Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Two-stage single-arm trials are rarely analyzed effectively or reported adequately

Grayling, Michael J. and Mander, Adrian P. 2021. Two-stage single-arm trials are rarely analyzed effectively or reported adequately. JCO Precision Oncology 5 , pp. 1813-1820. 10.1200/PO.21.00276

Full text not available from this repository.

Abstract

PURPOSE Two-stage single-arm designs have historically been the most common design used in phase II oncology. They remain a mainstay today, particularly for trials in rare subgroups. Consequently, it is imperative such studies be designed, analyzed, and reported effectively. We comprehensively review such trials to examine whether this is the case. METHODS Oncology trials that used Simon's two-stage design over a 5-year period were identified and reviewed. They were evaluated for whether they reported sufficient design (eg, required sample size) and analysis (eg, CI) details. Articles that did not adjust their inference for the incorporation of an interim analysis were also reanalyzed. RESULTS Four-hundred twenty-five articles were included. Of these, just 47.5% provided the five components that ensure design reproducibility. Only 1.2% and 2.1% reported an adjusted point estimate or CI, respectively. Just 55.3% provided the final stage rejection bound, indicating many trials did not test a hypothesis for their primary outcome. Trial reanalyses suggested reported point estimates underestimated treatment effects and reported CIs were too narrow. CONCLUSION Key design details of two-stage single-arm trials are often unreported. Their inference is rarely performed such as to remove bias introduced by the interim analysis. These findings are particular alarming when considered against the growing trend in which nonrandomized trials make up a large proportion of all evidence on a treatment's effectiveness in a rare biomarker-defined patient subgroup. Future studies must improve the way they are analyzed and reported.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Medicine
Centre for Trials Research (CNTRR)
Publisher: American Society of Clinical Oncology
ISSN: 2473-4284
Date of First Compliant Deposit: 25 May 2022
Date of Acceptance: 9 November 2021
Last Modified: 01 Jun 2022 11:45
URI: https://orca.cardiff.ac.uk/id/eprint/150019

Actions (repository staff only)

Edit Item Edit Item