The FDA ODAC decided that the overall benefit-risk of the investigational therapy UGN-102 is not favorable in patients with recurrent low-grade, intermediate-risk NMIBC.
US FDA
In a narrow 5 to 4 vote, the Oncologic Drugs Advisory Committee (ODAC) decided that the overall benefit-risk of the investigational therapy UGN-102 (intravesical mitomycin) is not favorable in patients with recurrent low-grade, intermediate-risk non-muscle invasive bladder cancer (LG-IR-NMIBC).1
The ODAC convened on May 21, 2025, to discuss the new drug application (NDA 215793) for UGN-102, a novel intravesical formulation proposed for the treatment of adults with recurrent LG-IR-NMIBC.
UGN-102, developed by UroGen Pharma, Inc., was set to become the first FDA-approved therapy specifically indicated for this patient population.
Discussion: Given uncertainty regarding interpretation of the duration of response in low-grade intermediate-risk non-muscle invasive bladder cancer (LG-IR-NMIBC), discuss whether randomized trials should be required in the future to assess the effectiveness of therapies in this disease setting.
Voting Question: Is the overall benefit-risk of the investigational therapy UGN-102 favorable in patients with recurrent LG-IR-NMIBC?
For bladder cancer that is non-muscle invasive at the time of diagnosis, the current standard of care involves repeated transurethral resection of bladder tumors (TURBT), often accompanied by intravesical chemotherapy or Bacillus Calmette-Guérin (BCG). However, this strategy is associated with high recurrence rates and burden on patients, especially elderly individuals with comorbidities.
“Low-grade, intermediate-risk disease recurs frequently and is inadequately controlled with TURBT. Each time you do a resection, the patient has a potential for complications, and we know that to TURBT has substantial cost, not only to the patient, but to the patient’s family, not only in terms of financial costs, but social costs,” explained Sam S. Chang, MD, chief division of urologic oncology, chief surgical officer, Vanderbilt Ingram Cancer Center, during his presentation at the meeting.
UGN-102 is a novel formulation of mitomycin suspended in a reverse thermal hydrogel designed for intravesical delivery. It transitions from liquid to gel upon reaching body temperature, allowing prolonged local exposure of mitomycin to the bladder wall. UGN-102 is intended to be given once weekly for 6 weeks in an office-based setting without anesthesia.
“We need a more scientifically based approach that treats the entire bladder effectively for a longer period of time. I think patients want and need another choice right now,” added Chang.
UroGen Pharma is seeking FDA approval for UGN-102 based on data from several studies, including the pivotal ENVISION trial (NCT05243550), the earlier OPTIMA II study (NCT03558503), and the partially completed ATLAS randomized trial (NCT04688931). Together, these studies aimed to demonstrate that UGN-102 can produce complete responses (CRs) that are durable, safe, and preferable to TURBT, from the patient’s perspective.
The single-arm ENVISION trial was a single-arm, multicenter trial which enrolled 240 patients with recurrent LG-IR-NMIBC. The study’s primary end point of complete response rate (CRR) at 3 months was achieved in 77.6% of patients (95% CI, 71.5%-82.9%). Notably, among patients achieving a CR, over 80% remained recurrence-free at 18 months follow-up.
For the secondary end points and among patients who achieved a CR at 3 months, UGN-102 was associated with a clinically meaningful duration of response (DOR). With a median follow-up was 19.1 months, the median DOR was not estimable. At 12 months after 3-month CR, the probability of remaining event free by Kaplan-Meier estimate was 83.3% (95% CI, 76.7%-88.1%). At 18 months after 3-month CR, durability of response was maintained at later time points, with a Kaplan-Meier estimate of 81.2% (95% CI, 74.4%-86.4%).
“ENVISION is a large trial with adequate follow-up. The patient population we studied accurately reflects the target population for the medicine. The results demonstrate clinically meaningful complete response and duration of response, and the study met all criteria set forth by the FDA to support approval of UGN-102,” said Mark Schoenberg, MD, chief medical officer of UroGen Pharma, during the meeting.
Clinically meaningful disease control rates were also seen. Of the 173 patients who had a CR at the 3-month visit, 137 (79.2%) were directly observed to remain in CR 12 months after the 3-month CR, and 123 (71.1%) were directly observed to remain in CR 18 months after the 3-month CR.
“These results can be reliably compared [with] results obtained following TURBT. Patients benefit from treatment UGN-102, which decreases reliance on TURBT with its associated risks,” added Schoenberg regarding the ENVISION trial data.
