Introduction:

In patients with intermediate-risk non-muscle–invasive bladder cancer (IR NMIBC), measurable disease by marker lesions provides an opportunity to assess on-target efficacy of novel therapies. FGFR3/2 alterations are frequent in NMIBC and FGFR inhibition may be beneficial to patients with IR NMIBC with FGFR3/2 alterations. Erdafitinib is an oral selective pan-FGFR tyrosine kinase inhibitor approved to treat locally advanced or metastatic urothelial cancer in adults with FGFR3/2 alterations who have progressed during or following ≥1 line of platinum-containing chemotherapy. We report a longer follow-up of an exploratory marker lesion cohort of patients with IR NMIBC (Cohort 3) from the multicohort phase 2 THOR-2 study (NCT04172675) of erdafitinib in patients with NMIBC with FGFR3/2 alterations.

Methods:

Patients were ≥18 years, with histologically confirmed IR NMIBC with FGFR3/2 alterations and recurrent IR disease. All previous tumors must have been grade 1-2, Ta/T1, with no previous carcinoma in situ, a <5% risk of progression for 2 years, and a >50% risk of recurrence using the EORTC risk calculator. Patients had all visible bladder tumors resected by transurethral resection except for one 5-10mm marker tumor. Patients received oral erdafitinib at 6 mg daily in 28-day cycles. Surveillance cystoscopy was performed at 3 months and if complete response (CR) was achieved, urine cytology was performed. Patients with partial response (PR) or CR ≤3 months after starting treatment continued erdafitinib for maximum 2 years or until high-risk disease recurrence, intolerable toxicity, consent withdrawal, investigator decision, or study closure. Exploratory end points included CR rate, duration of response (DOR), best overall response, and safety.

Results:

At data cutoff (27-Jun-2023; median follow-up 10.0 months), 18 patients were enrolled (median age: 63.5 years [range, 47-77]; tumor stage: n=13 Ta, n=5 not staged) and received erdafitinib for median duration 7.1 months (range, 0.7-17.4). Fifteen patients had CR (CR rate, 83.3%); 2 had PR (11.1%); median time to response was 1.15 months; 1 had high grade recurrence (Table). In 17 responders, median DOR was 12.8 months, with 12 responses ongoing, 3 censored, 2 ending with recurrence (1 low-grade; 1 high-grade) (Figure). Most common treatment-emergent adverse events (TEAEs): hyperphosphatemia (100%; n=18), diarrhea (83.3%; n=15), dry mouth (72.2%; n=13), dry skin (50.0%; n=9), dysgeusia (50.0%; n=9). Grade 3 TEAEs (n=1 each): abdominal pain, diarrhea, dysuria, and gastritis. Seven grade 1 treatment-related central serous retinopathy events were reported in 3 (16.7%) patients; events resolved in 1 patient (ongoing in 2). Three (16.7%) patients discontinued due to treatment-related TEAEs. No deaths were reported. 

Conclusion:

In this marker lesion study of an FGFR-targeted therapy, erdafitinib demonstrated efficacy in all treated patients with IR NMIBC with FGFR alterations. Safety data were consistent with the known safety profile of erdafitinib.

Funding: Janssen Research & Development