Evidax Insights
Patient-level external data can sometimes augment a randomised control group. The external patients must answer the same question as the trial, and the amount borrowed must fall when the evidence conflicts. Here is a field guide.
Minimising the sample size required for a pivotal rare disease trial is a challenge every orphan drug developer faces. Meta-analytic predictive (MAP) priors can use historical trials to reduce the number assigned to control, but the amount borrowed depends on the evidence and agreement with the new trial.
External evidence can make small clinical trials more feasible, but only when the data are relevant, high quality and supported by robust analysis. Professor James Wason explains why, when and how to borrow information.
Inavolisib improved survival against one existing treatment. But standard care in Scotland involved three, and SMC did not recommend it for reimbursement. This case shows why one trial comparator may not be enough.
Two medicines supported by single-arm trials received opposite regulatory decisions on the same day. Fayuvi was approved and Xervyteg was recommended for refusal. The lesson is that external evidence must show what would have happened without the medicine.
A meta-analysis gives an effect estimate. GRADE helps decision-makers judge how much confidence to place in it, one outcome at a time.
Three EU JCA dossiers were discontinued after important evidence remained missing. Here is what they teach us about making comparative evidence assessable.
NICE did not accept an uncertain indirect comparison as proof of equal efficacy. It still recommended dostarlimab. Here is what evidence teams can learn from the decision.