Influence of timing of surgery on perioperative morbidity after neoadjuvant therapy for locally advanced rectal cancer.

Abstract

754 Background: Timing of surgery following completion of neoadjuvant therapy (NT) for locally advanced rectal cancer (LARC) has important implications for treatment response. However, it was recently reported in the GRECCAR 6 trial that delayed surgery beyond 8 weeks from completion of NT is associated with increased complications. Within a cohort of LARC patients treated with NT (CRT alone, Total NT (TNT) and chemotherapy alone) we examine perioperative complications based on time from NT to surgery. Methods: Patients with Stage II/III LARC ≤15cm from the anal verge who received NT from 06/01/09 – 03/01/15 were identified and preoperative morbidity collected on those undergoing rectal resection. Patients were grouped according to time of surgery from completion of NT (5-8 weeks – early surgery / 8-12 weeks – late surgery). Results: 798 patients were identified and 547 underwent rectal resection within 12 weeks of completing NT (440 LAR and 107 APR). Surgery was performed 5-8 following NT in 252 pts and 8-12 weeks following NT in 246 pts. 204 patients (41%) had a post-op complication: 53 (10%) Grade 3-5 complication and 83 (17%) SSI. There were no statistically significant differences in rates of all complications (44% vs 38%), grade 3-5 complications (9% vs 11%), SSI (17% vs 17%), and LOS (median 6 days vs 6 days) between the early and late surgery groups. Similar results were obtained when evaluating the subgroups by type of NT (CRT alone, chemo alone or TNT), surgical approach (open vs minimally invasive and sphincter preservation vs colostomy), post-treatment TNM stage and year of treatment (all NS). In addition, we did not observe differences in rates of downstaging responses: T downstaging (63% vs 64%), N downstaging (61% vs 54%), > 95% regression (34% vs 34%) or pCR rates (18% vs 18%) between the early and later surgery groups. Conclusions: In patients undergoing radical surgery for LARC post NT, we do not observe an effect of timing of surgery on surgical complications. Although timing of surgery is reported to influence response rates, we did not reproduce these findings, likely as a consequence of the high rate of deferral of surgery/ non-operative management in this cohort.