TY - JOUR

T1 - Beyond the Cox Hazard Ratio

T2 - A Targeted Learning Approach to Survival Analysis in a Cardiovascular Outcome Trial Application

AU - Chen, David

AU - Petersen, Maya L.

AU - Rytgaard, Helene Charlotte

AU - Gron, Randi

AU - Lange, Theis

AU - Rasmussen, Søren

AU - Pratley, Richard E.

AU - Marso, Steven P.

AU - Kvist, Kajsa

AU - Buse, John

AU - van der Laan, Mark J.

PY - 2023

Y1 - 2023

N2 - The Hazard Ratio (HR) is a well-established treatment effect measure in randomized trials involving right-censored time-to-events, and the Cardiovascular Outcome Trials (CVOTs) conducted since the FDA's 2008 guidance have indeed largely evaluated excess risk by estimating a Cox HR. On the other hand, the limitations of the Cox model and of the HR as a causal estimand are well known, and the FDA's updated 2020 CVOT guidance invites us to reassess this default approach to survival analyses. We highlight the shortcomings of Cox HR-based analyses and present an alternative following the causal roadmap-moving in a principled way from a counterfactual causal question to identifying a statistical estimand, and finally to targeted estimation in a large statistical model. We show in simulations the robustness of Targeted Maximum Likelihood Estimation (TMLE) to informative censoring and model misspecification and demonstrate a targeted learning analogue of the original Cox HR-based analysis of the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial. We discuss the potential reliability, interpretability, and efficiency gains to be had by updating our survival methods to incorporate the recent decades of advancements in formal causal frameworks and efficient nonparametricestimation.

AB - The Hazard Ratio (HR) is a well-established treatment effect measure in randomized trials involving right-censored time-to-events, and the Cardiovascular Outcome Trials (CVOTs) conducted since the FDA's 2008 guidance have indeed largely evaluated excess risk by estimating a Cox HR. On the other hand, the limitations of the Cox model and of the HR as a causal estimand are well known, and the FDA's updated 2020 CVOT guidance invites us to reassess this default approach to survival analyses. We highlight the shortcomings of Cox HR-based analyses and present an alternative following the causal roadmap-moving in a principled way from a counterfactual causal question to identifying a statistical estimand, and finally to targeted estimation in a large statistical model. We show in simulations the robustness of Targeted Maximum Likelihood Estimation (TMLE) to informative censoring and model misspecification and demonstrate a targeted learning analogue of the original Cox HR-based analysis of the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial. We discuss the potential reliability, interpretability, and efficiency gains to be had by updating our survival methods to incorporate the recent decades of advancements in formal causal frameworks and efficient nonparametricestimation.

KW - Causal roadmap

KW - LEADER

KW - Semiparametric efficiency

KW - TMLE

KW - CAUSAL INFERENCE

KW - REGRESSION

KW - MODELS

KW - BIAS

U2 - 10.1080/19466315.2023.2173644

DO - 10.1080/19466315.2023.2173644

M3 - Journal article

VL - 15

JO - Statistics in Biopharmaceutical Research

JF - Statistics in Biopharmaceutical Research

SN - 1946-6315

IS - 3

ER -