Partial Identification of Personalized Treatment Response with Trial-reported Analyses of Binary Subgroups

Natural Experiments to Inform Clinical Practice

AbstractNatural experiments refer to events or practices that result in similar individuals receiving different services or interventions for arbitrary reasons. In the clinical context, researchers may wish to leverage natural experiments to estimate the causal impact of a particular treatment on a health outcome in situations where randomized clinical trial data are unavailable and other observational research designs are likely to yield biased results. This review provides an overview of natural experiments, discusses the potential for natural experiments to establish cause and effect, illustrates applications to specific clinical questions, and outlines situations and practices where natural experiments are most likely to answer the question at hand. Overall, while natural experiments have become popular in health policy, the widespread application of these approaches to specific clinical questions faces several challenges.

Cost-Effectiveness Analysis of Contemporary Advanced Prostate Cancer Treatment Sequences

There has been a proliferation of novel treatments for the management of advanced prostate cancer (PCa), including androgen receptor pathway inhibitors (ARPI). Although there are health economic analyses of novel PCa treatments, such as ARPIs for specific health states, there is a lack of sequential analyses. Our paper aims to fill this gap. We developed a Monte Carlo Markov model to simulate the management of advanced PCa to end-of-life. We modeled patients who begin in metastatic and nonmetastatic castration-sensitive PCa (mCSPC and nmCSPC), with risk stratification for mCSPC, progressing to metastatic castration-resistant PCa (mCRPC). Using current guidelines and recent literature, we simulated admissible treatment sequences over these states along a 15-year horizon. We report the best treatment sequences in terms of efficacy and cost-effectiveness. We find that the most cost-effective use of ARPIs is early in advanced PCa for a cost-effectiveness threshold (CET) of CAD 100K per QALY. For a CET of CAD 50K per QALY, early ARPI use is most cost-effective in mCSPC-starting patients but not nmCSPC-starting. We conclude that the most cost-effective way to use ARPIs is when patients first enter advanced PCa. The most cost-effective ARPI at current Canadian prices is abiraterone, mostly due to abiraterone's lower price level.

Using Limited Trial Evidence to Credibly Choose Treatment Dosage when Efficacy and Adverse Effects Weakly Increase with Dose

It has become standard in medical treatment to base dosage on evidence in randomized trials. Yet it has been rare to study how outcomes vary with dosage. In trials to obtain drug approval, the norm has been to compare some dose of a new drug with an established therapy or placebo. Standard trial analysis views each trial arm as qualitatively different, but it may be credible to assume that efficacy and adverse effects weakly increase with dosage. Optimization of patient care requires joint attention to both, as well as to treatment cost. This article develops a methodology to use limited trial evidence to choose dosage when efficacy and adverse effects weakly increase with dose. I suppose that dosage is an integer t ∊ (0,1,..., T ), T being a specified maximum dose. I study dosage choice when trial evidence on outcomes is available for only K dose levels, where K < T + 1. Then the population distribution of dose response is partially identified. I show that the identification region is a convex polygon. I characterize clinical and population decision-making using the minimax regret criterion. A simple analytical solution exists when T = 2. Computation is tractable when T is larger.