Author: Gregory A Azzam, Danielle Cerbon, Laura Huang, Panayiotis Mavroidis, Diana Molinares, Stuart E Samuels, Sotirios Stathakis 👨🔬
Affiliation: Mary Bird Perkins Cancer Center, University of Miami Sylvester Comprehensive Cancer Center, Mount Vernan Rehabilitation Medicine Associates, Department of Radiation Oncology, University of Miami, University of North Carolina 🌍
Purpose: This study aims at performing a normal tissue complication probability (NTCP) modeling by fitting the doses to the skin, sternocleidomastoid (SCM) muscle and subcutaneous tissue (subcut) structures with the skin systemic sclerosis outcome data.
Methods: 88 patients with head & neck tumors were used in this study. The prescription doses were 60-70Gy. Scleroderma skin PRO (SSPRO) was recorded 12 months post-RT and a difference of ≥48 defined the clinical endpoint. The LKB NTCP model was used to fit the DVHs of the skin (total, left, right), SCM (total, left, right) and subcut (total, left, right) structures with the individual SSPRO scores. Metrics like area under the receiver operating characteristic curve (AUC) and odds ratio (OR) were used to evaluate the goodness-of-fit of the model.
Results: Skin systemic sclerosis was observed in 16.3% and 13.3% of the patients with and without prior surgery, respectively. The AUC values for skin, SCM and subcut were 0.65, 0.57 and 0.60 for the patients with prior surgery and 0.58, 0.75 and 0.68 for the patients without prior surgery, respectively. The model parameter values of the patients without prior surgery were TD50=40.2Gy, m=0.30 and n=0.15 for skin, 70.9Gy, 0.14 and 0.26 for SCM and 40.9Gy, 0.20 and 0.20 for subcut, respectively. Statistically significant gEUD cutoff doses with p-values of 0.03-0.05 were identified. Those doses of 30Gy, 60Gy and 30Gy for skin, SCM and subcut were associated with OR=9.3, 7.8 and 6.8, respectively.
Conclusion: Dose-response curves were derived for the skin, SCM and subcut structures for the endpoint of skin systemic sclerosis at 12 months post-RT. gEUD<30Gy, 60Gy and <30Gy to the skin, sternocleidomastoid muscle and subcutaneous tissue structures, respectively, were found to significantly reduce the risk of skin systemic sclerosis in the patients without prior surgery.