Dataset Information

Assessment of Hazard Ratios in Oncology Clinical Trials Terminated Early for Superiority: A Systematic Review.

ABSTRACT:

Importance

Group sequential designs allow potential early trial termination at the interim analysis, before study completion. Traditional maximum likelihood estimate is commonly used to quantify the treatment effect in group sequential design trials; however, in published clinical trials, a bias-adjusted estimator has rarely been reported.

Objective

To emphasize the need for considering overestimation of treatment effect by applying 2 bias-adjusted estimators to previously published, early-terminated oncology clinical trials.

Evidence review

Trials published from 2013 to 2017 were identified by searching MEDLINE and Embase on February 23, 2018. This review was restricted to oncology clinical trials using group sequential designs with a single preplanned interim analysis as well as 2-arm randomized clinical trials that were subsequently stopped for efficacy reasons. Each article was independently reviewed by 3 biostatisticians during text screening, and differences in opinion were resolved by discussion. This report presents the unadjusted hazard ratio (HR) of an experimental arm to a reference arm and 2 bias-adjusted HRs calculated by using the conditional mean-adjusted estimator (CMAE) and weighted CMAE (WCMAE).

Findings

In total, 198 abstracts were screened for eligibility, of which, 19 eligible clinical trials were identified as applicable to the bias-adjusted estimators. Unadjusted HRs ranged from 0.203 (95% CI, 0.150-0.276) to 0.71 (95% CI, 0.60-0.84), number of events at the interim analysis from 58 to 540, and information time from 48% to 82%. In each study, the HRs adjusted by CMAE and WCMAE were higher than the unadjusted HR. Bias-adjusted estimates in large trials (243 and 414 events at the interim analysis) were similar to the unadjusted HR. However, in small trials (eg, with 58 events at the interim analysis), bias-adjusted estimates were highly disparate from the unadjusted HR. In trials with large treatment effects (eg, HRs of 0.20 and 0.22), the difference between unadjusted and bias-adjusted HRs was small even though the number of events at the interim analysis was small; larger differences were observed when the unadjusted HR was greater than 0.5.

Conclusions and relevance

In this systematic review of oncology clinical trials that were stopped for efficacy at the interim analysis, relatively large differences were noted between the unadjusted and adjusted HRs when the number of events at the interim analysis was small or when the unadjusted HR was close to the boundaries. These findings suggest presenting the 2 bias-adjusted HRs along with the unadjusted HR in the data monitoring committee meeting.

SUBMITTER: Shimura M

PROVIDER: S-EPMC7312398 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The objective of this study was to investigate the occurrence and determinants of non-publication of clinical drug trials in the Netherlands.All clinical drug trials reviewed by the 28 Institutional Review Boards (IRBs) in the Netherlands in 2007 were followed-up from approval to publication. Candidate determinants were the sponsor, phase, applicant, centers, therapeutic effect expected, type of trial, approval status of the drug(s), drug type, participant category, oncology or other disease area, prospective registration, and early termination. The main outcome was publication as peer reviewed article. The percentage of trials that were published, crude and adjusted odds ratio (OR), and 95% confidence interval (CI) were used to quantify the associations between determinants and publication. In 2007, 622 clinical drug trials were reviewed by IRBs in the Netherlands. By the end of follow-up, 19 of these were rejected by the IRB, another 19 never started inclusion, and 10 were still running. Of the 574 trials remaining in the analysis, 334 (58%) were published as peer-reviewed article. The multivariable logistic regression model identified the following determinants with a robust, statistically significant association with publication: phase 2 (60% published; adjusted OR 2.6, 95% CI 1.1-5.9), phase 3 (73% published; adjusted OR 4.1, 95% CI 1.7-10.0), and trials not belonging to phase 1-4 (60% published; adjusted OR 3.2, 95% CI 1.5 to 6.5) compared to phase 1 trials (35% published); trials with a company or investigator as applicant (63% published) compared to trials with a Contract Research Organization (CRO) as applicant (50% published; adjusted OR 1.7; 95% CI 1.1-2.8); and multicenter trials also conducted in other EU countries (68% published; adjusted OR 2.2, 95% CI 1.1-4.4) or also outside the European Union (72% published; adjusted OR 2.0, 95% CI 1.0-4.0) compared to single-center trials (45% published). Trials that were not prospectively registered (48% published) had a lower likelihood of publication compared to prospectively registered trials (75% published; adjusted OR 0.5, 95% CI 0.3-0.8), as well as trials that were terminated early (33% published) compared to trials that were completed as planned (64% published; adjusted OR 0.2, 95% CI 0.1-0.3). The non-publication rate of clinical trials seems to have improved compared to previous inception cohorts, but is still far from optimal, in particular among phase 1, single-center, not prospectively registered, and early terminated trials.

Project description:BackgroundClinical significance in a randomized controlled trial (RCT) can be determined using the minimal clinically important difference (MCID), which should inform the delta value used to determine sample size. The primary objective was to assess clinical significance in the pediatric oncology randomized controlled trial (RCT) treatment literature by evaluating: (1) the relationship between the treatment effect and the delta value as reported in the sample size calculation, and (2) the concordance between statistical and clinical significance. The secondary objective was to evaluate the reporting of methodological attributes related to clinical significance.MethodsRCTs of pediatric cancer treatments, where a sample size calculation with a delta value was reported or could be calculated, were systematically reviewed. MEDLINE, EMBASE, and the Cochrane Childhood Cancer Group Specialized Register through CENTRAL were searched from inception to July 2016.ResultsRCTs (77 overall; 11 and 66), representing 95 (13 and 82) randomized questions were included for non-inferiority and superiority RCTs (herein, respectively). The minority (22.1% overall; 76.9 and 13.4%) of randomized questions reported conclusions based on clinical significance, and only 4.2% (15.4 and 2.4%) explicitly based the delta value on the MCID. Over half (67.4% overall; 92.3 and 63.4%) reported a confidence interval or standard error for the primary outcome experimental and control values and 12.6% (46.2 and 7.3%) reported the treatment effect, respectively. Of the 47 randomized questions in superiority trials that reported statistically non-significant findings, 25.5% were possibly clinically significant. Of the 24 randomized questions in superiority trials that were statistically significant, only 8.3% were definitely clinically significant.ConclusionsA minority of RCTs in the pediatric oncology literature reported methodological attributes related to clinical significance and a notable portion of statistically insignificant studies were possibly clinically significance.