Crossover in trials
When patients in a trial's control arm are allowed to switch to the experimental drug after their cancer progresses. It is fair to patients but blurs the survival comparison, because the control group has now had the drug too.
Crossover is common in trials of drugs already approved elsewhere or clearly active (imatinib in GIST, osimertinib in FLAURA, 177Lu-PSMA in VISION and PSMAfore, where 84% of controls crossed over) and explains why many trials show large progression-free survival gains but little or no overall survival difference. Statistical corrections (rank-preserving structural failure time, inverse probability of censoring weighting) estimate what survival would have been without crossover but rely on assumptions and are treated cautiously by regulators, who may still demand an OS trend. In regulatory debates, crossover cuts both ways: it can hide a true survival benefit or excuse its absence.
Pages like this
not linked directly; found by shared links- PersonMichael J. Morris
- IdeaSeamless phase 2/3 with pre-registered go rules as the default for new agents
Shares Overall survival (OS), Progression-free survival (PFS).
- Key paperVISION: lutetium-177 PSMA-617 radioligand therapy extends survival in advanced prostate cancer
Shares PSMAfore, VISION, Overall survival (OS), Progression-free survival (PFS).
- IdeaReport time toxicity, the days a treatment consumes, in every trial and decision aid
Shares Overall survival (OS), Progression-free survival (PFS).
- IdeaReport time toxicity, the days spent in healthcare, as a standard outcome for older patients
Shares Overall survival (OS), Progression-free survival (PFS).
- TermPrimary, secondary and co-primary endpoints
Shares Surrogate endpoint, Overall survival (OS), Progression-free survival (PFS).
- IdeaProvisional prices for surrogate-endpoint approvals, reset when survival data arrive
Shares Overall survival (OS), Progression-free survival (PFS).
- TermmCRPC and mHSPC (castration-resistant vs hormone-sensitive prostate cancer)