Study finds no significant survival benefit from participating in cancer clinical trials

By Pooja Toshniwal PahariaReviewed by Benedette Cuffari, M.Sc.May 23 2024

In a recent study published in JAMA, researchers determined whether patient involvement in cancer trials is associated with improved survival rates compared to regular therapy.

Study: Survival Benefit Associated With Participation in Clinical Trials of Anticancer Drugs A Systematic Review and Meta-analysis. Image Credit: Halfpoint / Shutterstock.com

Introduction

Researchers often regard clinical trials as delivering superior patient care due to their frequent assessments and increased access to novel medicines. However, no conclusive evidence of therapeutic benefit from study participation has emerged.

Cancer trial reviews from the 1990s have reported inconsistent evidence that patient involvement improves survival rates. Likewise, systematic literature reviews in different illness categories have not indicated that individuals participating in clinical trials had improved outcomes as compared to those receiving standard care.

About the study

In the current study, researchers investigate differences in overall survival rates for individuals enrolled in clinical trials as compared to standard care recipients, during which they considered confounding and biased sources.

Pooled estimates of overall survival basis were calculated for both standard care recipients and trial participants while considering trial eligibility criteria, treatment effects, and confounding variables. These estimates were then used to identify design characteristics related to detecting survival advantages among trial participants.

The treatment effect was defined as the effect of participating in trials on study outcomes regulated by intervention allocation. The participation effect was that of clinical trial participation unmediated by allotment to the intervention. Combining both treatment and participation effects yielded trial effects, which comprised differences in outcomes between regular care recipients and trial participants as measured by trial participation.

The primary outcome measure was the hazard ratio (HR) for the overall survival rates of trial participants as compared to usual care patients.

The Embase and PubMed databases were searched for trials published between January 1, 2000, and August 31, 2022,that compared the overall survival of trial participants to standard care recipients. Reference lists of identified records were also used to obtain additional data.

Studies using HRs to compare overall survival between the trial participant group and those receiving standard care and trials that treated cancer patients with pharmacological or biologic therapy were also included in the analysis. Studies that investigated indirect interventions or surgical procedures, letters, editorials, comments, and studies published in languages other than English were excluded from the analysis.

Two independent researchers performed title-abstract screening, followed by full-text screening, and resolved any disputes through discussion. The Risk of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool was used to assess the methodological quality of the included studies and pool data using random-effects modeling. I² statistics was also utilized to analyze study heterogeneity and funnel plots with the Egger test to detect publication bias.

A graph was generated to detect bias-related features that influenced participation effects and those that might influence estimate values. Eligibility criteria, scheduling, demographics, and medical history were among the quality features that distinguished study participants from usual care patients.

Study findings

Initially, the researchers identified 12,791 records, 3,094 of which were duplicates, and 9,523 that were excluded after title-abstract screening. After applying eligibility criteria, the researchers analyzed 39 studies, which included 85 comparisons between standard care recipients and trial participants. The median sample sizes for the trial participant and standard care patient groups were 209 and 409, respectively.

A significant advantage was observed in the overall survival of trial participants with an HR of 0.8 when all studies were considered, irrespective of quality or design, with high study heterogeneity. However, survival advantages were less among subsets matching trial participants with standard care recipients for eligibility criteria and were not present when only high-quality records were pooled, with HRs of 0.9 each.

The funnel plots and Egger test indicated a publication bias against studies lacking a participation effect. Subgroup analyses showed HR values of 0.9 and 0.7 for studies conducted in the United States and other nations, respectively.

Conclusions

Clinical trial participation in cancer research often leads to survival benefits, with evidence primarily obtained from studies that do not account for factors that could bias or confound these estimates. However, studies that accounted for eligibility and prognostic confounders demonstrate that these effects are significantly reduced.

High-quality records reduced heterogeneity between trial participants and routine care patients. Moreover, indirect statistical tests indicated that a disproportionate number of small studies suggest high survival benefits for trial participants, which is consistent with publication bias.