Ex vivo drug screening and clustering of bladder cancers for pre-clinical treatment prediction.

backgroundBladder cancer (BC) is the tenth most common cancer and the ninth leading cause of cancer death worldwide. BC has high rates of treatment failure, so alternate approaches are needed to personalise treatments to individual patients in order to improve outcomes from this disease. One method that could provide actionable results to influence clinical decisions on alternative treatments is ex vivo drug screening.

methodsWe explored the feasibility of using ex vivo drug screening directly on patient tumour tissue from transurethral resection of the bladder tumours (TURBT) and cystectomies. We screened 38 BC patients investigating drug sensitivities to 15 agents, including standard of care treatments and some more exploratory compounds. In addition, we investigated ex vivo sensitivity and resistance over the 15 compounds and annotated common mutational profiles. We saw high methodological success (41/54 samples, 75.9%), in clinically useful timeframes (4 days) and identified distinct drug and tumour clusters.

resultsHere, we show that drug resistance is associated with aggressive clinical features, mutation burden, and differs with individual gene mutations. Cross-resistance between agents is common. Cisplatin-resistant tumours differ by mutational profiles and include those with multi-drug resistance and those sensitive to alternative agents. Observed clinical responses match our ex vivo response (5/6 patients, 83.3%). Proliferative responses are observed to some receptor tyrosine kinase inhibitors, cautioning against their unselected widespread use.

conclusionsEx vivo drug screening identifies drug clusters of patients' tumours that could potentially respond to standard of care and alternative therapies. Our approach offers a platform to potentially individualise treatments, especially in drug-resistant tumours.