Toxicity and Surgical Complication Rates of Neoadjuvant Atezolizumab in Patients with Muscle-invasive Bladder Cancer Undergoing Radical Cystectomy: Updated Safety Results from the ABACUS Trial

[Background] There are limited data on toxicity and surgical safety associated with neoadjuvant programmed death ligand 1 (PD-L1) inhibitors prior to radical cystectomy (RC) in patients with muscle-invasive bladder cancer (MIBC).
[Objective] To present a comprehensive safety analysis of the largest neoadjuvant series, with focus on timing and severity of toxicity and surgical complications occurring after neoadjuvant atezolizumab in patients with MIBC enrolled in the ABACUS trial.
[Design, setting, and participants] ABACUS (NCT02662309) is an open-label, multicenter, phase II trial for patients with histologically confirmed (T2-T4aN0M0) MIBC, awaiting RC. Patients either were ineligible or refused cisplatin-based neoadjuvant chemotherapy.
[Intervention] Two cycles of neoadjuvant atezolizumab (1200 mg, every 3 wk) followed by RC.
[Outcome measurements and statistical analysis] Description of atezolizumab toxicity profile in the neoadjuvant setting, impact on surgery, and delayed immune-mediated adverse events (AEs) were assessed.
[Results and limitations] Ninety-five patients received treatment. Of them, 44% (42/95) had atezolizumab-related AEs during the neoadjuvant period (fatigue [20%], decreased appetite [6%], and transaminases increased [6%]). Treatment-related grade 3–5 AEs occurred in 11% (10/95) of patients during the study. Of the patients, 21% (20/95) received only one cycle of atezolizumab due to AEs; 92% (87/95) underwent RC. No surgery was delayed due to atezolizumab-related toxicities. Surgical complications occurred in 62% (54/87) of patients. Of these patients, 43% (37/87) and 20% (17/87) had minor (grade 1–2) and major (grade 3–5) complications, respectively. Thirteen of 87 (15%) patients had post-RC atezolizumab-related AEs, including adrenal insufficiency and transaminases increased. Three deaths occurred during the period of study-related interventions (one non–treatment-related aspiration pneumonia, one immune-related myocardial infarction, and one cardiogenic shock after RC). Not all surgical safety parameters were available.
[Conclusions] Two cycles of neoadjuvant atezolizumab are well tolerated and do not seem to impact surgical complication rates. Owing to the long half-life, AEs may occur in the postoperative period, including endocrine abnormalities requiring attention and intervention.
[Patient summary] Here, we report a comprehensive dataset of patients receiving neoadjuvant immune checkpoint inhibitors before radical cystectomy. Treatment with neoadjuvant atezolizumab is safe and does not seem to complicate surgery significantly.
Queen Mary University of London was the Sponsor of the study. Roche granted QMUL funding for the study. J. Bull and M. Jacobson also provided financial support for aspects of the biomarker analysis. We acknowledge Cancer Research UK, the UK Experimental Cancer Medicine Network, and La Roche-Hoffmann for funding.