HPB P04 The DISSECT trial, ‘The effect of periaDventitIal SMA diSsECTion on margin status during pancreaticoduodenectomy for resectable pancreatic cancer’ Results of safety and feasibility pilot study

Abstract Background Survival following pancreatic cancer resection remains poor despite advances in adjuvant and neoadjuvant therapy. The incidence of positive margin status (R1) is high and an R1 resection is associated with poorer survival. The most commonly positive margin is the SMA, accounting for between 47–77% of R1 resections. Periadventitial dissection of the SMA includes the lymphoneural tissue around the adventitia in addition to the ‘mesopancreas’ in the resection specimen and is used to varying extent across different centres. Theoretically this technique should reduce R1 resection rate, however, there is a lack of prospective evidence in the published literature. Methods The DISSECT trial is a single centre, randomised control trial. All patients with a resectable as per NCCN staging criteria head of pancreas tumour are randomised to either ‘periadventitial SMA dissection’ or ‘no periadventitial SMA dissection’ (1:1 block randomisation, sealed envelope system). A standardised technique for length and circumference of dissection is used and intra-operative photography with blinded assessment of dissection status ensures quality control. This safety and feasibility stage assesses recruitment, technique and complications and will be followed by a study powered to detect the primary outcome of reduction in positive SMA margin status. Primary outcomes to be assessed in the fully powered study will be the SMA margin status as detailed by the AJCC 8th edition. Secondary outcomes are status of other margins, length of stay, complications, overall and disease-free survival, complications, and adjuvant treatment details. These are assessed prospectively by an investigator blinded to the study arm. Results For the feasibility stage, the following criteria were set to proceed: recruitment of 20 patients in 6 months, at least 80% of recruited patients to undergo correct arm of randomisation, and an increase in complication rate of no more than 10%. Within the first 6 months, 32 patients were successfully recruited and randomised. 6 were excluded post-randomisation but before SMA approach (N=2 cancelled and not re-booked, N=2 abandoned after opening, N=2 converted to total pancreatectomy). N=3 were eligible but not recruited, (2 refused after approach and 1 was missed due to research team capacity). N=25 (96.2%) underwent their allocated study arm (confirmed by blinded photo assessment). For potentially procedure related complications (SMA bleed, post-operative haemorrhage, chyle leak, collection, and anastomotic leak) there was an overall rate of 38.5%, with no significant difference between the 2 arms (42.9% in No SMA dissection arm and 33.3% in the SMA dissection arm). There were no incidences of SMA branch bleed in the SMA dissection arm and a single post-operative haemorrhage in the SMA dissection arm that was due to a mesenteric venous bleed. Conclusions The DISSECT trial is a randomised controlled trial that aims to evaluate the use of a standardised technique of periadventitial SMA dissection to reduce the rate of R1 resection in early stages of pancreatic cancer. The feasibility and safety pilot is complete and demonstrates successful recruitment and randomisation, excellent adherence to randomisation arm and no operative concerns from the surgical team. The study will now proceed to full recruitment with complications and blinded photo reviews for quality control assessed contemporaneously. Primary and secondary outcomes will be recorded prospectively by a member of the research team blinded to randomisation arm.