Bekaii-Saab Brings Insight to Precision Medicine in GI Cancers

Tanios Bekaii-Saab, MD

A precision medicine approach in oncology is entering nationwide clinics and across various tumor types, but many unanswered questions still remain before this strategy can apply to gastrointestinal (GI) cancer treatment, said Tanios Bekaii-Saab, MD.

For example, many patients with colorectal cancer (CRC) remain limited to traditional chemotherapy regimens, aside from a small population of those whose tumors are microsatellite instability-high (MSI-H). This rare subset can derive significant benefit from checkpoint inhibitors.

Beyond immunotherapy, targetable alterations and corresponding agents are emerging.

For NTRK-mutated tumors, larotrectinib (Vitrakvi) was granted a tumor-agnostic accelerated FDA approval in November 2018. The decision was based on findings from patients with NTRK fusion—positive tumors enrolled across 3 clinical trials and multiple disease types.

“The main challenge is how we bring these agents into the clinic,” said Bekaii-Saab, a professor of medicine at Mayo Clinic. “Today, in GI cancers, there is a point of entry into a study with 2, 3, or more nonspecific histology-agnostic platforms.”

OncLive: How has precision medicine impacted the field of GI cancers?

In an interview with OncLive, Bekaii-Saab discussed ways to further develop precision medicine within the treatment of patients with GI cancers.Bekaii-Saab: Across cancer medicine overall, we are making strides with precision medicine—understanding the targets and bringing more targeted therapies into the mix. The concept is that we hit the cancer cells and spare the normal tissue. However, we are still a long way from that because we are still hitting "innocent bystanders" and normal organs when we treat patients.

We don't have a lot of opportunities in the clinic yet [for precision medicine in GI cancers], unlike lung cancer or some of the hematologic malignancies. The good news is that the discovery process is accelerating quite significantly. We have a number of promising targets we can go after and a number of agents that are making headway.

Other agents have been making their way into the clinic through single trials. FGFR inhibitors in cholangiocarcinoma are a perfect example; there are 2 or 3 of them that seem very promising. In CRC, we only have a few targets that we can match with corresponding therapies. One of them is MSI-H with PD-1 inhibitors, and the other is NTRK fusions with NTRK inhibitors. These are examples of histology-agnostic alterations and these are very rare.

When we look at the others that we are starting to learn about, such as HER2, BRAF, ROS1, and MET amplification, how do we develop studies specific to these? Through one of our research platforms at Mayo Clinic, which actually involves over 120 centers, we started a CRC specific platform. We think it is very innovative because it helps to clear out the challenges we typically face.

We allow 2 points of entry: liquid and tissue biopsies. All patients will have both, so we will also learn how these 2 modalities relate. We have multiple arms already up and running or approved for funding. For example, we have an EGFR arm; this is specific to CRC. We also choose endpoints based on how rare or common the patient's targetable alteration is. Response rate is now an accepted endpoint to allow for an accelerated approval in these very rare mutations. Essentially, we are trying to make this as meaningful as possible. This allows us to create platforms that are more disease-specific. We are also working on a platform specific to pancreatic cancer and other GI cancers.

What are some of the challenges with these agents?

There is a lot of activity in this area, and the hope is that all types of GI cancer will be broken up into multiple diseases that have specific drivers. It may not be 100% of them, but if we can hit 30% or 40%, that is a great achievement.With all of these drugs, although they seem to bind to the target very well, we are still seeing a lot of resistance. This is either from the get-go—what we call acquired resistance, where the patient simply does not respond to the therapy—or secondary resistance, meaning the patients eventually progress.

Again, we have a target; we have an agent that hits the target, but we still may not see responses. We are trying to better understand this. We do understand that cancer may have multiple drivers. Secondly, we know that when cancer becomes resistant to an agent, there are other drivers that may have moved to become the dominant driver.

Traditionally, one of the problems we face is that we have to biopsy tumors. However, to do so, you have to put patients through the cost and the invasiveness of a tissue biopsy. How many times can you do this? When we are talking about patients in the clinic, you can't continue to do that. Liquid biopsies look at circulating-tumor DNA.

What is the biggest take-home message from this?

In our study that I mentioned earlier, we are doing liquid biopsies at time of diagnosis and time of progression to determine what clones are being suppressed and what ones are taking over. The other way to curb this is to combine targeted agents together, but we need to learn about the biology. We are on track to do this, but we are not there yet.For GI cancers, it is important to keep in mind that change is coming. We have a lot of agents and a lot more understanding of the different alterations and drivers. We also have studies looking at specific diseases. This is something we are very passionate about. For community oncologists, it is important to consider screening patients and doing genomic analyses. A lot of platforms are commercially available. Then, look for studies and find the right match for patients.