Dosimetric Analysis of Major Wound Complications Following Preoperative Ultra-Hypofractionated Radiation Therapy for Soft Tissue Sarcoma

Abstract

Preoperative radiation therapy (RT) for soft tissue sarcomas (STS), delivered with conventional fractionation, has been shown to reduce long-term toxicity at the expense of increased postoperative major wound complications (MWC). Ultra-hypofractionated RT has emerged as a potential alternative preoperative modality with early but comparable outcomes to conventional regimens. However, limited data are available evaluating dosimetric, patient, and treatment specific factors associated with development of MWC in this setting. This IRB approved review included STS patients treated with preoperative 5 fraction daily RT followed by surgical resection within 7 days. Patients were evaluated for MWCs in association with patient and tumor characteristics, dosimetric parameters, and treatment techniques. MWCs were defined as a return to operating room, readmission for wound care or IV antibiotics, and persistent deep packing for >120 days. Prescription isodose line, PTV mean dose and the PTV volume exposed to 105% and 110% of prescribed dose were recorded. Dose to tissue likely to be involved in wound healing was assessed by creating a 1 cm thick superficial skin strip within 2 cm of the PTV which was then evaluated for volume, mean dose, V15, V21, V27, and V30. Secondary endpoints were locoregional control (LRC), metastasis free survival (MFS), and overall survival (OS). A total of 31 patients with a median age of 66 years (range 28-87) and a median follow up of 21 months (IQR 8-43) were included. All patients received 30 Gy in 5 fractions using IMRT/VMAT. There were 11 upper limb (36%) and 20 lower limb (65%) tumors included. Median time to resection following RT was 1 day (IQR 0-3). Median tumor size was 8 cm (IQR 5-13). MWC occurred in 13 patients (42%) with 10 patients (32%) requiring additional surgery. Dehiscence and infection requiring IV antibiotics occurred in 12 (39%) and 6 patients (19%), respectively. RT plans were predominately prescribed to the PTV mean (87%) with a median prescription isodose of 97% (IQR 96-97) and PTV mean dose of 3110 cGy (IQR 3089-3142). Median PTV volume, mean dose, and volume of PTV receiving 105% and 110% of the prescribed dose were higher in the MWC cohort although none reached significance. Similarly, for the 1 cm skin strip the median volume, mean dose, V30, V27, V21, and V15 were all higher in the MWC cohort without significant difference. Among patient, treatment, and tumor factors: tumor size, location, grade, margin status, type of wound closure, and prior non-oncologic resection were not associated with MWC. LRC, MFS, and OS at 3-years were 96%, 67%, and 76%, respectively. Although not reaching significance, increased plan homogeneity and reduced dose/volume relationships in proximity to the skin trended to reduced MWC in this limited cohort. Multi-institutional collaboration may be warranted to better identify factors associated with MWC in patients treated with preoperative ultra-hypofractionated RT.