Major ASCVD Trials Frequently Alter Endpoints From Initiation to Publication

Daniel Shepshelovich, MD

Credit: Columbia University

A new investigation identified notable heterogeneity in primary endpoints used for large atherosclerotic cardiovascular disease (ASCVD) randomized controlled trials (RCTs) in eminent medical journals, particularly from trial initiation to publication.¹

Investigators from Columbia University indicated these discrepancies in contemporary ASCVD RCTs could influence the reported results and complicate trials with similar interventions or data collection for meta-analyses.

“Regulators and journal editors should generally require that ASCVD RCTs use uniform endpoints and report the original protocol endpoints to generate higher-quality data and improve informed decision-making for patients with ASCVD,” wrote the investigative team, led by Daniel Shepshelovich, MD, division of general medicine at the Columbia University Irving Medical Center.

Shepshelovich and colleagues reviewed all ASCVD-related RCTs published between January 2019 and December 2023 in high-quality medical journals, including The New England Journal of Medicine (NEJM), The Lancet, JAMA, the European Heart Journal, Circulation, and the Journal of the American College of Cardiology (JACC). They aimed to determine the heterogeneity of primary endpoints and compare them with the original endpoint described at RCT initiation, given the lack of appraisal on the extent of heterogeneity in ASCVD endpoints.²

Those trials evaluating revascularizations and acute care were excluded, owing to differences in endpoints. Published endpoints were extracted from ClinicalTrials.gov and investigators assessed the effect of endpoint modification on the published results.¹

Most RCTs used time-to-event analysis and original data were typically unavailable, rendering an accurate comparison of trial results using different composite endpoints unfeasible. Instead, Shepshelovich and colleagues used a Fisher exact test to assess the cumulative incidence of various composite endpoints in both study cohorts.

Overall, 50 RCTS were involved in the analysis, with most (98%) using composite endpoints and the most common (28%) being 3-point major adverse cardiovascular events (MACE). The composite endpoint was comprised of nonfatal myocardial infarction, nonfatal stroke, and cardiovascular death.

All other RCTs (72%) involved 29 different endpoint combinations. Among 22 RCTs reporting other endpoints, the results of 2 trials included endpoint components not included in the 3-point composite MACE endpoint.

Approximately 23 RCTs (46%) had original primary endpoints different than the published endpoints. Across 15 trials (30%), components of the original composite endpoints were edited after patient enrollment. The analysis indicated these changes were reported by 11 publications (73%), with the most common reason (40%) being to boost statistical power.

Approximately 11 trials (73%) reported similar results using the original endpoint, but 4 (27%) did not report enough data for estimation, according to Shepshelovich and colleagues. Across 8 trials (16%), they noted the initial entry in ClinicalTrials.gov reported an undefined original endpoint and was later edited to include a more detailed definition.

Limitations, identified by Shepshelovich and colleagues, noted these results may not extend to trials not in the sample. Estimation of results using alternative composite endpoints could also be inaccurate, as primary data were unavailable. Overall, the team indicated this reported heterogeneity could lead to lower barriers to expanding endpoints, rather than increasing trial sample size or follow-up time when the event rate is lower than expected.

“Endpoint modifications between study initiation and publication compromise the trial’s scientific validity,” they wrote. “This is underlined by RCTs that reported positive results driven by endpoint components not included in the 3-point MACE."

References

1. _{Shepshelovich D, Yahav D, Rome DR, et al. Heterogeneity of Primary Outcomes in Large Atherosclerotic Cardiovascular Disease Trials Published in Prominent Medical Journals. JAMA Intern Med. Published online February 17, 2025. doi:10.1001/jamainternmed.2024.7694}
2. _{Armstrong PW, Westerhout CM. Composite End Points in Clinical Research: A Time for Reappraisal. Circulation. 2017;135(23):2299-2307. doi:10.1161/CIRCULATIONAHA.117.026229}