 Very good. You’ve got the floor. Why don’t you end talking about research and the future? Walk us through what the current studies are and what you’re seeing out there. Not asking you to bet on a horse, but I think we’ve got some pretty good evidence that these trials should prove positive if they’re properly enriched and well designed. Give us a sense of what currently is out there accruing.
 

 Oh, sorry. As we learn about the molecular classification of cholangiocarcinoma, certainly new therapies are coming, new ways to target the FGFR pathway, to target IDH, looking at novel subsets, for example, we alluded to earlier EGFR and ERBB2 family members, so can we come back and repurpose some existing drugs in this population. Again, mutations matter, and being able to do these studies collectively is challenging because it’s a rare cancer, looking at rarer subsets, but fortunately we’ve learned how to do this for FGFR, for IDH. The other advantage to these even smaller subsets, for example, BRAF in cholangiocarcinoma, is that the response rates, like you said in the highly selected population, are higher. So with an enrichment of the population in a disease where alternative therapies aren’t that beneficial, the bar is a little easier to do smaller studies, and that’s fortunate because we’re talking about smaller and smaller subsets of these populations. So new targeted therapies, work going on in immunotherapy as well in this population, but also new drugs that are improvements on gemcitabine or cisplatin. We’re excited to see these new opportunities, new combinations. Again, I come back to the earlier comment about keeping doctors away from our patients, who think our patients are not going to survive their cancer. There is hope, there are good therapies, don’t give up, whether it’s standard therapies or clinical trials, we’ve got great options for these patients. Don’t be afraid of this disease.
 

 Milind, I see a lot of these frontline studies are single agent versus GEM/CIS, so a non-chemo, whereas what we used to do is three drugs versus two drugs. Is it the toxicity? Is it you think they can stand alone and beat the current standard?
 

 We learned from the infigratinib study, John, that patients who receive this agent in the second-line setting did much better in terms of response than the third- or subsequent-line setting. We do not know why. There may be a rationale to use these. And also they’re very effective. If you look at the response rate to gemcitabine and cisplatin, it’s in the 20% range, and in the frontline, the PFS is six months. With FGFR inhibitors, it’s certainly higher than that.
 

But a case can be made perhaps of combining agents. Of course, this is caveat of antiproliferative activity of IDH1, how will that combine with chemo. But it’s been done with leukemia. That’s how they use it in AML. So can we use those sort of agents in the first-line setting in combination with chemo? I think we’ll have to identify the right setting for the right drug. I certainly think first-line single agent FGFR makes sense, for some single agent, perhaps BRAF V600E makes sense. But some of the more static drugs with a low response rate, we’ll be looking at combination approaches.
 

 I’m assuming that all these frontline studies have crossover, you almost would have to.
 

 And I would also make the comment that the FDA perhaps needs to change their standard(?) in this disease, with an accelerated approval been given in 100 patients, requiring a confirmatory like 400-patient study in, you know, 10% of cholangios. We have to be realistic in terms of expectations.
 

 Yeah, I get both sides of that argument, too. It’s nice to have safety, that when you play this out in the real world, do you learn something different than you do when you see them in centers like ours. So I do think you’re right, that somehow real-world evidence might be supporting versus these more controlled trials. But they’re real close by, I’ll drive up there and let them know that you said that. They are listening. I think to be fair, the FDA has never been, in my memory, more open to trying to get these medicines out for patients. They’ve in fact, I think, been pretty good partners, so maybe, in fact, they will listen to those kinds of pieces of advice from us in the trenches.
 

 … to the first-line studies comparing GEM/CIS and infigratinib and other formulations of that approach, meaning combination, of course, is being asked about, but also the idea of cyclical. I’ve seen a handful of patients who are great responders to GEM/CIS, then progress, get an FGFR drug in the second-line setting, do that therapy for a very long time, maybe two years, and with that long gap from their original GEM/CIS exposure, I’ve gone back to GEM/CIS and obtained responses, shrinkage of tumors. So the concept of either combination or maybe even cyclical therapy with GEM/CIS and FGFR therapy is a possibility. We don’t know yet, it has to obviously be studied in a trial to show are we really benefitting people. One advantage of doing that is we may, through a cyclical approach, lessen cumulative toxicities from either, right? Anyway, so the possibility with GEM/CIS, which is a very beneficial therapy, is out there and it’s to be determined how we’ll put them in order.
 

You also alluded earlier to the mutations, the subsequent mutations you think of now. You’re grateful that you’ve got multiple drugs for GIST. So we’re probably, after an FGFR inhibitor, a drug like futibatinib or other second-generation drugs. At some point, we will have an armamentarium, a mutation panel like we do for ABL, for KIT, and ALK, where we can sequentially go from different FGFR agent and retain benefit. That’s sort of in the next couple of years, that I hope we’ll start to see some of that data. But we’re excited that more FGFR drugs are coming. Even the first generation of drugs are benefitting our patients greatly, and so the second generation holds even better hope.
 

 As an aside, are we seeing FGFR fusions in other diseases and are we going to get to a pan-cancer kind of approvals for these, Milind?
 

 As you know, erdafitinib is approved for bladder cancer with FGFR3 fusions, and I think we’re seeing that about 5% of patients, or perhaps even smaller, a lower number, are actually FGFR kinase mutations, in the kinase domain. Some of these drugs seem to be effective, for instance, futibatinib had some very nice data, now recently pemigatinib had a cancer discovery paper. So I think we will find multiple options in terms of both FGFR aberrations and the diseases. I’ve seen a patient with colon cancer with an FGFR fusion.
 

