{"entity":{"id":"futility","kind":"term","name":"Futility analysis (stopped for futility)","aka":["stopped for futility","stopped early for futility","futility boundary","futility analysis","futility stopping","discontinued for futility","terminated for futility","stopped early for efficacy","stopped early","early stopping","stopping boundary","data monitoring committee","DMC","IDMC","DSMB","independent data monitoring committee","futility boundaries"],"tldr":"A planned check partway through a trial to see whether the new treatment has any realistic chance of proving better; if not, the trial is stopped to spare patients and money. The opposite rule stops a trial early when benefit is overwhelming.","summary":"Independent data monitoring committees review unblinded interim data against pre-specified boundaries: futility boundaries (often a conditional power below 10-20%) end trials such as TrilynX (xevinapant) and a string of phase 3 immunotherapy combinations; efficacy boundaries (O'Brien-Fleming or Lan-DeMets alpha spending) allow early positive readouts, as in KEYNOTE-006 and CheckMate 067. Stopping early for efficacy can overestimate effect size and truncate safety and survival follow-up, while futility stops sacrifice the chance of a late benefit. Sponsors also stop trials for 'business reasons' or poor accrual, which is not the same as futility.","asOf":"2026-09-09","wikipedia":"https://en.wikipedia.org/wiki/Interim_analysis","links":[{"label":"Wikipedia","url":"https://en.wikipedia.org/wiki/Interim_analysis"}],"tags":[],"related":["interim-analysis","statistical-significance","primary-endpoint","accrual","data-monitoring-committee","group-sequential-design","sample-size-re-estimation","trial-failure-modes"],"cancers":[],"sections":["drug-discovery"],"technologies":[],"targets":[],"drugs":[],"companies":[],"institutions":[],"pathways":[],"terms":[],"trials":["javelin-hn-100","trilynx","magnitude","dream3r"],"people":[],"bottlenecks":[],"keyPapers":[],"journals":[],"dependsOn":[],"notes":[],"category":"Trials"},"route":"/terms/futility/","neighbours":{"term":[{"id":"accrual","kind":"term","name":"Accrual and enrolment","route":"/terms/accrual/"},{"id":"data-monitoring-committee","kind":"term","name":"Data monitoring committee (DSMB, IDMC)","route":"/terms/data-monitoring-committee/"},{"id":"group-sequential-design","kind":"term","name":"Group sequential design, stopping rules and alpha spending","route":"/terms/group-sequential-design/"},{"id":"interim-analysis","kind":"term","name":"Interim analysis, readout and data cut-off","route":"/terms/interim-analysis/"},{"id":"primary-endpoint","kind":"term","name":"Primary, secondary and co-primary endpoints","route":"/terms/primary-endpoint/"},{"id":"response-adaptive-randomisation","kind":"term","name":"Response-adaptive randomisation","route":"/terms/response-adaptive-randomisation/"},{"id":"sample-size-re-estimation","kind":"term","name":"Statistical power, sample size and re-estimation","route":"/terms/sample-size-re-estimation/"},{"id":"statistical-significance","kind":"term","name":"Statistical significance (P values, alpha, multiplicity)","route":"/terms/statistical-significance/"},{"id":"trial-failure-modes","kind":"term","name":"Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover","route":"/terms/trial-failure-modes/"}],"section":[{"id":"drug-discovery","kind":"section","name":"Drug Discovery Platforms","route":"/fronts/drug-discovery/"}],"trial":[{"id":"dream3r","kind":"trial","name":"DREAM3R","route":"/trials/dream3r/"},{"id":"nct04447235","kind":"trial","name":"Ivermectin plus losartan for COVID-19 in cancer patients (ICESP phase 2, terminated)","route":"/trials/nct04447235/"},{"id":"javelin-hn-100","kind":"trial","name":"JAVELIN Head and Neck 100","route":"/trials/javelin-hn-100/"},{"id":"magnitude","kind":"trial","name":"MAGNITUDE","route":"/trials/magnitude/"},{"id":"trilynx","kind":"trial","name":"TrilynX","route":"/trials/trilynx/"}]}}