Sequential curative surgery and outcome in a multicenter randomized phase III trial comparing conversion chemotherapy followed by chemoradiotherapy vs continuous chemotherapy in patients with unresectable pancreatic cancer (CONKO-007 trial).

e16458 Background: Secondary surgery is increasingly used in patients with unresectable pancreatic cancer. Objections have been raised about the potential harmful effect of preoperative radiotherapy on the surgical complication rate and a discussion on the benefits of major arterial and venous resection is being held. Methods: Within the CONKO-007 Trial an observational analysis of patients (pts.) with secondary surgery (S) was performed. The trial randomly assigned (1:1) pts with pancreatic cancer regarded as anatomically unresectable after initial chemotherapy (FOLFIRINOX or Gemcitabine, physicians’ choice) to Radiochemotherapy (RCT,50.4Gy+Gemcitabine) or further chemotherapy (CT) followed by second evaluation of resectability and surgical intervention at the discretion of the treatment center. The analysis consisted of a per-protocol approach. The endpoints of the present analyses were the rate of postoperative complications (non-hematological CTC-AE and Clavien-Dindo-CD), in-house mortality and survival. Results: This analysis included surgically treated 122 pts.; 60/167 following CT, 62/169 RCT of 336 randomized and previously 525 registered pts. 96/525 (18.2%) received curative intended surgery and 26 (4,9%) a surgical exploration. 11/122 pts. died closely to surgery (9.0%); 6 following RCT+S (d2-37, median d15), and 5 following CT+S (d 2-11, median d4). Cause of death after CT were related to impaired perfusion of the liver, lungs, colon; after RCT 3/6 had leakage or hemorrhage. CD-Grade 3/4: 10% (CT) vs. 11% (RCT; p = 1) and CTC-AE Grade 3/4: 10% vs. 6% (p = 0.52) did not differ. Median Overall Survival after resection did not differ by randomized treatment (p = 0.40), however by the extend of pancreatic resection (Pancreatoduodenectomy/distal pancreatectomy/central pancreatectomy vs. total pancreatectomy (23.0 vs. 13.0 mo., univariate, p = 0.01), the extend of resection of arterial vessels (TC, AMS, AH) vs. venous (VP; VMS) or no major vessel resection (10.0 vs. 23.0 mo.) multivariate, p = 0.002, HR: 1.42) and reaching R0 vs. R1/R2: (24.0 vs. 14.0 mo.) multivariate, p = 0.012; HR:1.97). Conclusions: Surgery was feasible in initial unresectable pancreatic cancer following RCT or CT in a multicenter approach with specific challenges in postoperative management. Survival was determined by extend of vessel resection and reaching R0. Clinical trial information: EudraCT Number: 2009-014476-21 .