BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: &quot\;Causal machine learning for biomarker subgroup
  discovery in randomised trials&quot\;.   - Paul Newcombe\, GlaxoSmithKlin
 e
DTSTART:20240123T140000Z
DTEND:20240123T150000Z
UID:TALK209761@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Decreasing costs of high-throughput ‘omics\, as well as new 
 technologies such as the Olink platform\, has driven wider application in 
 clinical trials\, for example to inform precision medicine strategies. How
 ever\, data-driven characterisation of patient subgroups with enhanced (or
  weaker) treatment effect remains a challenging problem\, particularly whe
 n searching over high-dimensional biomarkers. With growing recognition tha
 t traditional approaches (e.g. exhaustive biomarker-treatment interaction 
 testing) are sub-optimal\, several promising methods have recently emerged
  that combine machine learning tools with concepts from causal inference. 
 In principle\, they offer greater power through a combination of less cons
 ervative multiplicity control\, and the ability to capture complex multiva
 riate signatures which may be missed during one-at-a-time testing.\n\nI wi
 ll describe three causal machine learning methods for responder subgroup d
 etection\; the “Modified covariate Lasso”1\, “Causal Forests”2\, a
 nd the “X-Learner”3. I will compare and assess their performance in a 
 modest simulation study motivated by real biomarker trial datasets being g
 enerated in GSK. I will then share some (anonymised) results from on-going
  application of these methods to detect and predict responder subgroups fr
 om transcriptomic data measured in two Phase 3 Lupus trials. Finally\, I w
 ill close with a discussion on our experience of the benefits and limitati
 ons of existing approaches in this space.\n
LOCATION:MRC Biostatistics Unit\, East Forvie Building\, Forvie Site Robin
 son Way Cambridge CB2 0SR.
END:VEVENT
END:VCALENDAR