High-risk non-muscle-invasive bladder cancer (NMIBC) remains a therapeutic challenge, with substantial rates of recurrence and progression despite standard treatment. Current standard of care consists of transurethral resection of the bladder tumour (TURBT), followed by intravesical Bacillus Calmette-Guérin (BCG) induction and maintenance.1 However, outcomes are often suboptimal. The phase III CREST trial evaluated the efficacy and safety of adding sasanlimab, a subcutaneously administered PD-1 inhibitor, to intravesical BCG in patients with high-risk, BCG-naïve NMIBC.2
The global, open-label, phase III CREST trial randomised patients with BCG-naïve, high-risk NMIBC in a 1:1:1 ratio to one of three arms: sasanlimab plus BCG induction and maintenance (arm A), sasanlimab plus BCG induction only (arm B), standard of care BCG induction and maintenance (arm C). Sasanlimab was administered subcutaneously every four weeks for up to 25 cycles. BCG was delivered intravesically per standard guidelines, with re-induction and maintenance allowed based on response. The primary endpoint was investigator-assessed event-free survival (EFS) for arm A vs arm C, defined as time from randomisation to recurrence of high-grade disease, persistence of carcinoma in situ (CIS), progression of disease, or death from any cause. Secondary endpoints included EFS in arm B vs arm C and overall survival (OS).
A total of 1,055 patients were enrolled across 140 centres in 14 countries between January 2020 and November 2021. At a median follow-up of 36.3 months, the primary endpoint was met. Patients treated with sasanlimab plus BCG induction and maintenance (arm A) had a significantly longer EFS compared to those receiving BCG alone (arm C) (HR[95% CI]: 0.68[0.49-0.94], one-sided p= 0.0095). The estimated 36-month EFS was 82.1% in arm A compared to 74.8% in arm C. This benefit was consistent across key subgroups, including patients with CIS and those with T1 disease.
In contrast, the addition of sasanlimab without BCG maintenance (arm B) did not improve EFS compared to BCG (arm C). At a median OS follow-up of 40.9 months, no OS differences were observed between arms.
Treatment-related adverse events (TRAEs) of grade ≥3 were more frequent in the sasanlimab arms: 29.1% in arm A, 21.8% in arm B, and 6.3% in arm C. The most common in arm A were increased lipase (6.0%), haematuria (4.0%), increased amylase (2.3%), and increased alanine transaminase (2.3%).
The CREST trial demonstrates that combining sasanlimab with both induction and maintenance BCG significantly prolongs EFS compared to BCG alone in patients with high-risk, BCG-naïve NMIBC. The combination appears tolerable, with manageable toxicity and no unexpected safety signals.
References
Top
