Incomplete Toxicity Reporting Common in Phase III Oncology Trials
Key Insights
A review of 407 phase III oncology (search) trials revealed that only 32% reported complete toxicity data, including total adverse events (search), serious adverse events, deaths, and treatment discontinuations.
Toxicity-minimizing language was used in 46% of trials, subjectively downplaying the harm of therapies, regardless of sponsorship source.
Industry-sponsored trials demonstrated more comprehensive toxicity reporting compared to cooperative group-sponsored trials, highlighting the need for standardized reporting.
A comprehensive analysis of phase III oncology (search) trials has revealed significant gaps in toxicity reporting and a frequent use of language that minimizes the severity of adverse events (search). The study, published in JCO Oncology Practice, highlights the need for a more standardized approach to reporting toxicity data to improve understanding of new treatments.
The research team, led by Avital M. Miller et al., reviewed 407 two-arm superiority-design phase III oncology (search) trials published between 2002 and 2020, encompassing a total of 322,645 patients. The primary objective was to determine the prevalence of complete toxicity reporting (CTR) and the use of subjective toxicity-minimizing language (TML).
Key Findings on Toxicity Reporting
CTR was defined as the reporting of total adverse events (search) (TAEs), total serious adverse events (SAEs), total deaths, and study therapy discontinuations due to toxicity. Guideline concordance was assessed based on recommendations published in the BMJ, focusing on the reporting of total SAEs, total deaths, and study therapy discontinuations due to toxicity.
The analysis revealed that while a majority of trials reported SAEs (51%), total deaths (88%), and study therapy discontinuation due to toxicity (84%), only 32% (n = 131; 95% credible interval, 28 to 37) met the criteria for CTR. Notably, CTR was more prevalent in industry-sponsored trials (37%) compared to cooperative group-sponsored trials (4%).
Use of Toxicity-Minimizing Language
The study also assessed the use of TML, defined as terms that subjectively downplay the harm of therapies. TML was identified in 46% of the trials (n = 186; 95% credible interval, 41 to 51). Interestingly, no trial-related factors, including the source of sponsorship, were associated with the odds of using TML.
Implications for Clinical Practice
The authors emphasize that incomplete toxicity reporting and the use of TML can impede patients' and oncologists' understanding of the true risks associated with new treatments. Comprehensive and transparent reporting of adverse events (search) is crucial for informed decision-making and patient safety.
"Toxicity in phase III oncology (search) clinical trials is often incompletely reported and is frequently minimized in its interpretation," the authors stated. They further noted that industry-sponsored trials tend to report toxicity more comprehensively than cooperative group-sponsored trials.
Call for Standardization
The study underscores the urgent need for a more standardized approach to reporting toxicity data in oncology (search) trials. This would involve clear guidelines for reporting all relevant adverse events (search), including TAEs, SAEs, deaths, and treatment discontinuations, as well as avoiding the use of subjective language that minimizes the potential harm of therapies. By improving the completeness and transparency of toxicity reporting, clinicians and patients can make more informed decisions about treatment options, ultimately leading to better outcomes.