Skip to content

S3B

Novel Methods for Multiple and Hierarchical Outcomes in Clinical Trials

Chair:  David Yanez (Duke)

Speaker: Lu Mao, PhD (University of Wisconsin-Madison)
Title: On Restricted Mean Time in Favor of Treatment
Abstract: The restricted mean time in favor (RMT-IF) of treatment is a nonparametric effect size for complex life history data. It is defined as the net average time the treated spend in a more favorable state than the untreated over a prespecified time window. It generalizes the familiar restricted mean survival time (RMST) from the two-state life-death model to account for intermediate stages in disease progression. The overall estimand can be additively decomposed into stage-wise effects, with the standard RMST as a component. Alternate expressions of the overall and stage-wise estimands as integrals of the marginal survival functions for a sequence of landmark transitioning events allow them to be easily estimated by plug-in Kaplan-Meier estimators. The dynamic profile of the estimated treatment effects as a function of follow-up time can be visualized using a multilayer, cone-shaped “bouquet plot.” Simulation studies under realistic settings show that the RMT-IF meaningfully and accurately quantifies the treatment effect and outperforms traditional tests on time to the first event in statistical efficiency thanks to its fuller utilization of patient data. The new methods are illustrated on a colon cancer trial with relapse and death as outcomes and a cardiovascular trial with recurrent hospitalizations and death as outcomes. The R-package rmt implements the proposed methodology and is publicly available from the Comprehensive R Archive Network (CRAN).

Speaker: Roland Matsouaka, PhD (Duke)
Title: Trial Design with Win Ratio or Win Odds Based on Hierarchical Endpoints
Abstract: Win statistics, such as win ratio and win odds, have become a popular approach to analysis of hierarchical endpoints in clinical studies. While several sample size or power calculation formulas are available for the design of randomized trials using these methods, these formulas require investigators to specify clinically significant effect for win ratio or win odds, as well as the probability of ties. Usually, it is difficult to come up with these assumptions since prior studies published in clinical literature can have different prioritization of the outcomes, substantial drop out, different maximum follow up, or may be conducted in a different population compared to the study being planned. However, clinical investigators usually can provide clinically meaningful effect (that can be converted to win ratio or win odds) for each individual outcome used in the hierarchical endpoint. In this talk, we derive the relationship between the win ratio (or win odds) for the hierarchical endpoints with individual outcome win ratios (or win odds), under the assumption of independence of the individual outcomes. We also provide the expression for the probability of ties. We examine the impact of the assumption of independence via simulation. These new formulas can provide a way to specify meaningful and defensible magnitudes of win ratio (or win odds) for trial design; they can also be used to investigate the impact of adding outcomes to the hierarchical endpoint on the magnitude of the win ratio or win odds.

Speaker: Ying Cui, PhD (Stanford)
Title: Evidence Synthesis Analysis with Prioritized Benefit Outcomes in Oncology Clinical Trials
Abstract: Overall survival, progression-free survival, objective response/complete response, and duration of (complete) response are frequently used as the primary and secondary efficacy endpoints for designs and analyses of oncology clinical trials. However, these endpoints are typically analyzed separately. In this article, we introduce an evidence synthesis approach to prioritize the benefit outcomes by applying the generalized pairwise comparisons (GPC) method, and use win statistics (win ratio, win odds and net benefit) to quantify treatment benefit. Under the framework of GPC, the main advantage of this evidence synthesis approach is the ability to combine relevant outcomes of various types into a single summary statistic without relying on any parametric assumptions. It is particularly relevant since health authorities and the pharmaceutical industry are increasingly incorporating structured quantitative methodologies in their benefit-risk assessment. We apply this evidence synthesis approach to an oncology phase 3 study in first-line renal cell carcinoma to assess the overall effect of an investigational treatment by ranking the most clinically relevant endpoints in cancer drug development. This application and a simulation study demonstrate that the proposed approach can synthesize the evidence of treatment effect from multiple prioritized benefit outcomes, and has substantial advantage over conventional methods that analyze each individual endpoint separately. We also introduce a newly developed R package WINS for statistical inference based on win statistics.

Speaker: Rachel Marceau West , PhD (Merck)
Title: Time-to-Event Analysis Approaches for Randomized Clinical Trials with Composite Endpoints
Abstract: Composite endpoints are commonly used in randomized clinical trials. For example, the IMPROVE-IT trial of ezetimibe + simvastatin vs. placebo + simvastatin in 18,144 patients with acute coronary syndrome used a composite endpoint with five component outcomes: (1) cardiovascular death, (2) non-fatal stroke, (3) non-fatal myocardial infarction, (4) coronary revascularization ³ 30 days after randomization, and (5) unstable angina requiring hospitalization. In such settings, the standard analysis compares treatments using the observed time to occurrence of the first (i.e., earliest) component outcome for each patient. In this talk, we will use real data and simulations to contrast this approach with several alternative approaches that simultaneously use data for all the intra-patient component outcomes, not just the first. A potential enhancement of the recommended approach (Wei-Lachin method) using 5-STAR (objective risk-stratification) for each component will also be introduced. Issues related to study design, clinical interpretation and regulatory labeling will be highlighted.