181 Background: Several reports have linked noncompliance with radiotherapy (RT) protocol guidelines with inferior clinical outcomes. Here we perform a meta-analysis of prospective cooperative group trials to determine the impact of RT quality assurance (QA) deviations on disease control and overall survival (OS). Methods: We searched MEDLINE and the Cochrane Central Register of Controlled Trials for multi-institutional trials that reported clinical outcomes in relation to RT quality assurance (QA) results. Hazard ratios (HRs) describing the impact of RT protocol noncompliance on outcomes were extracted directly from the original studies or calculated from survival curves. Analyses were performed to assess the impact of RT QA deviations on OS and secondary outcomes (local/locoregional control, event-free survival, relapse), which were grouped together. Pooled estimates were obtained using the inverse variance method. A random effects model was used in cases of significant effect heterogeneity (p<0.10 using Q test). Results: Eight studies met all inclusion criteria and were incorporated into this analysis. Four were pediatric trials (POG 8346, SFOP.TC 94, POG 9031, SIOP/UKCC PNET-3), and four studied adult patients (RTOG 73-01, SWOG 7628, TROG 02.02, RTOG 97-04). Six of these trials reported the impact of RT QA deviations on overall survival, and six described the effects of RT QA deviations on secondary endpoints. The frequency of RT QA deviations ranged from 8% to 71% (median: 40%). Using a random effects model, RT deviations were associated with a significant decrease in OS (HR = 1.74, 95% CI: 1.28 to 2.35, p<0.001). A similar effect was seen for secondary endpoints (HR = 1.79, 95% CI: 1.15 to 2.78, p=0.009). No evidence of publication bias was detected using the Egger test (p=0.361 for OS, p=0.468 for secondary endpoints). Conclusions: In clinical trials, RT protocol deviations are associated with increased risk of treatment failure and overall mortality. The magnitude of these effects demonstrates that RT QA results should be considered in the interpretation of clinical trial results. More importantly, the delivery of high-quality RT is critical for the successful treatment of cancer patients.
Modeling overdispersed longitudinal binary data using a combined beta and normal random-effects model
