The American Society of Clinical Oncology (ASCO) Value Framework and the European Society for Medical Oncology (ESMO) Magnitude of Clinical Benefit Scale (ESMO-MCBS) share weak correlations, suggesting different methods of measuring clinical benefit.
Additionally, quality-adjusted life-years (QALYs) and funding recommendations do not demonstrate a strong correlation with the frameworks.
While the ASCO Value Framework uses a scale to measure the clinical benefit and toxicity of a treatment regimen to calculate a net health benefit score, the ESMO-MCBS uses a structured and consistent approach to derive a ranking of the degree of clinically meaningful benefit that can be expected from a treatment, relative to other similar treatments. Whether the frameworks measure similar constructs of clinical benefit and their relations to QALYs and funding recommendations remain unclear.
Kelvin Kar-Wing Chan, MD, FRCPC, MSc, PhD, Sunnybrook Research Institute (Toronto, Canada), and colleagues conducted a literature review of 109 clinical trials of oncology drugs approved by the FDA, European Medicines Agency, and Health Canada between 2006 and 2015. Trials were scored using the ASCO version 1 framework (n = 108), ASCO version 2 framework (n = 111), and ESMO Magnitude of Clinical Benefit Scale (n = 83). Researchers utilized Spearman correlation coefficients to measure construct between frameworks.
The study was published in the Journal of Clinical Oncology (online June 6, 2017; doi:10.1200/JCO.2017.73.5704).
Results of the analysis showed weak-to-moderate correlation coefficients between the frameworks. The correlation coefficient was 0.36 between ASCO version 1 and ESMO-MCBS, 0.17 between ASCO version 2 and ESMO-MCBS, and 0.5 between ASCO version 1 and ASCO version 2. Sensitivity analyses conferred similar results.
After using the Spearman rank coefficient to measure criterion validity of the frameworks against QALYs and National Institute for Health and Care Excellence (NICE) and the pan-Canadian Oncology Drug Review (pCODR) recommendations, researchers found similar results. Among the 22 incremental QALYs from NICE reports, the correlation coefficients were 0.45 on ASCO version1, 0.53 on ASCO version 2, and 0.46 on ESMO-MCBS. Among the 50 incremental QALYs from pCODR reports, the correlation coefficients were 0.19 for ASCO version 1, 0.2 for ASCO version 2, and 0.36 for ESMO-MCBS.
Researchers noted that univariable logistic regression analysis conferred funding recommendations by NICE and pCODR were not associated with any of these scores.
“Revisions are recommended for both the ASCO and ESMO frameworks to reflect how scoring should be carried out in situations the current framework versions did not address explicitly, leaving room for inconsistent interpretation and scoring,” authors of the study concluded.—Zachary Bessette