Updated Analysis Of Bowel Bag (Bb) Dose Volume Parameters And Acute Gastrointestinal (Gi) Toxicity During Preoperative (Preop) Radiation Therapy (Rt) For Retroperitoneal Sarcoma (Rps)

There are no reports correlating bowel dose and acute GI toxicity for pre-op RT for RPS. Previously, we examined a cohort of such patients (pts) and found very low Grade (Gr) 3 toxicity (3%) despite dose delivery to BB exceeding published constraints for GI and GYN cancers (V15<830 cc, V25<650 cc, V45<195 cc). The purpose of this study is to examine an additional cohort from another institution to further define dose-volume constraints to BB for RPS. From 2003-13, 56 pts with RPS underwent pre-op RT at two institutions (36 and 20 pt cohorts). Acute GI toxicity was assessed using weekly treatment visit notes, and scored using RTOG Acute Radiation Morbidity Scoring Criteria for upper (anorexia, nausea, pain) and lower (diarrhea) GI toxicity. BB was contoured per RTOG atlas guidelines. Relationships between clinical factors [tumor size and location, recurrent RPS, prior abdominal surgery, chemotherapy (CT)] and dose-volume metrics to toxicity were analyzed using paired t-tests, Wilcoxon rank-sum tests or Fisher’s exact tests. The association between BB volumes in intervals of 5 Gy (5 to 55 Gy) and toxicity (Gr 0-1 vs. Gr ≥2) were tested using the paired t-test. Receiver operating characteristics (ROC) were calculated for each 5 Gy interval. 31 pts (55%) were male. Median age was 61 years (range: 18-83). RPS was recurrent for 16 (29%). Median tumor size was 13.8 cm (range: 0.8-35.9). 36 tumors (64%) were above the pelvic brim, 13 (23%) below the brim, and 7 (13%) were both above and below the brim. Median dose was 50 Gy (range: 21.6 - 58.1); 34 pts (61%) were treated with 3D RT and 22 with IMRT (39%). 10 pts (18%) received induction or concurrent systemic therapy (4 concurrent); 31 pts (55%) had prior abdominal surgery. Median V15 for the BB was 1425 cc; 42 pts (75%) had V15 ≥830 cc. Median V25 was 1152 cc; 39 pts (70%) had V25 ≥650 cc. Median V45 was 580 cc; 36 (75%) pts had V45 ≥195 cc. Median maximum point dose to the BB was 53.6 Gy (range: 48.8-62.6 Gy). One pt (2%) had Gr 5 toxicity: she was 83 years-old and after 21.6 Gy, developed sepsis from a tumor/bowel fistula, stopped RT, and died 5 months later from sepsis/progression. Two pts (4%) had Gr 3 toxicity (nausea, anorexia); 36 (64%) had Gr 2; 11 (20%) had Gr 1; and 6 (11%) had no toxicity. Nausea was the most common Gr 2 toxicity, affecting 31 pts (55%); 13 (23%) had Gr 2 diarrhea and 11 (20%) had Gr 2 pain. There were no significant differences in Gr ≥2 toxicity between cohorts or based on clinical factors. Mean and median BB volume predicted for Gr ≥2 toxicity with statistical significance. On ROC analysis, V30 was the best discriminator for Gr ≥2 toxicity (highest area under the curve: 0.80; p <0.0001). Despite delivery of BB doses exceeding established constraints for >70% of pts, significant toxicity was very low. The lack of demonstrated dose-limiting toxicity at 50 Gy is encouraging given the interest in dose escalation for RPS.