Analyses of Adverse Events and Patient-Reported Outcomes with the Toxicity Index in Cancer Clinical Trials.

BACKGROUND:Advancements in cancer therapies and improvements in survival have led to an increasing need to understand the impact of treatment on both severe and low-grade persisting symptoms and treatment-related toxicities. The National Cancer Institute launched the Cancer Moonshot Tolerability Consortium in 2018 to develop methods to analyze clinical and patient-reported tolerability data. METHODS:Through this initiative, the toxicity index (TI) was applied in the analysis of clinical trial tolerability data, and its performance was compared to other existing scoring methods. The TI is a tool developed to quantify the cumulative burden of all levels of clinician-graded toxicity data using Common Terminology Criteria for Adverse Events and patient-reported toxicity. This review describes the statistical properties of the TI and highlights key findings on the utility of the TI in summarizing tolerability data and its applications to existing clinical trial data. RESULTS:Compared to other commonly used summary measures, the TI was shown to be more efficient and increased the statistical power to detect treatment differences. The TI was also applied to existing toxicity and patient-reported symptom data from selected trials, where it was used to predict dose-limiting toxicities and treatment discontinuation. Visualization methods were explored and incorporated into interactive web-based research and patient-facing tools to enhance the interpretability and clinical utility of the TI. CONCLUSION:In conclusion, the TI is a useful summary measure that can be applied to a range of adverse event and patient-reported data to maximize the information collected and provide a more accurate and complete account of the tolerability experience. Future research will focus on extending applications of the TI and exploring its use for the design of clinical trials.