Chemoradiotherapy for anal cancer: A population-based study of treatment interruption, treatment completion, and associated outcomes.

3573 Background: Chemoradiotherapy (CRT) is the standard treatment for squamous cell anal carcinoma (SCCA). Here, we describe CRT delivery in routine practice and explore the association between treatment interruption, non-completion and outcomes. Methods: The Ontario Cancer Registry was used to identify all incident cases of SCCA treated with curative intent RT in Ontario, Canada (2007-2015). Treatment interruption was defined a priori as > 7 days between consecutive fractions. Completed CRT was defined as receiving ≥50 Gy and 30+ fractions of RT along with 2 concurrent doses of chemotherapy. Log-binomial regression models were used to estimate risk ratios (RR) between patient characteristics and 1) failure to complete CRT and 2) salvage abdominoperineal resection (APR). Cox proportional hazard models were used to estimate hazards ratios (HR) between treatment interruption or non-completion and 1) cancer specific survival (CSS) and 2) overall survival (OS). Results: We identified 1593 patients with SCCA; 73% (n = 1161) initiated curative intent RT. Median RT dose and duration was 54 Gy (IQR, 50.4-67.8) and 46 days (IQR, 42-53), respectively. Treatment interruption > 7 days occurred in 23%. CRT was completed by 59%. Factors associated with CRT non-completion were age > 70 vs. < 50 (RR 0.70, 95% CI: 0.59-0.93) and greater comorbidity (1+ vs. 0, RR 0.57, 95% CI: 0.39-0.85). Treatment interruption > 7 days appears to be associated with salvage APR (RR 1.39, 95% CI: 0.91-2.12). In an exploratory analysis, the association between treatment interruption > 10 days and salvage APR reached statistical significance (RR 1.63, 95% CI, 1.06-2.53). Treatment interruption was not associated with inferior CSS (HR 0.87, 95% CI: 0.60-1.25) nor OS (HR 0.96, 95% CI: 0.76-1.23). Failure to complete CRT was not associated with higher rates of salvage APR nor inferior CSS, but was associated with inferior OS (HR 1.40, 95% CI: 1.13-1.72). Conclusions: In routine clinical practice, treatment interruption and non-completion among patients with SCCA are common. Quality improvement initiatives to optimize treatment continuity and completion are needed. The observed association between failure to complete CRT and OS is likely a reflection of confounding by indication, which is highlighted by the lack of association between CRT completion and CSS. Publisher's Note The abstract by Raphael et al entitled, “Chemoradiotherapy for anal cancer: A population-based study of treatment interruption, treatment completion, and associated outcomes,” published in the Journal of Clinical Oncology 37, no. 15 suppl (May 20 2019) 3573–3573, contained errors. In updated analyses, the authors discovered that the main exposure (radiation therapy) may have been incorrectly coded into the population-level databases from several individual treatment centers. Some of the coded radiation doses and fractionation numbers are considerably beyond what would be clinically plausible. Until this issue has been resolved, the authors believe the results of their study cannot be considered reliable. This article was retracted on 17-07-2019.