A study protocol for a comparative effectiveness evaluation of antiandrogenic medications against Standard of Care

TL;DR

This paper outlines a study protocol to evaluate the effectiveness of antiandrogenic medications versus standard care in metastatic prostate cancer patients using observational data and causal inference methods.

Contribution

It introduces a novel observational study design based on the Rubin Causal Model to compare NAMs with standard care in real-world data.

Findings

Design ensures covariate balance on observed variables

Use of rich registry data supports validity of comparisons

Sensitivity analyses are recommended for unobserved confounders

Abstract

This paper presents a study protocol for an effectiveness evaluation of antiandrogenic (NAM) medications (i.e. abiraterone acetate and enzalutamide) against standard of care in patients suffering from metastatic castrate resistant prostate cancer. The study protocol takes stock in the Rubin Causal Model for observational data in using historical comparisons for the analysis on three outcomes: mortality, pain and skeleton related events. We restrict the evaluation to 1,285 NAM patients with a prostate cancer diagnosis after June 2012 and find the comparisons in the group of men that had a prostate cancer diagnosis between June 2008 and June 2010. The rich registry data and the lack of NAMs as an option for the comparisons provide support for the validity of the design. However, while the design yields balance on the observed covariates, one cannot discard the possibility that unobserved confounders are not balanced. Thus, sensitivity analyses are suggested.