Understanding the Implications of Mutational Status in Intrahepatic Cholangiocarcinoma

Jesper B. Andersen, MSc, PhD

There continues to be a dramatic increase in both incidence and mortality rate of cholangiocarcinoma worldwide, according to Jesper B. Andersen, MSc, PhD.

In a study recently published in Hepatology from the Andersen Group at Biotech Research and Innovation Center, it was shown that using a single-gene dissection approach in patients with intrahepatic cholangiocarcinoma can characterize divergent cancer programs and drug vulnerabilities. Through this, therapeutic sensitives for this patient population can be identified, aiding in the process of targeted genotyping.

This study evaluates the genomic, epigenetic, and pharmacologic landscapes of intrahepatic cholangiocarcinoma, which is both molecularly heterogeneous and resistant to chemotherapy.

"There is a clinical need to highlight these patients,” said Andersen in an interview with OncLive. “This is not only a question of highlighting our study. This disease is a rare malignancy that has progressed since the 1980s. Whereas the therapeutic options have improved for many other diseases, the clinical opportunities for most patients diagnosed with cholangiocarcinoma are limited—this is a dismal disease.”

The appropriate stratification of patients with intrahepatic cholangiocarcinoma comes with implications for the development of precision medicine in this disease, Andersen adds. This could be groundbreaking, as there are no approved therapies for this patient population, and only 10% to 30% of patients are eligible for curative surgery.

OncLive: Can you please provide an overview of this study?

During an interview with OncLive, Andersen, who is an associate professor and leader of the Andersen Group at Biotech Research and Innovation Centre, Department of Health and Medical Sciences at the University of Copenhagen, discussed the findings of this study and the implications it could have on the future treatment of patients with cholangiocarcinoma.Andersen: The study is focused on tumors that arise in the bile duct epithelium within the liver, meaning they are anatomically classified as intrahepatic cholangiocarcinomas. That was a deliberate strict inclusion of patients into this study to introduce minimal genomic heterogeneity, because cholangiocarcinoma is a tumor type that exists both in the liver and right outside the liver (extrahepatic). We used a total cohort size of 496 intrahepatic tumors, and analyzed these by whole-exome sequencing, targeted-exome sequencing, transcriptomics, structural variances, and DNA methylation.

We have defined a novel strategy that stratified the patient based on the recurrent genetic alterations that occur and give rise to unique or distinct regulatory networks between groups. We did that by stratifying the patient based on 3 driver-gene groupings—IDH1/2, KRAS, and TP53—that led to the discovery of a fourth group that we termed “undetermined.” This means we did not know the cause of the underlying driver of this disease subgroup. As such, the undetermined group is wild-type for the 3 driver genes. Those could be distinctly classified based on distinct mutational signatures, structural alterations and DNA-methylation profiles.

This classification scheme led to a stratification of unique pathway enrichment for each of these 4 patient groups who were associated with pharmacogenomic and -epigenetic signatures, clinical significance, and specific therapeutic response when we tested patient-matched cell lines in a large-scale drug screening of 525 different drugs that are late-stage or FDA approved.

The undetermined group became quite interesting in the sense that we didn't know what was causing that group. There, we found an enrichment of FGFR2 gene fusions and amplification in the chromosomal focal region of the gene (METTL13). This gene is a putative novel methyltransferase with very little information in the literature (metyltransferase-like 13). We are continuing to identify the role of METTL13 in general and its oncogenic potential in patients with intrahepatic cholangiocarcinoma.

What are the next steps moving forward?

In conclusion, our study has led to elucidating 4 different patient groups —IDH1/2, KRAS, TP53, and undetermined—based on different genomic and epigenetic signatures, resulting in unique pathways enrichment and specific drug sensitivity with putative therapeutic application to treat these patient subgroups. The IDH group indicated a specific enrichment and potential for drugs that would target metabolic modifiers. The KRAS group indicated a role for microtubule modulators and potential for immune therapy, TP53 indicated potential for topoisomerase inhibitors, and the undetermined group was only highlighted by inhibitors of the mTOR pathway. Theoretically, there are 2 potential next steps. In this study, we matched based on the genetic substitutions (the different mutational profiles) of these patients the best-fitting cell lines for the drug screening in our library.

The next step would be either in vivo modeling or directly test in the clinic these drug families for efficacy in the 4 enriched patient subsets. The problem with in vivo testing is that there currently are no animal models specifically for each of the 4 different groups. Of course, one interesting approach could be to generate patient-derived xenografts (PDXs) of each of these 4 groups. This challenging in a rare malignancy.

Are there any roadblocks that you foresee moving forward?

What are the main takeaways from this study for oncologists?

Alternate, this approach could be tested in the clinic since the drugs are approved for other indications. This means that we would do genotyping of all the patients coming into the clinic, then associate them with one of the 4 groups, and then give the drugs that are already approved for clinical use. That is the advantage of precision medicine. At the end of the day, we are using a drug library not of investigative drugs, but FDA-approved drugs or drugs in late stages of clinical trials. These drugs are already tested and safe to use in the clinic, but they are not approved for cholangiocarcinoma (approach is termed drug repositioning). The main roadblock would be that this is a rare disease. It is challenging to acquire enough fresh tissue from patients to generate PDXs that would match the 4 different groups. Going straight to a clinical trial outlining these 4 different groups, and getting approval for testing the drug families in each of their assigned patient group would require a longer enrollment, and would likely need the involvement of multiple institutions.The main message is that we need to subgroup the patients to see the benefit of drug treatment. There are possibilities to uniquely treat these patients rather than treating them as one group. There is a need for precision medicine for these patients. Do not treat cholangiocarcinoma patients as one.

Nepal C, O’Rourke, CJ, Oliveria D, et al. Genomic perturbations reveal distinct regulatory networks in intrahepatic cholangiocarcinoma [published online December 26, 2017]. Hepatology. doi: 10.1002/hep.29764.