855 OPTIMIZATION OF SURGICAL DELAY AFTER INDUCTION CHEMORADIATION FOR ESOPHAGEAL CANCER: 10 YEARS RETROSPECTIVE MULTI-CENTER STUDY

The treatment of esophageal cancer is in constant evolution. Most of the esophageal cancer receive induction chemoradiation therapy. Surgical delay has been studied but the optimal timing has not been clarified. Through the years, surgical delay has been modified by surgeons in our institutions, going from an average of 6 weeks delay to an average of 10 weeks delay. It is time to ask if this change has a real positive impact on our patient. Methods In this retrospective multi-center study, we combined data from two center in Quebec city that performs oncologic esophagectomy. The surgical delay went from 6 to 10 weeks around 2014. All surgeons changed their practice at that moment. We retrospectively analysed 5 years before and after the change of practice and created two cohorts of patients. Our primary outcome compared complete pathologic response rate. Our secondary outcomes were surgical complications, anastomotic leak, disease free survival and overall survival. Results Thirty-eight patients had surgery under 8 weeks (mean: 6 weeks) after their induction chemoradiation compared to 64 patients that had surgery after 8 weeks (mean: 10 weeks). There was no statistical significant difference between groups for the complete pathologic response (32% vs 25%, p = 0,16). Important complications were similar, with a rate of 24% vs 28% (p = 0,69). Anastomotic leaks were less frequent in the less than 8 weeks group, but no statistical significance was obtained (13% vs 27%, p = 0,14).No difference in the disease-free survival rate and overall survival rate was noted (DFS 40% vs 55% (p = 0,32), OS 38% vs 38% (p = 0,29)). Conclusion The treatment of esophageal cancer is in constant evolution, induction therapy and surgical technics involve over time. Surgical delay has no impact on complete pathologic response, complication and overall survival. There is no advantage to wait longer before surgery.

854 OPTIMIZATION OF SURGICAL DELAY AFTER INDUCTION CHEMORADIATION FOR ESOPHAGEAL CANCER: 10 YEARS RETROSPECTIVE MULTI-CENTER STUDY

The treatment of esophageal cancer is in constant evolution. Most of the esophageal cancer receive induction chemoradiation therapy. Surgical delay has been studied but the optimal timing has not been clarified. Through the years, surgical delay has been modified by surgeons in our institutions, going from an average of 6 weeks delay to an average of 10 weeks delay. It is time to ask if this change has a real positive impact on our patient. Methods In this retrospective multi-center study, we combined data from two center in Quebec city that performs oncologic esophagectomy. The surgical delay went from 6 to 10 weeks around 2014. All surgeons changed their practice at that moment. We retrospectively analysed 5 years before and after the change of practice and created two cohorts of patients. Our primary outcome compared complete pathologic response rate. Our secondary outcomes were surgical complications, anastomotic leak, disease free survival and overall survival. Results Thirty-eight patients had surgery under 8 weeks (mean: 6 weeks) after their induction chemoradiation compared to 64 patients that had surgery after 8 weeks (mean: 10 weeks). There was no statistical significant difference between groups for the complete pathologic response (32% vs 25%, p = 0,16). Important complications were similar, with a rate of 24% vs 28% (p = 0,69). Anastomotic leaks were less frequent in the less than 8 weeks group, but no statistical significance was obtained (13% vs 27%, p = 0,14). No difference in disease-free survival rate and overall survival rate was noted (DFS 40% vs 55% (p = 0,32), OS 38% vs 38% (p = 0,29)). Conclusion The treatment of esophageal cancer is in constant evolution, induction therapy and surgical technics involve over time. Surgical delay has no impact on complete pathologic response, complication and overall survival. There is no advantage to wait longer before surgery.

Retrospective analysis of multiparametric MRI to predict complete pathologic response after neo-adjuvant chemotherapy for muscle invasive bladder cancer.

e16535 Background: Muscle invasive bladder cancer (MIBC) is a life-threatening disease. Treatment is multimodal combining neo-adjuvant cisplatin-based chemotherapy (NAC) and radical cystectomy (RC). Accurately predicting complete pathologic response (pCR) using multiparametric MRI (mpMRI) could impact peri-operatively treatment. Methods: MIBC patients receiving NAC were evaluated at our institution with mpMRI before, after 2 cycles and after 4 cycles of dose dense MVAC (ddMVAC). Response after 4 cycles was retrospectively assessed using the method earlier described by Necchi et al with blinding of the readers for the pathological result of RC. Two radiologists independently evaluated 3 questions: residual disease at T1/T2-weighted images, presence of spots of restrictive diffusion within the bladder wall on diffusion-weighted imaging, and presence of focal contrast enhancement in the bladder wall on dynamic contrast enhanced images. Radiographic complete response (rCR) was defined as “No” on all three questions. Results: A total of 46 patients were identified having received ddMVAC for urothelial MIBC. Six patients did not undergo RC after NAC and were excluded from this analysis. Eleven out of 40 (28%) patients showed a complete pathologic response (ypT0). Baseline characteristics were similar compared to non-complete pathologic responders ( > ypT0), with the exception of hydronephrosis (9% in ypT0 vs. 52% in > ypT0). mpMRI questions could be assessed in 37 of 40 patients (93%). rCR was seen in 5 patients and was significantly associated with pCR (1-sided p value 0.021). Although sensitivity was low (36%), specificity was very high (96%) of this 3-step assessment. Positive likelihood ratio was 9.45, negative likelihood ratio 0.66. Concordance of assessment was very high. Conclusions: Using the 3-step imaging approach of Necchi et al, mpMRI can predict pCR after neo-adjuvant cisplatin-based chemotherapy with high specificity but low sensitivity. mpMRI should be included in future trials of multimodal management of MIBC and is an important predictive asset in routine clinical management.[Table: see text]