A discussion on current prognostic and predictive testing assays used to detect FGFR mutations in patients with metastatic urothelial cancer (mUC).
Arlene O. Siefker-Radtke, MD: Hello. It's a pleasure speaking with you today about targeting FGFR [fibroblast growth factor receptor] mutations in metastatic urothelial carcinoma, a program presented by OncLive® Insights. I am Arlene Siefker-Radtke, a professor of genitourinary medical oncology at The University of Texas MD Anderson Cancer Center in Houston, Texas. And I'm joined today by my colleague Dr. Scott Tagawa [Scott T. Tagawa, MD, MS, FACP], who's a professor at Weill Cornell Medicine Urology group in New York, [New York]. Thank you for joining us today, Scott.
Scott T. Tagawa, MD, MS, FACP: Thank you very much for the invitation. Look forward to the discussion.
Arlene O. Siefker-Radtke, MD: Great. I know a lot of us have been talking about the impact of FGFR3 alterations and urothelial cancer and the new agents that have been approved for FGFR3-targeted mutant tumors. And a lot of discussion has been focused about molecular testing, when do you test, who do you test, what does it mean. What are your thoughts on testing? When do you test, Scott?
Scott T. Tagawa, MD, MS, FACP: As a general principle, testing can be important [in] a couple of ways. I'm just speaking in general, particularly for advanced oncology tumors [for which] we can get some prognostic information in general. But even more importantly, what we've always been looking for is predictive information. We've had some prognostic information for a long time. Some of it was thought, [to be] maybe predictive, such as p53 for muscle-invasive [bladder cancer, but] that didn't quite work out. [I]n any case, what is very nice in the current era is to have, not just a therapeutic target, but a drug that goes with it since the accelerated approval of erdafitinib. It's nice that there's an activating mutation in FGFR3 or FGFR2 fusion that's actionable. Some are variants of uncertain things, but the common ones, particularly FGFR3 mutations. What is very nice is we have a drug to go with that.
Arlene O. Siefker-Radtke, MD: That's key, what you pointed out: having a target that not only is prognostic but can be predictive of benefit, because that's when you're going to use it the most. You mentioned the old studies on p53, and boy, that was a challenging marker because other p53 proteins go up and down, p63 and the like. We've even seen PD-L in expression fluctuate, so it's a very dynamic marker…that changes based on whether you've given prior treatment which perhaps has limited some of its utility in urothelial cancer. But what makes FGFR3 so attractive is that it's a reproducible marker; it's based on a gene, and now we have something that works. We have a treatment that may benefit patients. But when you have patients who come to your clinic, when do you do testing? How early or how frequently, and do you do it in all patients?
Scott T. Tagawa, MD, MS, FACP: One aspect is germline, and that has been part of at least some discussion, particularly for upper tract, but even as it looks like for bladder primaries, there's—at least to me—a surprising percentage of patients that might harvest a germline alteration. That for me is one of the most important aspects of knowing that is what we call the cascade effect. Who else in the family might be affected, not just with urothelial carcinoma, but some other cancer? Certainly, we have screening for certain cancers as well as preventative strategies. It is a consideration for many of our patients and we should, at the minimum, consider it and take a good family history. And if available then we refer.
In terms of somatic testing, meaning the alterations in the genome of the tumor itself, on a practical basis, again, what we started with is when does it potentially make a difference? It's for patients with advanced disease. Certain alterations such as FGFR3—I don't want to stick on that only—but there can be changes, as you stated before. Not specifically on FGFR, but other alterations can change, and that can include tumor mutational burden, so if possible, as just a general rule, [I use] the most proximal specimen in time, meaning that if someone has metastatic disease today and I have a specimen from yesterday and I've got the specimen from a TURBT (transurethral resection of bladder tumor) 10 years ago, if available, I’m going to test the one from yesterday, just as a rule. There are some caveats in terms of degradation of tissue over time and the quality of samples, so it's better to test an available specimen even if it's old then not to test at all.
I also advise panel testing when possible. Is there an activating FGFR3 mutation or a fusion? I'm not asking for FGFR only: I'm asking for panel testing just as a general rule. Number (1) because that is generally more cost-effective, and number (2) maybe I'm going to use that information down the line, either for prognostic information or if a patient runs out of the regular types of available therapies. Maybe there's going to be a targeted clinical trial available for them. I would say, as I talk about to 100% of my patients with advanced urothelial carcinoma, we need to test. If a specimen is already available. I will pick a metastatic biopsy that is the most recent. If they haven't had any testing, I generally will try to get a metastatic biopsy, number 1, just to confirm the diagnosis but also for a genomic characterization. I would also say that we have a reasonable platform of plasma testing, with the caveat that most of the commercial platforms do not control for germline, so there can be some occasional problems with clinical metamorphosis.But, if something is there, I believe that to be true although the sensitivity might be a little less.
Transcript edited for clarity.