O-020Impact of surgical site experience on treatment outcomes of fixed-cT3 and cT4 rectal cancer patients in phase III study comparing preoperative radiochemotherapy and short-course radiotherapy with consolidation chemotherapy (Polish-II study)

Introduction: We have recently presented results of Polish II study comparing conventional radiochemotherapy with addition of weekly oxaliplatin (CRTx) with 5 Gy per day for 5 days short-course radiotherapy (SCRTx) combined with 3 cycles of consolidation chemotherapy (5FU/LV and oxaliplatin in FOLFOX4 regimen) in locally advanced rectal cancer patients (fixed cT3 or cT4 tumors). Although R0 rate, local control and relapse-free survival (RFS) were not different between study arms, we have observed a statistically significant 10% overall survival (OS) improvement after median follow up of 35 months for SCRTx arm (75% vs 65%, p = 0.046). The question is whether the experience of surgical center has an influence on outcomes of the study. Methods: 545 patients were randomized and 515 patients were eligible for analysis (254 and 261 patients in CRTx arm and SCRTx arm, respectively). Patients underwent surgery in 35 centers (varied from 1 to 114 patients per center). We divided the patients into two groups based on the experience of the center performing colorectal surgery. The number of surgeries during the study (SDS) classified a center as a high (HVC) and low (LVC) volume surgical center. 68% of the study population was operated on in HVC (25-114 SDS); 32% of patients underwent surgery in LVC (1-19 SDS). Results: We observed a trend for improved R0 resection rate in HVC (80% vs 72%; p = 0.09). There was a significant improvement in overall survival (OS) – 78% vs 64% patients were alive in SCRTx arm and CRTx arm after 3 years of follow-up, respectively (p < 0.05). This corresponded to 75% vs 60% of patients alive after 5 years of follow up (relative risk of death 0.625). There were no differences in R0 resection rates and OS in the subgroup of patients treated in LVC (p values 0.54 and 0.718, respectively). Similar results were obtained when surgical sites were divided by other cut-off points (proportion of HVC to LVC like 50%:50% or 80%:20% of the study cohort), ie. R0 resection rate was not worse for SCRTx (either trend or significant difference for higher rate of R0 resections only in HVC). For all tested cut offs we have not seen any differences between chemoradiation group in HVC and neither of treatment arms in LVC. Consequently, OS benefit for the experimental arm is only seen in HVC. Conclusion: The new treatment strategy combining the short-course radiotherapy and consolidation chemotherapy is effective and, in the experienced colorectal surgery centers, provides a substantial survival benefit over preoperative chemoradiation. Longer follow up is needed to confirm the observed phenomenon of higher survival rates without significant improvement in the local control.