Progress in cholangiocarcinoma treatment was long stalled, but now scientists are identifying genetic drivers likely to respond to novel drugs.
A new understanding of the genetic mutations that drive the rare and difficult-to-treat cancer cholangiocarcinoma, along with the testing of targeted drugs and immunotherapies in people with the disease, are laying the groundwork for long-awaited advances in therapy.
Cholangiocarcinomas are tumors that arise in the bile ducts — tubes from the liver to the small intestine that carry bile, a fluid that helps digest fats and excrete wastes. Diagnosed in over 8,000 Americans per year, the disease has long been treated with surgery, chemotherapy and/or radiation.
Interest in discovering more effective, less toxic treatments — and funding for those efforts — have just recently begun to swell. Now, scientists are focused on learning as much as possible about the genetic glitches that drive individual cholangiocarcinomas, which can be quite different from each other. This could open the door to a variety of specialized treatments, some of which may already be available because they’re approved for use in other cancer types, says Nabeel Bardeesy, Ph.D., who holds the Gallagher Endowed Chair in Gastrointestinal Cancer Research at Massachusetts General Hospital and is an associate professor of medicine at Harvard Medical School in Boston.
“There is good evidence that abnormalities in 40 or 50 genes (contribute to) this cancer; a number of these mutated genes produce proteins that make cancer grow and are potential drug targets,” Bardeesy says. “A cancer cell has different complements of genes that are turned on or off. The field has been asking: What mutations are present in cancer cells, and which genes are on or off? If we can identify subtypes of (this) cancer with different sets of alterations, that will predict therapeutic approaches.”
Bardeesy presented the information during the 2018 Cholangiocarcinoma Foundation Annual Conference, held Jan. 31 through Feb. 2 in Salt Lake City. He and conference co-chair Katie Kelley, M.D., an associate professor of clinical medicine at the University of California, San Francisco, worked together over the past year to develop the direction and agenda of the conference. Speakers included laboratory-based scientists, epidemiologists and clinical researchers from around the world whose work focuses on biliary tract cancers or relevant molecular pathways. The conference also included patients and caregivers, many of whom received funding to travel there through grants from the Patient-Centered Outcomes Research Institute.
Progress is being made toward identifying the disease’s subtypes.
Cholangiocarcinomas that occur in the small bile ducts within the liver are termed “intrahepatic.” Those that arise in the bile ducts but outside the liver are “extrahepatic.” In some cases, these tumors arise at the junction of the left and right hepatic ducts immediately upon exiting the liver (“perihilar” or “Klatskin”) or further down the bile duct, closer to the pancreas (“distal”). The incidence of these cancers, particularly those that are intrahepatic, is rising significantly in the United States, as well as worldwide.
Four or five years ago, Bardeesy says, there was no clear therapeutic distinction between intrahepatic/distal versus extrahepatic cholangiocarcinomas. “We knew there were a small number of mutations that were common to both types,” he says, “so there was no reason to have a distinct treatment hypothesis. Now, some mutations (are emerging that are) unique to intrahepatic cholangiocarcinoma.”
The current school of thought, Bardeesy says, is that there are multiple subtypes of cholangiocarcinoma, distinct based on the genetic alterations, histopathologic features and profile of immune cells associated with the tumor. Moreover, he says, existing medications may treat a number of the alterations identified, such as abnormalities in the FGFR2, IDH1, BRAF and PIK3CA genes.
“These cancers need to be studied,” he says. “We want to know what’s common and to resolve these into molecular subtypes, and think of them as a spectrum of disease with related features but different underlying genetics.”
Overall, at least 40 to 50 percent of intrahepatic cholangiocarcinomas may have mutations that currently available medications could potentially target; the percentage is somewhat lower for extrahepatic cases, said Kelley, who also presented data on targeted therapies and immunotherapy for cholangiocarcinoma at the conference.
Enough is known about the genomics of cholangiocarcinoma, however, to state that all patients with advanced stages of disease who require systemic therapy should have their cancer cells genetically sequenced as part of the treatment decision process, which could include clinical trials, Kelley says. “If you have advanced cholangiocarcinoma,” she says, “these data do warrant molecular profiling.”
What genetic abnormalities are doctors looking for when they conduct this sequencing?
For one thing, Bardeesy says, deletions or gains of chromosomes that occur in cholangiocarcinoma can result in too many or too few copies of certain genes and the cell-growth messages they send. Deletions can be especially problematic if they remove an area of the chromosome that encodes a gene important to preventing abnormal growth, he says, noting that this can happen if the genes p53 or BAP1 disappear.
A similarly meaningful event is the translocation of chromosomes that get broken and improperly fused together during cell division. For instance, fusions involving the FGFR2 gene affect 20 percent of patients with intrahepatic cholangiocarcinoma. “We’re trying to fully understand them,” Bardeesy says.
FGFR2 fusions remove signals that would otherwise control the division and multiplication of cells, allowing cancer cells to grow out of control. A number of promising experimental drugs have been designed to stop that process by disturbing the FGFR2-driven cellular signaling pathways that the cancer relies on to grow.
In a clinical trial, “most patients with cholangiocarcinoma and FGFR2 fusions in their tumors showed a reaction to the (experimental FGFR inhibitor) drug BGJ398,” Bardeesy says.
Several clinical trials are being conducted with these experimental drugs. One recently published study included 63 patients with cholangiocarcinoma and various types of FGFR abnormalities who took BGJ398, most of whom experienced tumor shrinkage, Kelley says. “The most pronounced responses were in the patients with fusions,” she says, “and the responses lasted a long time — in some cases, well over a year.”