OPTIMA II, a phase 2, open-label, single-arm study, also supported the case for UGN-102. In that trial, 63 patients received once-weekly intravesical UGN-102 for 6 weeks. At 3 months, the CRR was 65%, and 72.5% of those who achieved a CR remained recurrence-free at 12 months.
Adverse events (AEs) in the study were primarily mild to moderate and localized to the urinary tract, including dysuria, frequency, and urgency. Importantly, there was no systemic toxicity or treatment-related mortality.
While smaller in size than ENVISION, OPTIMA II helped establish both the dosage and treatment schedule used in later studies and contributed to the consistency of the overall efficacy and safety profile for UGN-102.
Further, the ATLAS trial initially promised to provide more rigorous data through a randomized design. The study compared UGN-102 with or without TURBT to TURBT alone. Early results suggested higher CR rates and longer DOR in the UGN-102 group.
However, the trial was terminated early for operational reasons and never achieved its intended enrollment or statistical power. As a result, the FDA has stated that the data from ATLAS cannot reliably support claims of superiority or noninferiority vs TURBT.
“In addition, ATLAS provides strong supportive evidence of efficacy and durability in patients with recurrent disease. The safety profile UGN-102 is acceptable and manageable in standard neurologic practice, and the totality of the data from our studies, which enrolled almost 600 patients, demonstrates a favorable benefit-risk profile,” said Schoenberg.
In its presentation to the committee, UroGen Pharma emphasized the totality of the evidence across studies, highlighting the consistency of the CRR and the long-term disease control observed in both ENVISION and OPTIMA II.
“Ninety percent of patients interviewed would recommend UGN-102 over TURBT, and patient-reported outcomes showed the quality of life was not adversely impacted by UGN-102,” said Michael J. Louie, MD, MPH, MSc, executive vice president, clinical development and medical affairs, UroGen Pharma, during his presentation. “UGN-102 efficacy results were consistent across the late phase trials and the FDA analysis population of 329 patients. UGN-102’s efficacy compares favorably to that of standard of care with TURBT with or without intravesical chemotherapy.”
According to the applicant, patients reported fewer disruptions to daily life, work, recreation, and sexual activity following UGN-102 treatment compared with TURBT. In addition, the therapy’s safety profile was described as acceptable and manageable, with most treatment-emergent AEs being mild to moderate and primarily localized to the lower urinary tract. The therapy did not negatively impact health-related quality of life and avoided the systemic complications typically associated with chemotherapy.
“The totality of the data indicates that UGN-102 is a highly efficacious treatment for patients with recurrent low-grade intermediate-risk NMIBC. The pivotal vision trial demonstrated robust and clinically meaningful efficacy. The randomized ATLAS trial demonstrated that UGN-102 was associated with a higher complete response rate and a longer duration of response than TURBT. Also, UGN-102 reduced the need for TURBT, which is associated with a mortality rate of close to 1% and which patients with intermediate-risk disease typically experience multiple times during their lifetime,” added Louie.
From a clinical utility standpoint, UGN-102 can be administered in a urologist’s office, eliminating the need for operating room time and hospitalization. This potentially lowers overall health care costs and expands treatment access to patients who may otherwise defer care due to surgical risks.
“If approved, the UGN-102 would provide patients with a treatment option that can break the cycle of repeated TURBTs,” added Schoenberg.
While acknowledging the drug's promise and activity, the FDA expressed significant concerns about the robustness of the data supporting UGN-102’s efficacy and safety. The FDA agreed that the observed 77.6% CRR in ENVISION demonstrates drug activity, especially given that LG-IR-NMIBC lesions are not expected to resolve spontaneously. However, it disputed the applicant’s claim that UGN-102 is superior or more durable than TURBT, citing that the ATLAS trial was not designed or powered to establish such a comparison.
Concerns that undermine ATLAS’s reliability, as stated by the FDA, include the absence of guideline-recommended intravesical chemotherapy in the control arm, inconsistent definitions of disease-free survival (DFS) between study arms, and disagreement with the use of a noninferiority design for a time-to-event end point like DFS. The trial's early, sponsor-led termination without FDA consultation further limited its statistical power, rendering its findings largely uninterpretable and insufficient to support efficacy claims.
“In evaluating the overall benefit-risk assessment, it appears UGN-102 can induce complete responses in this patient population; however the duration of response is challenging to interpret,” said Sundeep Agrawal, MD, clinical team lead, Genitourinary Malignancies, Division of Oncology I, Office of Oncologic Diseases, FDA.