 I have, too. So this is going to be interesting to see how that evolves.
 

 There are a number of basket studies addressing that question right now, that’s far less common outside of cholangio and urothelial, the head-and-neck cancer, uterus cancer, that not only have fusions involving FGFR2 and 3, but also, as Milind alluded to, mutations in different parts of the gene, the kinase domain, the extracellular part of the gene and protein. My advice today is if you see a cholangiocarcinoma patient and they have an atypical FGFR mutation, find one of our academic consultants and we’ll get them on a clinical trial, because we want to know if those patients are going to benefit. In the absence of other drive mutations, a mutation that’s somatic in the tumor, that’s FGFR3, could be a novel mutation that’s actionable. I wouldn’t assume it, but we want to get those patients in clinical trials.

Videos

An overview of various treatment opportunities under study that are looking at using novel techniques to help personalize treatment for patients with bile duct cancers. 

Cholangiocarcinoma: Potential Therapies Under Investigation

 Katie, back to you to kind of back up to another major target, and that’s the IDH. Take us a little bit through that if you would.
 

 IDH is a different gene and IDH mutations, IDH1 mutations in particular, rather than driving a purely proliferative signaling pathway in the tumor cells, they actually cause the accumulation of a downstream oncometabolite, 2-hydroxyglutarate, in the cell microenvironment, that causes a differentiation block. That differentiation block leads to proliferation of these baby, immature cells that haven’t fully formed as an epithelial cell, and that’s the proliferative stimulus, is just a collection of baby cells in the environment that leads to the oncogenesis. So it’s sort of a novel way of causing a cancer, and the IDH1 inhibitor, ivosidenib, is novel in that it is trying to reestablish this normal differentiation. So not so much of a cytotoxic mechanism, but rather causing cells to differentiate the way they should into mature cholangiocytes. This is a really fascinating biology, obviously. But what we see in the clinic, again just to reiterate that around 15% of intrahepatic cholangiocarcinomas have an IDH1 mutation. We think this is probably also a distinct clinical phenotype, just like FGFR maybe, and targeting the IDH1 mutation with ivosidenib in a phase 1 trial showed not so much a lot of responses as we might expect with the mechanism as I mentioned, but a prolongation of progression-free survival compared to what we expect with second-line therapy, which, unfortunately, most of our second-line regimens have a PFS, in this chemo realm, that is like FOLFOX and FOLFIRI, in the two- to three-month range really. So based on the phase 1/2 data, the ClarIDHy trial was a randomized phase 3 trial that looked at ivosidenib versus placebo in patients with second- or third-line intrahepatic cholangiocarcinoma with IDH1 mutations. They didn’t have to be intrahepatic, but as I mentioned earlier, almost all of IDH1 mutations are intrahepatic, occur in intrahepatic locations. The ClarIDHy trial primary endpoint was progression-free survival, and they showed a very significant, statistically significant improvement in progression-free survival for ivosidenib over placebo.
 

Now, there were a lot of challenges to interpreting this data, including the heterogeneity of second- and third-line patients being included, the fact that scans were done every six weeks, and ultimately the PFS absolute values were relatively low, with PFS in the treatment arm of only around three months versus in the one-month range, 1.4 months I believe for placebo. But the hazard ratio was very significant, and if we look, there’s distinctly a tale of the curve. We think this is a statistically positive trial, and although the absolute numbers are low, there’s a subset of patients that seem to be deriving a benefit.
 

 Yeah. Sameek, let me turn to you to comment on this. This is a drug that’s approved for our AML patients, I think, so many of our listeners may actually have some experience with the drug. The way I pulled the data out was about 25% positive rate from those who screened, that found a mutation, and as Katie pointed out, I think about 90-plus percent were intrahepatic, but not exclusively intrahepatic. Is this something, you get one of these tests back and you get this IDH1 mutation, is there where you’re calling the guy to say can I get this drug for my patient? Where are we with this?
 

 It’s not yet approved to my knowledge.
 

 Not that I know of either yet, for this indication.
 

 In my practice, we generally try to qualify people for clinical trials if we can, but certainly if they’re not trial eligible, we’re certainly willing to go outside the normal realm to use our expert experience and use therapy that we think could benefit our patients. I agree, going above and beyond to try to help our patients or to get them access to a therapy that we think and we know has data that can help them, I’d be very supportive of that. I also say this is very promising and exciting for this subset of cholangiocarcinoma, but there’s new therapies coming.
 

Sometimes the first generation of a drug isn’t always the best generation and there’s going to be some refinements not only to IDH drugs, but also FGFR drugs. One example is the RET inhibitors that we’ve seen, the latest RET inhibitor is producing far greater responses in some of our patients than the first generation of RET inhibitors from maybe a decade ago now. So I think this is early promising signs, these patients are benefitting clinically, the toxicity I’ll let Milind talk about, but I think it’s a very well tolerated drug, and I’d advocate wholeheartedly for our patients for this.

Considerations for treating appropriate patients with advanced cholangiocarcinoma with an IDH inhibitor. 

R. Kate Kelley, MD, University of California-San Francisco

Articles

Cholangiocarcinoma: Potential Therapies Under Investigation

IDH Targeting in Advanced Cholangiocarcinoma