Side effects from FGFR2 inhibitors included elevated phosphorus, which can require additional medications, as well as swelling and pain of the hands and feet, dry eyes and fingernail changes, Kelley says.
Just as interesting to Bardeesy’s colleagues was the fact that some tumors eventually stopped responding to the drug. The researchers sought to find out why by monitoring tumor mutational status in a cohort of patients after disease progression on BGJ398. They found that the tumors had developed new mutations in FGFR2 that specifically prevented the drug from working.
Luckily, Bardeesy says, several other pharmaceutical companies have developed experimental FGFR inhibitors that work differently and are not affected by these new mutations. In a trial, 23 patients with cholangiocarcinomas harboring FGFR mutations were treated with the investigational drug TAS-120, and most of the tumors shrank on this second treatment — including those of some patients whose disease had progressed while they were taking BGJ398. “So, we understand part of the mechanism of resistance and also, in part, how to overcome it,” Bardeesy says. “These are important steps in the right direction.”
A next step will be to study scientific models of cholangiocarcinoma with altered FGFR2, Bardeesy says. One model involves mice that are genetically engineered to have the disease, so that researchers can work to understand and possibly prevent the cancer. A second model involves cell lines established from patient tumors that carry FGFR2 alterations. The researchers are establishing a large bank of cholangiocarcinoma cell lines, which could help them better understand the biology of this cancer and identify vulnerabilities to existing drugs or experimental therapies, he says.
In addition to FGFR2 mutations, many other genetic abnormalities can occur in cholangiocarcinoma and are potentially treatable with drugs, Kelley says. These include IDH1 gene mutations, ERBB2 (also known as HER2) gene amplification, BRAF mutations and the presence of mismatch repair deficiency. In the latter condition, cancer cells can’t repair their own DNA when it is damaged, resulting in high rates of mutations in DNA microsatellites throughout the tumor genome, called microsatellite instability (MSI). MSI makes tumors more recognizable to the immune system when treated with new immune checkpoint inhibitor antibodies.
Like FGFR2 fusions, IDH1 mutations occur in about 20 percent of intrahepatic cholangiocarcinomas, but they are almost never seen in the presence of an FGFR2 fusion and almost never occur in extrahepatic cases, Kelley says. A phase 1 clinical trial tested the experimental IDH1-targeting drug AG-120 (ivosidenib) in 73 patients, most of whom had intrahepatic cholangiocarcinoma.
Tumor shrinkage was substantial in only 5 percent of participants, but 56 percent experienced prolonged stable disease for more than a year, and side effects have been very tolerable, she said.
“The ClarIDHy study of the IDH1 mutation-targeting drug has become the first biomarkertargeted phase 3 randomized trial in cholangiocarcinoma matching a patient’s tumor mutation to a targeted treatment,” Kelley says.
“This is a true milestone in the field and is how we hope to conduct more cholangiocarcinoma trials in the future.” If approved, ivosidenib may later be combined with other treatments, such as chemotherapy or other targeted therapies, and be used to an even greater advantage, she suggests. Other promising targets in cholangiocarcinoma include ERBB2 amplification; BRAF, KRAS, PIK3CA or IDH1 or 2 mutations; FGFR1-3 fusions or amplifications; CDKN2A/B loss; ARID1A loss; MET amplifications; and rare fusions including ROS1, ALK and NTRK1.
All those abnormalities might respond to targeted drugs. But what might be the role of immunotherapy in treating cholangiocarcinoma?
“There’s a great deal of interest in understanding how cancers can co-opt immune function,” Bardeesy says, so that the body doesn’t recognize or fight cancer. That can happen when tumor cells recruit normal cells that should be fighting the disease. Studying these patterns “can give a sense of whether different subsets of cholangiocarcinoma patients have different recruitment of immune cells,” ultimately leading to ways to better modulate immune system function, he says.
About 2.5 percent of patients with cholangiocarcinoma have tumors with mismatch repair deficiency, which can be detected by MSI testing showing MSI-high status. They tend to respond exquisitely to immunotherapies known as PD-1 or PD-L1 immune checkpoint inhibitors, which help remove the brakes from the body’s immune system so that it can more effectively fight cancer, Kelley says. The immunotherapy Keytruda (pembrolizumab) is already approved by the Food and Drug Administration (FDA) to treat any solid tumor that is MSI-high or has mismatch repair deficiency, she notes, based upon clinical trials showing dramatic tumor shrinkage and prolonged, durable responses across histologic subtypes. “The approval for pembrolizumab in mismatch repair-deficient solid tumors represents the first FDA approval for a drug agnostic of histology,” meaning cancer subtype, Kelley says.
“These data warrant testing for mismatch repair deficiency in cholangiocarcinoma,” she adds, “because immune checkpoint inhibitors can achieve a remarkable benefit in the majority of patients with MSI-high or mismatch repair-deficient solid tumors, including cholangiocarcinoma, with tumor responses that are dramatic and sustained.
“One of my patients had a spontaneously occurring MSI-high mutation in her intrahepatic cholangiocarcinoma that had widely metastasized and progressed on chemotherapy,” Kelley continues. “She was treated with pembrolizumab as part of a clinical trial, and after two years, no residual tumor could be found. She then stopped the drug and continues to have an ongoing complete response, now approaching four years after she initially started treatment on the trial.”
Clinical trials are being conducted to determine whether patients who have cholangiocarcinoma without mismatch repair deficiency (meaning that they have “microsatellite stable” tumors) could also benefit from checkpoint inhibitors, and whether the drugs will work better if combined with other treatments, Kelley says.