“Evaluable patients in ENVISION had a CRR at 3 months of 77.6%, which is consistent with subgroup analysis and exploratory ATLAS results in a similar evaluable population. [A total of] 79.2% evaluable patients in ENVISION maintained a complete response at 12 months post 3-month complete response; however, there are challenges in interpreting the duration of response in a single-arm trial. Primarily, it is unclear whether duration of response observed is due to the effect of UGN-102 or the natural disease history,” said Brian Heiss, MD, clinical reviewer, Genitourinary Malignancies, Division of Oncology I, Office of Oncologic Diseases.
“So, is a single arm trial sufficient to inform benefit-risk in this setting? There are several weeks of potential toxicity burden and unclear recurrence risk for individual patients with low-grade intermediate-risk NMIBC. And [there’s] uncertainty as to whether the duration of response observed in ENVISION can be attributed to the drug or the disease. The benefit-risk considerations differ in a low-grade intermediate-risk population compared with other settings where single-arm trials have supported drug approvals, such as BCG-unresponsive [carcinoma in situs (CIS)] for upper tract urothelial carcinoma,” Agrawal explained.
The FDA questioned the assertion that UGN-102 would reduce the frequency of future TURBTs, noting that long-term data on subsequent treatments were not collected. As such, the real-world impact on procedural burden and health system utilization remains speculative.
While UGN-102 toxicity was deemed manageable, the FDA highlighted that AEs persisted for weeks—a duration longer than those associated with TURBT. The FDA also pointed out that no direct comparison of safety or tolerability was made between UGN-102 and TURBT, undermining claims that the drug is safer or more tolerable.
“We should consider the acute genitourinary toxicity with UGN-102 treatment that results from weekly catheterization and treatment for 6 weeks, with toxicity in some patients taking additional time after the sixth dose to resolve. There is a lack of data to contextualize safety, which results in an inability to understand the risks of UGN-102 in the context of available standard of care,” added Agrawal.
“Comparison of UGN-102 to TURBT analysis is exploratory only and likely underestimates the effect that would be seen with TURBT intravesical chemotherapy, the most active standard of care,” Heiss added.
In what was arguably the most critical regulatory concern, the FDA emphasized that a randomized clinical trial using a time-to-event endpoint (e.g., DFS) would have provided more definitive evidence. Such a design would address:
“We would like the committee to consider, can a durable CR assessed in a single-arm trial establish efficacy in a disease setting? A CR may obviate the need for 1 TURBT but not necessarily future procedures. Prevention/delay of 1 TURBT is different from the prevention/delay of radical procedures. The length of time responding patients maintain CR is a key component of establishing clinical meaningfulness, but in a single-arm trial it is difficult to distinguish between treatment effect vs natural history of disease when interpreting DOR. Furthermore, there is a lack of well established historical control,” concluded Heiss.
Ultimately, the ODAC was tasked with advising the FDA on whether the benefits of UGN-102 outweigh its risks for the treatment of adults with LG-IR-NMIBC and decided that the overall benefit-risk profile of UGN-102 is not favorable for patients with recurrent LG-IR-NMIBC. While acknowledging the therapy's potential and patient preference for a nonsurgical option, the committee cited unresolved concerns regarding trial design, lack of comparative data, and uncertainties around durability of response.
“I was thinking about benefit-risk. I felt that the risk was lower and harder to compare it with apples and oranges because these are different procedures, but that the benefit was very hard to ascertain given that heterogeneity. I also think that this was a missed opportunity to run a randomized control trial given the frequency of the disease,” explained Neil Vasan, MD, PhD, physician-scientist in the Department of Medicine and the Herbert Irving Comprehensive Cancer Center at Columbia University Irving Medical Center, and standing member of the ODAC who voted no.
“We have heard a number of critiques of the data presented today that could have perhaps been mitigated by having a randomized trial. That being said, there seems to be precedent in this disease phase to use a single-arm trial design. While it's always sort of the highest degree of evidence to have randomized trial, I think saying it should be required is up to interpretation. I don't know it's required, although it is preferable,” Mark W. Ball, MD, FACS, an attending surgeon in the Urologic Oncology Branch of the National Cancer Institute and associate program director of the Urologic Oncology Fellowship Program, who voted yes to the voting question, explained during the discussion session.
“To me, this is far too short in this indolent disease process to know how this will impact subsequent salvage therapies,” said Daniel Spratt, MD, ODAC member and chairman and professor of radiation oncology at University Hospital Seidman Cancer Center, who voted also no.
