Systems and methods for designing augmented randomized trials

A non-transitory machine readable medium containing processor instructions for clinical trial analysis using a set of computing devices wherein each computing device comprises a set of processors, where execution of the instructions by the set of processors causes the set of processors to perform a process that comprises:

• receiving, at a set of one or more computing devices, each computing device comprising a set of one or more processors and a memory, a set of outcomes at a plurality of points in time for a set of samples in an observed randomized control trial (RCT) including untreated human subjects, and a set of characteristics for each of the set of samples in the observed RCT;
• generating, using a set of generative models implemented by the set of one or more computing devices, a set of prognostic scores for the set of samples, wherein the set of characteristics for each sample in the set of samples in the observed RCT are used as inputs to the set of generative models, wherein the set of prognostic scores comprises prognostic scores at each of the plurality of points in time for each untreated human subject in the observed RCT, wherein the set of prognostic scores comprises predicted outcomes for the untreated human subjects from the observed RCT;
• assessing, using the set of one or more computing devices, discrimination and bias metrics for the set of generative models based on a correlation between the set of prognostic scores and the set of outcomes for the set of samples, wherein the set of outcomes comprises outcomes at each of the plurality of points in time for each sample of the set of samples;
• determining, using the set of one or more computing devices, a set of target trial parameters for a target randomized control trial (RCT) including human subjects for treatment based on the assessed discrimination and bias metrics;
• receiving, at the set of one or more computing devices, a set of characteristics and result data for each of a set of human subjects for treatment in the target RCT:
• providing, to the set of generative models implemented by the set of one or more computing devices, the set of characteristics for each of the set of human subjects for treatment in the target RCT as inputs;
• generating, using the set of one or more computing devices, result data for the target RCT using the set of generative models using the set of characteristics for each of the set of human subjects for treatment in the target RCT as inputs, wherein the result data comprises predicted outcomes for a set of control-arm digital twins having the characteristics of the human subjects for treatment from the target RCT;
• and
• determining, using the set of one or more computing devices, treatment effects for the target RCT using the generated result data and the result data of each of the set of human subjects for treatment in the target RCT.