BALVERSA - Central Serous Retinopathy/Retinal Pigment Epithelial Detachment

CLINICAL DATA

To provide the most relevant information, the summary below is limited to information from the pivotal phase 3 (THOR; cohort 1) and phase 2 (BLC2001) studies in patients with locally advanced or metastatic urothelial carcinoma (UC).

Phase 3 Study in Patients with Locally Advanced or Metastatic UC (THOR)

Interim Analysis

Interim results from cohort 1 (n=266) were presented after a median follow-up of 15.9 months. The median overall survival (OS) was 12.1 months for patients receiving BALVERSA vs 7.8 months for patients receiving chemotherapy. A 36% reduction in risk of death was observed in patients receiving BALVERSA (hazard ratio [HR], 0.64; 95% confidence interval [CI], 0.47-0.88; P=0.005). Median progression-free survival (PFS) was 5.6 months for patients receiving BALVERSA vs 2.7 months for patients receiving chemotherapy. Objective response rate (ORR) was 45.6% for patients receiving BALVERSA vs 11.5% for patients receiving chemotherapy. The safety profiles were consistent with the known safety profiles of BALVERSA and chemotherapy.¹

• In the BALVERSA treatment arm (n=135), 62 (45.9%) patients had grade 3-4 treatment-related adverse events (TRAEs; most frequent grade 3-4 TRAEs were palmar-plantar erythrodysesthesia syndrome [13 patients; 9.6%], stomatitis [11 patients; 8.1%], onycholysis [8 patients; 5.9%], and hyperphosphatemia [7 patients; 5.2%]), 18 (13.3%) patients had treatment-related serious adverse events (AEs), and 1 treatment-related death occurred (reported as sudden death).
• The median duration of exposure in the BALVERSA treatment arm was 4.8 months (range, 0.2-38.2).
• In the BALVERSA group, AEs of any cause led to treatment discontinuation in 19 (14.1%) patients, and TRAEs that led to treatment discontinuation occurred in 8.1% of patients.
• Grade 3/4 AEs of interest based on the known safety profile of BALVERSA included central serous retinopathy (CSR; 2.2%). The incidence of CSR AEs of interest is summarized in Table: THOR: Cohort 1 CSR AEs of Interest: BALVERSA.
  • In 16 of 23 patients (70%) with any-grade CSR, events were resolved by the clinical cutoff date. Of the 7 patients with ongoing events, 5 experienced grade 1 events.

Phase 2 Study in Patients with Locally Advanced or Metastatic UC (BLC2001)

Final Analysis

In the final analysis, after a median 24-month follow-up for efficacy (interquartile range [IQR], 22.7-26.6) and median treatment duration of 5.4 months (IQR, 2.8-9), 101 patients were treated with an uptitration regimen of BALVERSA 8 mg daily (2 patients enrolled after the clinical cutoff date for the primary analysis). Overall, 60 patients received BALVERSA 8 mg daily and 41 were uptitrated to BALVERSA 9 mg daily.^6,7

• The safety profile of BALVERSA remained consistent with that reported in the primary analysis.⁶
• No grade 4 AEs were considered related to BALVERSA, and no new TRAEs were observed with a longer follow-up.⁶
• A post hoc analysis was conducted to determine the cumulative incidence of first-onset CSR by grade.⁶
  • The median time to first onset of all-grade and grade 3 CSR events was 53 days (IQR, 32-100 days) and 94 days (IQR, 72-154 days), respectively.
  • Overall, 27 of 101 patients (27%) reported CSR (BALVERSA 8 mg daily, n=15/60 [25%]; BALVERSA uptitration to 9 mg daily, n=12/41 [29%]).
    • At data cutoff, 17 CSR events resolved, with a median time to resolution of 27 days (range, 9-299 days). The 10 unresolved events were grade 1-2.
• Treatment-emergent CSR events and their outcomes are presented in Table: BLC2001: TreatmentEmergent CSR Events and Outcomes for Grade ≥3 Events.

• The incidence of treatment-emergent CSR events is summarized in Table: BLC2001: TreatmentEmergent CSR Events.

