CALIBER: a phase II randomized feasibility trial of chemoablation with mitomycin-C vs surgical management in low-risk non-muscle-invasive bladder cancer

Abstract

Objectives: To evaluate the activity of intravesical mitomycin-C (MMC) to ablate recurrent low-risk non-muscle-invasive bladder cancer (NMIBC) and assess whether it may enable patients to avoid surgical intervention for treatment of recurrence.

Patients and methods: CALIBER is a phase II feasibility study. Participants were randomized (2:1) to treatment with four once-weekly MMC 40-mg intravesical instillations (chemoablation arm) or to surgical management. The surgical group was included to assess the feasibility of randomization. The primary endpoint was complete response to intravesical MMC in the chemoablation arm at 3 months, reported with exact 95% confidence intervals (CIs). Secondary endpoints included time to subsequent recurrence, summarized by Kaplan-Meier methods.

Results: Between February 2015 and August 2017, 82 patients with visual diagnosis of recurrent low-risk NMIBC were enrolled from 24 UK hospitals (chemoablation, n = 54; surgical management, n =28). The median follow-up was 24 months. Complete response at 3 months was 37.0% (20/54; 95% CI 24.3-51.3) with chemoablation and 80.8% (21/26; 95% CI 60.6-93.4) with surgical management. Amongst patients with complete response at 3 months, a similar proportion was recurrence-free by 12 months in both groups (84%). Amongst those with residual disease at 3 months, the 12-month recurrence-free proportion was lower in the surgical management group (40.0%) than in the chemoablation group (84%). Recruitment stopped early as chemoablation did not meet the prespecified threshold of 45% complete responses at 3 months.

Conclusion: Intravesical chemoablation in low-risk NMIBC is feasible and safe, but did not demonstrate sufficient response in the present trial. After chemoablation there may be a reduction in recurrence rate, even in non-responders, that is greater than with surgery alone. Further research is required to investigate the role and optimal schedule of neoadjuvant intravesical chemotherapy prior to surgery for NMIBC.

Keywords: #BladderCancer; #blcsm; chemoablation; mitomycin-C; non-muscle-invasive bladder cancer; randomized trial; surgery.

Conflict of interest statement

Mr. Mostafid reports grants from NIHR, during the conduct of the study; personal fees from Astrazeneca, personal fees from Olympus, personal fees from Cepheid, personal fees from Medac, outside the submitted work. Prof Hall reports grants from Department of Health, during the conduct of the study; grants and non‐financial support from Merck Sharp & Dohme, grants and non‐financial support from Astra Zeneca, grants from Janssen‐Cilag, grants and non‐financial support from Bayer, grants from Kyowa Hakko UK, grants from Alliance Pharma (previously Cambridge Laboratories), grants from Aventis Pharma Limited (Sanofi), grants from Cancer Research UK, grants from Accuray Inc., outside the submitted work. Prof. Catto reports personal fees from Astra Zeneca, personal fees from Janssen, personal fees from Roche, personal fees from Ferring and personal fees from MSD outside the submitted work. Prof. Knowles reports grants from Cancer Research UK, outside the submitted work.

© 2020 The Authors BJU International published by John Wiley & Sons Ltd on behalf of BJU International.

Figures

Fig. 1

CONSORT diagram. Eighty patients were included in the primary and efficacy endpoints’ analysis: two patients without a 3‐month assessment in the surgical management group were excluded (one withdrew from trial treatment after randomization, one was lost to follow‐up before 3 months). All patients for whom there were completed post‐treatment and/or 3‐month adverse event forms were included in the safety analyses (N = 81). Nine patients (three surgical management, six chemoablation) were found ineligible after randomization but were included in all analyses in accordance with the CALIBER Statistical Analysis Plan.

Fig. 2

Response at 3‐month assessment: visual vs histological confirmation.

Fig. 3

Kaplan–Meier estimate of proportion of patients free of subsequent recurrence after 3‐month disease assessment, by allocated treatment (A) and by allocated treatment and disease status (B). Patients who had a second primary cancer or died for reasons other than bladder cancer without a prior recurrence were censored at date of second primary or date of death. Stratified log‐rank test and stratified Cox model to explore the differences between treatment groups were used as appropriate to account for disease response status at 3 months (A). When treatment and disease status were combined to form four groups, these were compared by log‐rank test (not stratified). Proportional hazards were tested using Schoenfeld residuals.

Fig. 4

Health‐related quality of life: change from baseline in QLQ‐C30 global health scale. High score at any timepoint represents high quality of life. Positive change from baseline (calculated score at timepoint – score at baseline) represents improvement in quality of life. Questionnaire return rates were 91% at baseline, 72% at 3 months after end of treatment, 92% at 6 months, and 85% at 12 months. BL, baseline.