Collisson Covers Cholangiocarcinoma and Colon Cancer Challenges

Eric A. Collisson, MD

Heterogeneity, diagnostics, and molecular profiling are examples of the clinical challenges in treating patients with cholangiocarcinoma, explained Eric A. Collisson, MD, but targeting FGFR2 and IDH1 could be a potential approach in improving outcomes.

“One of the themes in gastrointestinal (GI) oncology is the specialized development of these therapeutics,” said Collisson. “There are some specific challenges that each one of these different diseases hold in getting drugs approved and using them safely and effectively.”

There are diagnostic challenges in this landscape, too. Cholangiocarcinoma, a disease of the biliary system in the liver, can often be misdiagnosed as a liver metastasis, added Collisson.

Moreover, in colon cancer, there continues to obstacles in identifying the small subset of patients with microsatellite instability (MSI)-high tumors who are likely to respond to PD-1 inhibition. Researchers are continuing to investigate therapies against RAS, KRAS, NRAS, and HER2 mutations, as well.

OncLive: You presented on cholangiocarcinoma and colon cancer at this meeting. What is known now about cholangiocarcinoma and what are the unanswered questions?

Can you expand on some of the therapies that are being investigated?

There seems to be less research emerging with cholangiocarcinoma than in other GI cancers. Are there challenges with diagnosing this disease?

In an interview during the 2018 OncLive^® State of the Science Summit^TM on Gastrointestinal Cancers, Collisson, an associate professor at the University of California, San Francisco Helen Diller Comprehensive Cancer Center, spoke on the challenges and emerging advancements in cholangiocarcinoma and colon cancer.Collisson: There are lots of unanswered questions in cholangiocarcinoma. It is a very heterogeneous disease in that the forms that take root inside the liver appear to be genomically and therapeutically very different than those that initiate outside the liver. [This includes] the gallbladder, where some of these cells look similar under the microscope but they clearly have distinct origins; the way they transform also appears to be pretty distinct. We are just now catching up and finding some therapies for small subsets, which can be really hard to do in a rare disease. It does take a village and it takes concentration, but progress is being made in that area. We discussed some of the advances; the FGFR2 receptor is a kinase that is rearranged, mutated, and amplified in different fractions of prevalence in intrahepatic cholangiocarcinoma. We have also discussed the theoretic rationale for targeting tumors with mutations in IDH1, a gene that is interesting and [creates] somewhat strange bedfellows with liver tumors, cholangiocarcinomas, brain cancers, and leukemias. This is sort of the poster child for a genotype, not anatomy. These systems are completely different in the body, yet they seem to have a predilection for this metabolic mutation with some therapeutic success, so we spoke about those trials, as well. There is lots to learn. It is, diagnostically, a real challenge. Most oncologists, clinicians, and general practitioners are familiar with tumors in the liver that are detected radiographically. The correct assumption is that [it is] a metastasis most of the time—except when it isn’t.

Regarding colon cancer, what is imperative to highlight in that area?

It can be hard to figure out—to essentially rule out carcinoma from another source in the body that has traveled to the liver versus what can be a diagnosis. [We need to] exclude all of these other places it could have come from and [identify that] the origin is in the biliary system. That is increasingly done by sophisticated pathology approaches; the problem is that cells are quite infrequent. We have a hard time getting them; there is no equivalent of a colonoscope, mammogram, or an ultrasound-guided biopsy for the liver. Therefore, pathologists are challenged with very few cells. How to diagnose a disease efficiently and leave us adequate tissue for other analyses, like for FGFR2, is a real rationing strategy. We want to be efficient, accurate, and also save some for later. Unfortunately, colon cancer has not gotten as much in the forms of new drugs or new avenues in the last couple of years as some of these other disease entities. However, we reviewed some of the distinctions and whether anatomy can trump genotype. We discussed the distribution of side of the primary [tumor] versus the harboring of a RAS, KRAS, or NRAS mutation in the tumor and how that is changing the way we think about genotype driving everything.

What are your thoughts on the priority review designation with nivolumab (Opdivo) and ipilimumab (Yervoy) in colorectal cancer?

Then, we discussed the role of MSI testing, the emerging role of testing for HER2, and other potentially drivable oncogenes and where those might lead us in colorectal cancer. The easiest take-home message for the treating oncologist today is that MSI testing is an important part of the therapeutic workup. We knew the prognostic implication and the potential to screen for Lynch syndrome in the family, but now there is a therapeutic modality that sometimes gets lost in the wash when we're working up a patient and focusing on KRAS and NRAS in the first-line treatment decision making.

Looking to the next 5 years in these landscapes, what does treatment look like to you?

We are continuing to focus on getting that piece of knowledge. When medical oncologists do their MSI testing in later metastatic disease, as the National Cancer Institute recommends, they are seeing low single-digit responses. Many oncologists have never seen [an MSI-high tumor] if they don’t see a lot of patients with CRC. That is normal; nothing is wrong with the test there. It is still working fine. It’s just that we know there is such strong prognostic data in that test that the positives rarely metastasize. However, when they do, [these patients] now have a treatment option. The therapeutic landscape is expanding a great deal in that we are finding a “sweet spot” after a lot of confusion between molecular profiling and the emerging importance of immune profiling. All these things about how to use immunotherapy are all somewhat proxy measures of the actual immunity we're looking for with high tumor mutational burden and MSI. These aren’t measuring immune efficacy directly, they are measuring what the immune system may recognize. We are going to make a lot of advances in understanding the T-cell or other immune-cell repertoire that does damage on the tumor, and that will be an exciting diagnostic advance.

Within the targeted therapy world, we're kind of in “Generation 2.0” and learning the FGFR2 and cetuximab (Erbitux) stories. It’s clear that acquired resistance is the rule, not the exception and we will start to build in treatment maneuvers that anticipate resistance and prevent it. That may not increase the responsive subset, but it may prolong that responsiveness into years instead of months, which is something everyone is aligned with—patients, industry, and oncologists.