Research methods to change clinical practice for patients with rare cancers

Lancet Oncol. 2016 Feb;17(2):e70-e80. doi: 10.1016/S1470-2045(15)00396-4.


Rare cancers are a growing group as a result of reclassification of common cancers by molecular markers. There is therefore an increasing need to identify methods to assess interventions that are sufficiently robust to potentially affect clinical practice in this setting. Methods advocated for clinical trials in rare diseases are not necessarily applicable in rare cancers. This Series paper describes research methods that are relevant for rare cancers in relation to the range of incidence levels. Strategies that maximise recruitment, minimise sample size, or maximise the usefulness of the evidence could enable the application of conventional clinical trial design to rare cancer populations. Alternative designs that address specific challenges for rare cancers with the aim of potentially changing clinical practice include Bayesian designs, uncontrolled n-of-1 trials, and umbrella and basket trials. Pragmatic solutions must be sought to enable some level of evidence-based health care for patients with rare cancers.

Publication types

  • Research Support, Non-U.S. Gov't
  • Review

MeSH terms

  • Bayes Theorem
  • Biomedical Research / methods*
  • Clinical Trials, Phase III as Topic
  • Humans
  • Neoplasms / therapy*
  • Observational Studies as Topic
  • Outcome Assessment, Health Care
  • Patient Selection
  • Rare Diseases / therapy*
  • Research Design*