Roche and Genentech's checkpoint inhibitor atezolizumab did not improve overall survival in a confirmatory phase III trial in patients with advanced bladder cancer. The FDA granted accelerated approval to the anti-programmed cell death protein 1 ligand 1 (PDL1) antibody for this indication in 2016 on the basis of improvement on the surrogate end point of objective tumour response rate, the percentage of patients who experienced complete or partial tumour shrinkage with treatment. The failure of this confirmatory trial now leaves a question mark over the future of this drug in this setting.

Analysts expect the bladder cancer market to generate more than US$2 billion by 2026 in the United States alone. Anti-PD1 and anti-PDL1 antibodies are currently earmarked to capture around $1.7 billion of that market (Nat. Rev. Drug Discov. 15, 599–600; 2016).

The jury is still out on whether this failure represents an opportunity or cause for concern for other checkpoint inhibitors. Bristol-Myers Squibb (BMS)'s anti-PD1 nivolumab, Merck KGaA and Pfizer's anti-PDL1 avelumab and AstraZeneca's anti-PDL1 durvalumab all have accelerated approvals in bladder cancer. (Durvalumab scored this accelerated approval in May, becoming the fifth PD1-modulating antibody to make it to market.) Merck & Co's anti-PD1 pembrolizumab received FDA approval for two bladder cancer indications in May.

Further analyses of results from confirmatory and phase III trials of these drugs will provide clarity on whether atezolizumab's failure was due to specific properties of the antibody, clinical trial design parameters or a more troublesome disconnect between the surrogate and clinical end points in this setting.

Last year BMS's nivolumab unexpectedly failed in a trial that sought to pave the way for the supplemental approval of the antibody in first-line non-small-cell lung cancer (NSCLC). Merck & Co. fared better in this setting with pembrolizumab, possibly because the company used a different approach to select PDL1-positive patients. In May the FDA approved pembrolizumab in combination with chemotherapy for first-line metastatic NSCLC, irrespective of PDL1 expression, adding to its earlier approval as a monotherapy in patients with high PDL1 expression.