Construction of a digital twin of chronic graft vs. host disease patients with standard of care.

ABSTRACT

There is an unmet medical need for new clinical trials to evaluate novel therapies in chronic graft-versus-host disease (cGvHD). Disease rarity, ethical issues regarding placebo arms, time, and cost impede clinical trial conduct. Digital twin (DT) technology enables virtual clinical trial arm construction using historical data, circumventing these obstacles. We evaluated the feasibility of constructing a DT trial arm using a large database of real-world clinical trial data and performed an efficacy assessment of a standard-of-care (SOC) drug to examine agreement with literature data. We constructed a flGvHD DT cohort (cGvHD patients at first-line treatment) (2042 patients; 32 cohorts) using the Trial Accelerator™ Digital Twin platform and derived an SOC arm from this cohort (flGvHD DT SOC cohort) (438 patients; eight cohorts); we analyzed the efficacy of SOC (prednisone) (overall response rate (ORR)) at six months. Our analysis results are in agreement with literature flGvHD DT disease onset time 7.58 months post-allogeneic hematopoietic cell transplantation; most used graft source peripheral blood stem cells; flGvHD DT SOC ORR at six months for prednisone 52.7%. It is feasible to construct a DT cohort using existing clinical trial data; a DT SOC arm can potentially replace a control arm in clinical trials.