Key Considerations for Choosing Between Erdafitinib and Enfortumab Vedotin in Urothelial Carcinoma

When choosing between enfortumab vedotin-ejfv (Padcev) and erdafitinib (Balversa) in the second-line treatment of advanced or metastatic urothelial cancer, several factors such as FGFR mutational status, comorbidities, adverse effects (AEs), and available data must be considered, said Michael E. Devitt, MD.

“Genomic testing [is important] because you are never going to be able to use erdafitinib if you do not test for FGFR mutations. It may also help simplify your decision about what drug you are going to use in the third line. Only about 20% of the time or less will you detect an FGFR alteration and be able to use erdafitinib,” Devitt said. “There is no clear-cut winner between the 2 [agents] in terms of what should be your go-to [choice]. Ensure that you are familiar and comfortable with both drugs, and tailor which medication you are going to use based on safety profile and the efficacy data, [to some degree].”

In an interview with OncLive® during an Institutional Perspectives in Cancer webinar on genitourinary cancers, Devitt, a hematologist and oncologist and an assistant professor at the University of Virginia Medical Center, further underscored the importance of early genomic testing in patients with urothelial carcinoma, shared advice on factors to consider when choosing among second-line and later options, and highlighted enfortumab vedotin combination regimens that are under exploration.

OncLive^®: How do you approach choosing between enfortumab vedotin and erdafitinib in the secondline setting for patients with urothelial carcinoma?

Devitt: When you are choosing between the 2 drugs, the first step is always to get genomic information. We know that in order for a patient to be eligible and to derive benefit from erdafitinib, they have to have an alteration in their FGFR gene—either a mutation or a translocation of the gene—that changes its function in a way that can be targeted with [the drug]. Those mutations are generally present in only about 20% of patients, at most. The majority of the time that you go looking for this gene, you are not going to find it, and the choice is made for you: Your only option is enfortumab vedotin. Things get complicated when the patient does have the genetic mutation present; choosing between these 2 drugs becomes more challenging because we do not have randomized data to guide that decision.

[I consider] patient factors and comorbidities. One key comorbidity with erdafitinib is central serous retinopathy. [As such], patients with underlying visual issues or other retinal pathologies may not tolerate the drug quite as well. However, those AEs are generally reversible by withholding the drug. Other issues we see with erdafitinib is hyperphosphatemia. [We would need to] monitor that very closely and decide whether a patient can tolerate that potential toxicity.

With enfortumab vedotin, we see AEs like significant skin reactions. However, other more specific [factors] that might [sway] me in [my choice between the 2 agents] are [factors such as] underlying peripheral neuropathy. Many of these patients have been exposed to platinum chemotherapy, so they may already have some neuropathy, and enfortumab vedotin can worsen that. We also see some hyperglycemia with enfortumab vedotin in a small number of patients. If a patient [has] poorly controlled diabetes, that may be another [comorbidity] that you might factor into your decision.

[The drugs] have relatively similar response rates when you compare across trials, [although we are not really supposed to [do that]. The only other factor that I [would highlight] would be the randomized data supporting the use of enfortumab vedotin [from] the phase 3 EV-302 trial [NCT04223856] that were presented during the 2021 Genitourinary Cancers Symposium. Results showed that [the agent] does have a survival advantage, [instead of] just having the response rate as the main driver for efficacy. That helps strengthen the recommendation for enfortumab vedotin.

For those patients who are eligible for both options, obviously we would love to have head-to-head data to help us make that decision, but I am not sure that we will ever get [those]. A poster from a group out of Memorial Sloan Kettering Cancer Center [presented] at the same meeting looked at patients with FGFR alterations who had received both agents. [Results] suggested a slight benefit to receiving erdafitinib before enfortumab vedotin. However, this was a very small, single-institution study, so it is hard to make any big conclusions.

What combination approaches are utilizing enfortumab vedotin?

The most promising combination that is currently being investigated is enfortumab vedotin and pembrolizumab [Keytruda]. A phase 2 trial [NCT03288545] demonstrated an incredible response rate of 71% in the first-line setting with this combination in patients with metastatic urothelial carcinoma. Currently, the ongoing phase 3 EV-302 trial is comparing enfortumab vedotin plus pembrolizumab with standard-of-care chemotherapy in the frontline setting. We currently have the trial open at the University of Virginia, and several patients have enrolled; I am very excited about that. A response rate of 71% in this disease is unheard of. To see that in the phase 2 setting is really promising.

What biomarkers are being utilized? Are any emerging biomarkers under investigation?

The 2 biomarkers that we use in practice would be PD-L1 expression and FGFR mutational status. I use PD-L1 for patients who are not cisplatin eligible in the frontline setting to help guide my decision on whether they would be better suited to receive carboplatin-based chemotherapy vs immunotherapy. For patients who have no or low expression of PD-L1, generally, response rates to PD-1 inhibitors are much lower, and [these patients] are much less likely to benefit from frontline immunotherapy. I will generally favor carboplatin/gemcitabine chemotherapy for those patients.

The phase 3 JAVELIN Bladder 100 [NCT02603432] data, to some degree, challenge that idea in that perhaps we should be giving anyone frontline chemotherapy if we feel they are eligible, whether it be cisplatin or carboplatin based, given the fact that we can then move them onto maintenance avelumab [Bavencio]. For the time being, I am still using PD-L1 expression to help guide that decision for cisplatin-ineligible patients.

FGFR mutational status is important in the secondline metastatic setting or the postplatinum metastatic recurrence setting. Looking for alterations in FGFR2/3 as a marker for response with erdafitinib [is key] in the metastatic setting. What is really important with this biomarker is ensuring that you get it early. At this point, essentially, every patient who develops metastatic urothelial carcinoma should undergo genomic testing to look for FGFR alterations. That test does take time and sometimes there are quality issues, so you may need to resend it or attempt circulating tumor DNA [ctDNA] testing in place of traditional tissue testing to look for the marker. You do not want to have to do that if you have a patient who is actively progressing on their first or second line of therapy and you are trying to scramble to find out whether they might be eligible for erdafitinib. That is something that all oncologists should be thinking about doing for their [patients with] newly diagnosed metastatic urothelial carcinoma.

Do you feel that ctDNA is ready for prime time in this patient population?

It can certainly be useful in these patients, but I still believe that we have to default to tissue-based testing for the time being. In some other diseases, lung cancer being a great example, [ctDNA has] pretty good specificity and sensitivity for detecting genomic alterations; it is fairly comparable with tissue-based testing [and] obviously comes with a lot less risk to the patient, which is certainly beneficial.

I still currently default to tissue-based testing if I have tissue easily available. If I do not have archived tissue, whether it be [from] the primary resection or the biopsy performed at the time of diagnosis, and I do not have a safe or adequate site of disease to biopsy, I will use ctDNA as my backup to obtain genetic information. It is useful enough in that realm to help us guide that decision.