Background and study aims Scarred polyps are challenging to resect using conventional endoscopic mucosal resection (EMR) techniques. The aim of this pilot study was to assess the feasibility of the EndoRotor device in resecting scarred polyps arising from previous endoscopic resection attempts. Patients and methods This was a prospective pilot study of patients with scarred colonic polyps treated using EndoRotor carried out in two centers. Results A total of 19 patients were included in this study. The overall cure rate using EndoRotor was 84 %; 10 patients (52.6 %) achieved cure after one attempt and six patients (31.5 %) achieved cure after two attempts. A total of three patients who had polyp recurrence after the first EndoRotor resection were referred for either endoscopic submucosal dissection (2 patients) or surgery (1 patient) due to difficult access. There were no perforations, delayed bleeding, post-polypectomy syndrome or complications requiring surgery. Conclusions In this pilot study, the novel non-thermal device (EndoRotor) has been demonstrated to be a safe and effective technique in challenging management of scarred polyps. Further randomized controlled trials comparing this technique with APC, hot avulsion, ESD and endoscopic full-thickness resection are required to ascertain the utility of EndoRotor in the hands of non-expert endoscopists.

Introduction: We have previously reported our initial experience with prolonged initial chemotherapy with gemcitabine/abraxane followed by radiotherapy with concurrent infusional 5-FU. Continued application of this treatment approach has now resulted in 18 patients with unresectable pancreatic adenocarcinoma being resected (17 achieving R0 resections) with significantly prolonged survival times. Patients with tri-modality therapy showed a 45% four year survival. Methods: An electronic database search was carried out to identify all cases of locally advanced pancreatic cancer treated between 2 institutions. Case records, pathology, radiology and multidisciplinary team meeting records were then examined to determine type and dose of chemotherapy given together with radiological, pathological and survival outcomes. Patients were deemed unresectable if at multidisciplinary team meeting, they were shown to have vascular involvement >180 degrees and considered by the surgical, endoscopic ultrasound and radiological team to be not suitable for vascular reconstruction. Patients were then treated with up to 8 cycles of gemcitabine plus nab-paclitaxel followed by external beam radiotherapy 54Gy in 30 fractions. Follow-up with tumour markers and serial CT scanning was used to determine response and case records were examined for follow-up and survival data. Results: 89 patients were identified who fulfilled these criteria. 18 patients responded well enough to be deemed resectable at subsequent MDT meetings. These patients underwent Whipple's pancreatico duodenectomy. 3 pathological complete responses were seen and 17 of 18 patients achieved an R0 resection. Toxicity was related mainly to neuropathy from oxaliplatin and cytopenia. No treatment related deaths were seen. Medium length of stay following surgery was 18 days and there were no perioperative deaths. Median survival for those receiving or 3 modalities of therapy was greater than 2 years compared to less than 2 years for those not undergoing surgery and 10 months for those not undergoing surgery or proceeding with radiotherapy. Conclusion: Our data suggests that prolonged initial chemotherapy with up to 8 cycles of gemcitabine/nab-paclitaxel followed by radiotherapy with concurrent infusional 5-FU results in a significant number of patients being down staged from unresectable to resectable. Those undergoing tri-modality therapy have particularly impressive survival times.