Effect of accurate heart delineation on cardiac dose during the CONVERT trial

Objective: This study investigated the effect of deviation from protocol in heart delineation for the Concurrent Oncedaily Versus twice-daily Radiotherapy (CONVERT) Trial (ClinicalTrials.gov Identifier NCT00433563) quality assurance (QA) programme and the effect of that on mean cardiac dose and percentage of heart volume receiving ≥5Gy (V_5%) and percentage of heart volume receiving ≥30Gy (V_30%). Methods: Patients with limited-stage small-cell lung cancer in the CONVERT trial were randomized to receive 45Gy in 30# twice daily or 66Gy in 33# once daily radiotherapy, with concurrent chemotherapy in both arms. Of the 100 trial patient cases reviewed by the QA team [for patient selection, disease/organs at risk (OARs) outlining and treatment planning], 50 patient cases were selected, and the heart was reoutlined according to the "gold standard" trial protocol. Dose-volume histogram (DVH) data were extracted. The impact of change in heart volume on cardiac DVH is presented. Results: For gold standard cardiac outlines, an increase in V_5% and V_30% was seen in 77.3% and 81.8% of cases, respectively, in the control arm (45 Gy) and in 78.6% and 82.1% of cases, respectively, in the experimental arm (66 Gy). The median increase in V_5% was 1.4% and in V_30%, it was 3.4%. The average of the mean cardiac dose for the control arm increased by 2.1Gy (from 15.2Gy for the centres outline to 17.3Gy for the gold standard outline) and for the experimental arm by 2.4Gy (from 16.2 to 18.6 Gy). Conclusion: This study demonstrates the importance of having a robust QA programme in place to ensure accuracy of cardiac delineation to facilitate future studies investigating the impact of cardiac dose on toxicity. Advances in knowledge: This study highlights the importance of ensuring accurate delineation of OARs in clinical trials, where conclusions about normal tissue toxicity are to be drawn.