. We are probably hurting these patients with chemotherapy, which is generally thought to be more toxic. This is one group that you need to be aware of up front, especially in the metastatic setting, whether they are dMMR or MSI-H.

Other targets in CRC include BRAF V600E mutations. We have borrowed from BRAF-[targeted therapies] from our colleagues in melanoma, which have high rates of BRAF. In advanced CRC, [BRAF mutations] occur in about 10% or 11% of patients. We have FDA-approved drugs in the second- and third-line setting, combining BRAF-targeted therapy with anti-EGFR therapy.

Similarly, we have borrowed HER2-[targeted therapies] from our colleagues in breast cancer. There are active HER2 targets in CRC, though they are relatively rare, [occurring] in about 5% of patients.

There are even rarer subsets, such as NTRK fusions, which have [targeted therapies with] tumor-agnostic approvals across different disease types. The more targets we find, the more we need to do broad genomic sequencing to find these rare subtypes.

What is the significance of assessing ctDNA in CRC?

When we talk about ctDNA, we often mean 1 of 2 related yet different things. One is in the advanced setting of using a liquid biopsy. If we can't get tissue or do not want to wait for a broad genomic tissue-based test, we can send liquid tests, such as the Guardant360® CDx assay. [Assays from] different companies have the same idea, where you send a liquid or blood test, and you get whatever DNA is being shed into the circulation that can be identified as tumor specific. [Liquid biopsies are] easy, quick, and relatively cheap. However, if although you may not find [ctDNA] in the [blood], it does not mean it is not in the tumor.

[Liquid biopsies] also overcome concerns with tissue or tumor heterogeneity, where you are getting a global snapshot of ctDNA across different tumor sites in the body. Whereas, if you're sticking a needle in one site of a tumor, you may not be seeing what is going on globally. There are some potential advantages to [measuring] ctDNA with a liquid biopsy.

However, often when we talk about ctDNA, we are talking about minimal residual disease [MRD] testing in the more localized setting and the potential curative setting. Does a patient have residual, microscopically detectable, abnormal ctDNA in their bloodstream? Some tests are tumor informed, where we take a piece of the tumor, send it for broad genetic sequencing, and select a cohort or panel of genes for that individual patient.

There are also tumor-uninformed assays, where we do not use the tissue. [Instead], we use a blood test to try to see if there is any evidence of MRD in the body.

There are a couple of different approaches [for evaluating ctDNA]. When we talk about certain terms like ctDNA, sometimes we are talking about MRD testing, and sometimes we are talking about liquid biopsies in the advanced setting. They are related, but still a bit different.

What else must clinicians be aware of when treating patients with metastatic CRC?

We must be cognizant of our patients’ mutational profiles, especially in the advanced settings. It is not appropriate to be unaware of a patient’s dMMR or MSI-H status. If tissue is not available, this can be detected on a liquid biopsy, which may potentially be more sensitive than a tissue [biopsy] in some instances.

In the later-line settings, you will need to know a patient’s KRAS/NRAS [mutation] status if you are trying to select if a patient should get anti-EGFR therapy, especially for left-sided tumors in the metastatic CRC setting. We need to know the BRAF status because we have approved BRAF-[targeted] drugs that beat chemotherapy in the second- and third-line setting.

We need to find these rare NTRK fusions. We need to know if a patient’s tumor is HER2-amplified, because there are active drugs in that space. We need to be doing testing up-front if there is tissue available for patients with metastatic disease because it may inform [more than] their first-line treatment. It may inform treatment in the second line, third line, and beyond in terms of these rarer subgroups.

Discovering a patient’s KRAS status [is not enough]. We need to know which KRAS mutation it is. We need to know if there are other rare co-occurring things with mutations that we may not have thought of. Now, we must do broad genetic testing up front.