Saturday, November 23, 2024
11/23/2024
For decades, researchers have used the log-rank test/Cox’s hazard ratio (HR) test/estimation approach as the standard primary analysis for clinical trials with time-to-event outcomes. The statistical significance test based on the log-rank or HR-based test statistic offers a dichotomous decision about the superiority of one treatment over another to support the regulatory approval of new drugs or define the standard of care in a population. However, patients and clinicians often need additional information to make value-driven treatment decisions, including the magnitude of risks and benefits of the treatments they are considering. Despite its common use, the HR does not provide robust, reliable, and clinically interpretable quantitative information about the risks and benefits of a new treatment; new analytic methods are needed to inform patient-centered cancer treatment decisions. Our team recently developed a novel two-sample comparison method that has the potential to change the traditional analytic practice and significantly improve the interpretation of the magnitude of treatment effect on time-to-event outcomes. The new method utilizes the "average hazard" (AH) --- a new summary measure of the event time distribution, with three main advantages over alternative methods. First, the AH can be estimated nonparametrically and has a clear interpretation as the general Censoring-Free person-time Incidence Rate (gCFIR). Second, it provides treatment effect estimates in both absolute difference and relative terms, which increases the likelihood that clinicians, clinical trialists, and patients can understand the treatment effect magnitude correctly. Third, since the between-group treatment effect summaries provided by this method consist of "hazards", it would be more likely to be accepted in practice than existing alternative methods by clinical researchers accustomed to Cox's "hazard" ratio. The proposed study aims to establish the gCFIR approach for use as the primary analysis for treatment risk and benefit estimation and disseminate it to the clinical research community. In Aim 1, we will develop the following methods that are indispensable for this approach to be employed as the primary analysis in practice: 1) regression analysis, 2) interim analysis, 3) adjusted analysis, 4) analysis for competing risks, and 5) analysis for observational studies. In Aim 2, we will develop dissemination materials, including implementation software and a user-friendly guide with vignettes. We will provide clinicians and clinical trialists with free webinar courses. The overarching goal of this work is to use novel analytic methods to enhance communication and clinical interpretability of research findings, thereby facilitating translation of scientific results from clinical research into clinical practice. Ultimately, the proposed project can support informed conversations about treatment options between clinicians and patients.
HHS’ Tracking Accountability in Government Grants System (TAGGS) website provides access to detailed descriptions of grants, loans, aggregated direct payments and other types of financial assistance awarded by the HHS Operating Divisions (OPDIVs) and the Office of the Secretary Staff Divisions (STAFFDIVs).
