Do results of the EORTC dummy run predict quality of radiotherapy delivered within multicentre clinical trials?

Objective: The European Organisation for the Research and Treatment of Cancer (EORTC) Radiation Oncology Group (ROG) has performed radiotherapy quality assurance (QA) in clinical trials, including dummy runs (DR) and individual case reviews (ICR), since 1991. We investigated the influence of DR results on subsequent QA and patient outcomes. Methods: EORTC ROG studies were reviewed for DR inclusion, QA and mature clinical outcomes. A DR was classified as a failure if corrections necessitated re-submission. ICR were graded as acceptable, minor or major deviation overall. Fisher's exact test characterised potential correlations and the Mantel-Haenszel statistic quantified pooled odds ratios (OR). Results: DR and ICR data were available from 12 and 3 protocols, respectively. The proportion of institutions successful at first DR attempt varied per trial from 5.6% to 68.8%. Participants were 3.2 times more likely to pass at first attempt after previous DR participation (p = 0.0002). Pooled OR for an acceptable ICR was 1.69 (p = 0.06) for institutions successful at DR first attempt. The effect of DR participation was not significantly correlated with patient outcome in the trial available for analysis. Conclusions: Implementing QA measures in ROG clinical trials should ensure optimal radiotherapy delivery. Centres which previously participated in a DR were significantly more likely to be successful at subsequent QA procedures.