An exploratory analysis was conducted for 2 dosing regimens in the BLC2001 study to investigate the relationship between selected efficacy and safety endpoints and phosphate (PO₄) concentrations. Exposure analyses supported the use of pharmacodynamically (PD)-guided dose titration to optimize therapeutic benefit/risk ratio.⁸

Pooled Safety Analysis in mUC Studies and RAGNAR STudy

• Patients treated with BALVERSA across five metastatic urothelial carcinoma (mUC) studies (N=479, including THOR, BLC2001, BLC2001 drug-drug interaction [DDI] substudy, NORSE, EDI1001 study) and the phase 2 RAGNAR study (N=314) were included in a pooled analysis to evaluate fibroblast growth factor receptors inhibitors (FGFRi)-associated retinopathy.⁹
• FGFRi-associated retinopathy treatment-emergent adverse events (TEAEs) included diagnoses such as detachment of retinal pigment epithelium, serous retinopathy, subretinal fluid, retinal edema, retinal thickening, chorioretinopathy, choroidal effusion, macular detachment, maculopathy, retinal detachment, retinopathy, serous retinal detachment, chorioretinitis, and vitreous detachment.
• FGFRi-associated retinopathy was monitored using Amsler grid testing. Patients who had a positive Amsler grid test or patient-reported symptoms indicative of FGFRi-associated retinopathy were referred for ophthalmologic examination, including visual acuity assessment and optical coherence tomography (OCT).
  • Management consisted primarily of BALVERSA dose interruption or dose reduction, with treatment discontinuation required in a small proportion of patients.
• The incidence of FGFRi-associated retinopathy across clinical studies are presented in Table: Incidence of FGFRi-associated Retinopathy in the mUC Pooled Studies and RAGNAR Study.

• Most FGFRi-associated retinopathy events occurred within 90 days of BALVERSA initiation (mUC pooled studies: 81 [78.6%]; RAGNAR: 30 [70%]) and were managed with BALVERSA dose interruption or reduction, with few requiring treatment discontinuations.
  • Most events were grade 1 or 2. Grade 3 events were infrequent.
• The characterization of FGFRi-associated retinopathy across clinical studies are summarized in Table: Overview of FGFRi-associated Retinopathy in the mUC Pooled Studies and RAGNAR Study.

• In mUC pooled studies, discontinuations of BALVERSA due to ≥1 FGFRi-associated retinopathy event were attributed to retinal pigment epithelial detachment (RPED, 8 [1.7%]), chorioretinopathy (3 [0.6%]), maculopathy (2 [0.4%]), retinal detachment (1 [0.2%]), and subretinal fluid (1 [0.2%]).
• In RAGNAR, discontinuations of BALVERSA due to FGFRi-associated retinopathy events, were attributed to maculopathy, retinopathy, and serous retinal detachment (n=1 each).
• Age-stratified characterization of FGFRi-associated retinopathy across the pooled mUC studies are summarized in Table: Overview of FGFRi-associated Retinopathy by Age Group in the mUC Pooled Studies.
• In the RAGNAR study, FGFRi-associated retinopathy between >65 years vs ≤65 years was 11/79 (13.9%) vs 32/235 (13.6%). Other measures listed in the table below for the mUC pooled studies below were not reported for RAGNAR.

• In RAGNAR, 23/25 (92%) patients with available follow-up assessments returned to baseline visual acuity or better, and 22/28 (78.6%) patients with retinal pigment epithelium elevation were stable or resolved at the last assessment.

csr Screening

Screening of CSR in the THOR Study¹¹:

• The Amsler grid was used to screen for CSR at screening, at the beginning of every new cycle from cycle 2 onwards, and at the end-of-treatment visit.
• Screening was performed by the study physician or nurse.
• Observation of wavy, broken or distorted lines, or a blurred/missing area of vision was considered positive.
• Referral for a comprehensive ophthalmological examination was made within 7 days for a positive Amsler grid test.

csr MANAGEMENT

• Follow ophthalmological monitoring and dose modification guidelines for ocular adverse reactions as described within product labeling to include dosing interruption and discontinuation. The interventions investigators in the THOR study were instructed to perform to manage CSR-suspected ocular disorders included¹¹:
  • A consultation with a retina specialist
  • Performing an optical coherence tomography (OCT), and obtaining color fundus photos or OCT images for future reference
  • Consideration of fluorescein angiography in conditions such as suspected retinal vein occlusion (RVO)

case reports

• The occurrence and management of individual CSR and RPED events in patients receiving BALVERSA have been reported.^12-21

Literature Search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File (and/or other resources, including internal/external databases) was conducted on 21 July 2026.