Expert Highlights Precision Medicine Advances in GI Cancers

In Partnership With:

Partner | Cancer Centers | <b>UCSF Helen Diller Family Comprehensive Cancer Center</b>

Eric A. Collisson, MD, discusses ongoing efforts to advance precision medicine in GI malignancies to improve the standard of care.

Eric A. Collisson, MD

The availability of targeted agents and immunotherapy in gastrointestinal (GI) cancers is enabling physicians to address more molecular aberrations, such as high mutation rates and defects in homologous recombination, says Eric A. Collisson, MD.

Collisson noted, however, that while tremendous progress has been made, much remains unknown about optimizing genomic testing and implementing precision medicine in GI malignancies.

“To what depth every tumor should be examined and whether patients should undergo testing upfront or after they’ve failed frontline treatment has yet to be determined,” said Collisson.

OncLive: What is the state of precision medicine in GI cancers?

What are some trials that have highlighted approaches in precision medicine?

Could KRAS soon become a standard target?

In an interview during the 2018 OncLive® State of the Science Summit™ on Gastrointestinal Cancers, Collisson, associate professor, Department of Medicine, University of California, San Francisco (UCSF) Helen Diller Comprehensive Cancer Center, discussed ongoing efforts to advance precision medicine in GI malignancies to improve the standard of care.Collisson: At UCSF, our approach is to look at the annotation of GI malignancies. The focus is on precision medicine and targeting therapies to specific mutations when possible, mutational signatures, and other genomic attributes if not. GI cancers have been a little behind in terms of precision medicine. The field of medical oncology has had success in very narrow areas in melanoma and lung cancer. GI cancers haven't focused as much on individual mutations as in those malignancies, but we’re catching up now. We have more drugs than we had before, and we understand that we can move beyond single mutations in EGFR and BRAF to come up with therapeutic strategies.There have been some important foundational and deeper work profiling GI cancers in the clinic. We’re learning that though the primary tumor can be easily accessed with an endoscope, the metastatic site isn’t always perfectly reflected. Clinical trials are now being designed to look at metastases closer. The role of germline genetics in pancreatic cancer and other GI malignancies is becoming extremely clear because of breast and ovarian cancer data. Physicians are learning how to look for germline and somatic defects. What’s clear is that physicians have the technology and a larger armamentarium of drugs to combat the disease than they did 2 years ago.That is an area that is extremely exciting. Approximately 2 or 3 years ago, we had a couple of things downstream of KRAS that we thought were going to be great. However, they didn't show the efficacy we wanted. I don't think anyone thought of developing a KRAS-targeted drug in their lifetime. Now we have a couple of tangible pieces of chemical matter that stick to KRAS in certain mutant forms. We're getting our first signs of efficacy. Unfortunately, these mutations are not as common in GI cancers as in lung cancers and other smoking-related cancers. It does, however, break the negative hex we have had on mutations that are not druggable. That's a major intellectual barrier, if not biochemical barrier, that we’re clearly thinking about.

Please discuss immunotherapy and the treatment of microsatellite instability-high (MSI-H) tumors.

Other exciting research is looking downstream of KRAS at RAF, MEK, and ERK proteins. We had one MEK inhibitor for a long time; that was underwhelming. We’re looking at how to push these new targeted agents to the limit. Managing tolerability has proven to be difficult because of the challenges with the therapeutic index. Nonetheless, we have a rich chest of biochemistry to work on here. The immunotherapy revolution has not overtaken GI cancers, but we are still looking for its optimal use. We all know that MSI-H are the easiest to treat; they are the “slam dunk” tumors. What we don't understand is how rare those are in the metastatic setting. A lot of our colleagues in the community are disappointed when they don’t find any of these MSI-H tumors in the metastatic setting, despite seeing quite a few of them in the resectable setting.

That's a powerful testament to how prognostic these markers are. These MSI-H tumors don't metastasize whether they are found in colorectal, gastric, or possibly other cancers in the GI tract. When we find them, we know immunotherapy is the right treatment; however, we haven't been able to expand its use with few exceptions in gastric cancer and hepatocellular carcinoma.

We don't have clear directions on how to expand its use. Should we be using immune therapy with chemotherapy like some of our colleagues in lung cancer? Should we be combining it with other checkpoint inhibitors like some of our colleagues in melanoma and other diseases? A mix of molecular profiling beyond MSI is clearly going to tell us about neoepitopes and immunologically “hot” tumors; these are tumors we can concentrate immunotherapy approaches on. That's more of an orthogonal direction than targeting individual oncogenes. Hopefully, the result will be more options for patients.