Disparities in positive results and dissemination of randomized controlled trials in immuno-oncology

Abstract

Background

This cross-sectional and longitudinal analysis aimed to demonstrate the disparities in positive results and dissemination patterns of randomized controlled trials (RCTs) in global immuno-oncology (IO).

Methods

Phase II–IV RCTs with results reported by article publications registered on ClinicalTrials.gov in 2007–2018 studying immune checkpoint inhibitors (ICIs), adoptive cell transfer, cancer vaccines, and immune modulators were included.

Results

Twenty-eight percent of trials were positive (72 of 258), most of which were pharma-sponsored and focused on ICI and multiple IO therapies in lung cancer, melanoma, and multiple cancer types. The recent period of trial start year, upfront registration, large sample size, high strictness score on corticosteroid/infection-related criteria, and survival endpoints were associated with positive results. Trials from Mainland China had a faster publication timeline of positive results but lacked study diversity or full reporting of negative results compared with US and multinational trials. Compared with phase II trials, phase III–IV trials had a higher average proportion of positive results (28.9% vs. 22.2%) and a more stable change over the past decade (23.65% vs. 49.24%). Positive trials yielded more secondary manuscripts (10 vs. 4), a shorter publication process of approximately two years (P < 0.001), and a superiority in the dissemination of journals with an h-index >90 (P < 0.001) compared with negative trials.

Conclusion

Disparities in positive result dissemination are widespread in IO RCTs and affected by trial features. We proposed improvements in upfront registration, procedural integrity, and adequate inclusion of rival trials reporting negative results within the earlier two years in future reviews.

PLAIN LANGUAGE SUMMARY

Immuno-oncology (IO) is a novel treatment modality utilizing the natural ability of body’s immune system to fight against cancer. The acknowledged standard method to confer the best medical evidence for showing the efficacy of a new intervention is randomized controlled trials (RCTs), and the publication of trial results via journal articles usually modifies medical decisions. A trial labeled negative means that the pre-specified goal was not met, but it deserves more attention rather than a simple interpretation of scientific failure. Previous studies on oncology trials indicated that negative and positive trials have different patterns of result publication and varied trial features. Although IO-related RCTs obtain a continuously increasing number, the extent and tendencies (positive or negative) of their results have not been assessed. To conduct a timely summary and a comprehensive analysis focusing on the publication details and its relationship with the properties of IO trials, we included phase II–IV IO RCTs with results reported by article publications registered on ClinicalTrials.gov in 2007–2018. We found that disparities in positive results and publication are widespread in IO RCTs and severely affected by IO category, cancer type, sponsorship, trial phase, and geographic origin. Positive trials had a significantly shorter publication timeline of approximately two years, more secondary manuscripts, and a superiority in high-quality publications over negative trials. We propose that investigators should complete registration before trial launch, improve procedural integrity, and allow an adequate inclusion of rival trials reporting negative results within the earlier two years in future IO-related reviews.

Keywords:

• Immuno-oncology
• immunotherapy
• randomized controlled trial
• positive result
• publication

Author contributions

Drs. Ma J and Huang Y had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs. Xu C, Zhang S, and Zhu GL contributed equally to the work as first authors.

Concept and design: Xu C and Zhang S.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Xu C, Huang Y, and Zhang S.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Xu C, Zhang Y, Zhu GL, Yang KB, and Liu Q.

Obtained funding: Ma J and Xu C.

Administrative, technical, or material support: All authors.

Supervision: Ma J and Huang Y.

Disclosure statement

The authors declared no potential conflicts of interest.

Fundings

This work was supported by grants from the Key-Area Research and Development Program of Guangdong Province (2019B020230002), the Natural Science Foundation of Guangdong Province (2017A030312003), the National Natural Science Foundation of China (82102828), and the Medical Science and Technology Research Fund of Guangdong Province (A2019418).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The original data are publicly available from the website of clinicaltrials.gov (https://clinicaltrials.gov).