The R.O.A.D. to precision medicine

TL;DR

This paper introduces a novel framework that transforms observational data into a randomized-like dataset, enabling the creation of personalized treatment recommendations with validated clinical utility, especially in oncology.

Contribution

It presents a new two-step method to correct unobserved confounding in observational data and trains optimal policy trees for personalized treatment decisions.

Findings

Outperformed expert recommendations in GIST treatment.

Identified patient subgroups in sarcoma data who may not need treatment.

Validated treatment recommendations in external cohorts.

Abstract

We propose a prognostic stratum matching framework that addresses the deficiencies of Randomized trial data subgroup analysis and transforms ObservAtional Data to be used as if they were randomized, thus paving the road for precision medicine. Our approach counters the effects of unobserved confounding in observational data by correcting the estimated probabilities of the outcome under a treatment through a novel two-step process. These probabilities are then used to train Optimal Policy Trees (OPTs), which are decision trees that optimally assign treatments to subgroups of patients based on their characteristics. This facilitates the creation of clinically intuitive treatment recommendations. We applied our framework to observational data of patients with gastrointestinal stromal tumors (GIST) and validated the OPTs in an external cohort using the sensitivity and specificity metrics. We show that these recommendations outperformed those of experts in GIST. We further applied the same framework to randomized clinical trial (RCT) data of patients with extremity sarcomas. Remarkably, despite the initial trial results suggesting that all patients should receive treatment, our framework, after addressing imbalances in patient distribution due to the trial's small sample size, identified through the OPTs a subset of patients with unique characteristics who may not require treatment. Again, we successfully validated our recommendations in an external cohort.