One of the hardest things I do in #oncology clinic is translate results from a clinical trial to the #patient in front of me. In @NatureMedicine, we introduce TrialTranslator, a #machinelearning framework for individualized #cancer trial emulation. nature.com/articles/s4159…
TLDR: By emulating 11 RCTs in >150K real-world patients, we show that survival results from trials generalize poorly to real-world pts, but in predictable ways. #ML framework and web interface @ to facilitate individualized estimation of tx benefit.trialtranslator.com
Dec 12, 2018 • 12 tweets • 4 min read
In our @JCO_ASCO piece out today, @VPplenarysesh and I argue that recent drug approvals based on metastasis-free survival (MFS) in #ProstateCancer illustrate the trade-off between speed and certainty in cancer drug approvals. Long thread to follow... ascopubs.org/doi/pdf/10.120…
Good treatments have dramatically improved life expectancy for patients with advanced #ProstateCancer. This is a good thing. But as life expectancy has increased, overall survival becomes a difficult endpoint to achieve in clinical trials, requiring lots of patients, time, & $$$