skip to main content

Bayesian Hierarchical Basket — SCE Medical Oncology MCQ

Instant feedback + full explanation. One question, done properly.

HardClinical Trials & StatisticsBayesian Hierarchical BasketSCE Medical Oncology

A 62-year-old woman is enrolled in an oncology basket trial testing a novel FGFR inhibitor across multiple FGFR-altered tumour types (cholangiocarcinoma, urothelial, breast, gastric). The trial uses a Bayesian hierarchical model for analysis. What is the advantage of this statistical approach in basket trials?

Educational content. Not a substitute for clinical judgement or local policy.

Reveal the answer and explanation

Correct answer: A — Bayesian hierarchical models allow 'information borrowing' between tumour-type cohorts — if the drug shows activity in cholangiocarcinoma and urothelial cancer, this information can strengthen the evidence for activity in the smaller breast and gastric cohorts, improving statistical efficiency for rare tumour types

Bayesian hierarchical models in basket trials allow evidence from one tumour-type cohort to inform (strengthen or weaken) conclusions in other cohorts sharing the same molecular target. If an FGFR inhibitor shows consistent activity across cholangiocarcinoma and urothelial cohorts, the model 'borrows' this information to provide stronger evidence for smaller cohorts (breast, gastric). This improves statistical efficiency for rare molecular subtypes where individual cohorts may be underpowered. The degree of borrowing is data-driven — if a cohort shows discordant results, borrowing is automatically reduced.

Reference: JRCPTB Curriculum; Berry et al JNCI 2006; Cunanan et al JCO 2017