Diagnostic CT–guided online adaptive radiotherapy for locally advanced colon cancer: a prospective implementation and feasibility study

Purpose: This study aimed to systematically evaluate the geometric, dosimetric, and workflow characteristics of diagnostic CT-guided adaptive radiotherapy for locally advanced colon cancer (LACC), and to assess its feasibility and efficiency in routine clinical practice. Methods and materials: Forty patients undergoing short-course radiotherapy on a CT-linac within the TORCH-C protocol were included. ART was triggered when the predefined clinical judgment deemed target coverage on the anatomy of the day inadequate. Interfractional target variation was quantified using Dice similarity coefficient (DSC), Hausdorff distance (HD95), and average symmetric surface distance (ASSD). Dosimetric impact was assessed by recalculating the initial reference plan on the daily CT and comparing it with the adapted plan. In-vivo three-dimensional (3D) gamma verification was performed to evaluate the delivery accuracy. Workflow time and acute toxicity (CTCAE v5.0) were recorded for efficiency and safety analysis. Results: Among 200 fractions, 48 (24%) required adaptation. Adaptation was highly segment-dependent, occurring in 51.43% of transverse, 40% of ascending, 16% of descending, and 12% of sigmoid colon fractions. Interfractional instability was segment-dependent, with the largest geometric deviation in the transverse colon (median DSC = 0.70; HD95 = 37.9 mm; ASSD = 6.5 mm). Online ART significantly improved target coverage (PTV V100: 67.91% vs 98.25%) while reducing bowel exposure in the low-to-intermediate dose range. The median in-vivo gamma passing rates exceeded 95%. The median ART session duration was 24.6 min. Acute toxicities were mild (3 Grade 3, 1 Grade 2, 14 Grade 1, 22 Grade 0). Conclusions: Diagnostic CT-guided online ART for LACC is feasible, efficient, and safe within routine clinical throughput, and adaptation occurred in a substantial proportion of fractions, indicating that CT-guided ART can be selectively implemented to restore target coverage and reduce bowel dose in routine colon cancer radiotherapy.