Up to 13.5% of patients that undergo a magnetic resonance imaging (MRI) scan of their abdomen without pancreatic symptoms are found to have an incidental pancreatic cyst, with the frequency increasing with age. In a post-mortem series, 25% of patients had a pancreatic cyst, of which 32% were potentially premalignant and 3% malignant. Premalignant cysts are currently either observed or removed surgically according to international guidelines. Observation is associated with significant anxiety for patients and a growing cost to the National Health Service, while surgery for this usually benign condition is associated with not insignificant morbidity and mortality. Premalignant pancreatic cysts may be indolent for a number of years before malignant transformation, creating a window of opportunity for minimally invasive intervention and cure. New early treatment options for premalignant tumours are urgently required. This study will evaluate the safety and efficacy of a novel minimally invasive technique for the treatment of pancreatic cystic tumours; endoscopic ultrasound guided radiofrequency ablation (EUSRFA).

If successful it will offer an alternative to long term observation or surgery for patients with this condition.

Pancreatic cystic tumours are an emerging and growing diagnostic challenge for the clinician. Due to the widespread use of high quality crosssectional imaging techniques, between 1.2 to 2.6% of patients that undergo a CT for a nonpancreatic indication are found to have an incidental pancreatic cystic tumour. Pancreatic cysts encompass a range of pathologies from benign to malignant. Premalignant pancreatic cysts are known to be indolent for a number of years before transforming into an invasive cancer, creating a window of opportunity for intervention and cure. There is a strong correlation between pancreatic cysts and age. In a recent series of patients with incidental pancreatic cystic tumours, no cases were found in patients less than 40 years of age, while 8.7 percent of the patients aged 80 to 89 years had a pancreatic cyst. The natural history of pancreatic cystic tumours remains poorly understood, so significant uncertainty exists regarding the appropriate diagnostic workup, how to differentiate between benign and malignant lesions and choice of therapy.

Characterisation of pancreatic cystic lesions is usually made by a combination of noninvasive cross sectional imaging and where necessary, endoscopic ultrasound guided fine needle aspiration (EUSFNA) of tissue and/or cyst fluid. In those undergoing EUSFNA, fluid aspirates for cytology are obtained in up to 80% of cases but samples frequently have a low cellularity so sensitivity for diagnostic cytology ranges between 31% and 85%. Pancreatic EUSFNA is a safe test, with complication rates (mainly pancreatitis, rarely infection and bleeding) occurring in approximately 12.4%. Our own published complication rates demonstrated a 0% mortality and 0.6% morbidity (mild acute pancreatitis).

To guide treatment a consensus document was published by the International Association of Pancreatology in 2006 and updated in 2012. The guidelines recommend the surgical resection of all pancreatic cysts that cause symptoms or are larger than 3 cm, contain a mural nodule or are associated with dilatation of the main pancreatic duct. If the cyst is less than 3cm and does not have any of these features the guidelines recommend that the lesions can be safely followed up with serial imaging. They recommended annual imaging for lesions <10mm, 6 monthly imaging for lesions 1020mm and 3 monthly imaging for lesions >20mm. How long patients should be followed up for or when they should be referred for surgical resection is unclear and is associated with a growing surveillance cost to the NHS.

Surgical resection of the pancreas is a considerable undertaking and is associated with a significant morbidity (21 to 59%) and mortality (07%). In addition, many elderly patients have significant comorbidity and will not be fit enough to undergo pancreatic surgery. To date, there have been only a few small studies using ablative techniques as an alternative to surgical resection. EUSguided injection of alcohol has been associated with cyst ablation rates of 35 to 62%, on short term follow up. However, this technique did not achieve total cyst ablation in cysts with septations and was associated with complication rates (pain and pancreatitis) of 4 to 20%. A major potential advantage of EUS guided RFA is that ablation of septated cystic tumours is feasible and that treatment effects can be monitored under realtime EUS control.

Radiofrequency ablation:

Radiofrequency ablation is achieved through a high frequency alternating current which generates high temperatures causing a coagulative necrosis. It has been used widely, percutaneously and intraoperatively, to treat primary and secondary cancers in the liver, lung, kidney, bone, pancreas and oesophagus. The Habib EUS RFA catheter is a single use sterile catheter for use during endoscopic ultrasound. It is an endoscopic monopolar catheter that has been designed to ablate cystic tumours of the pancreas and has European Conformity approval for this purpose. Following identification and puncture of the pancreatic cyst, the EUS RFA catheter can be introduced through a standard 19G EUSFNA needle. The catheter has an active metallic electrode which will be placed at the most distal part of the pancreatic cyst under EUSguidance. The catheter can be used with a variety of commercially available RF Generators, such as the RITA 1500, RITA 1500X, or ERBE VIO200D/300D. A dose of 10W for 90 seconds will be administered to each cystic tumour. EUSguided needle placement into pancreatic cysts is a standard technique, and will be performed by experienced (at least 5 years) endoscopists carrying out regular EUS interventions (at least 150 cases per year).

Animal studies of pancreatic RFA:

The safety and efficacy of the monopolar Habib EUS RFA catheter to be used in this study has been examined in the porcine model. The pancreas was visualized in all five cases and in four cases multiple punctures of the pancreas was possible. Each animal received up to 12 minutes of EUSguided RFA to the pancreas at a variety of currents (46W), catheter exposure (610mm) and length of time (0.25 minutes). Following the procedure, blood tests did not reveal any evidence of pancreatitis but at post mortem on day 7, three pigs had focal fat necrosis surrounding the pancreas. Analysis of the size of the area of ablation in the normal porcine pancreas found it to be proportionally related to the catheter length, power or time to which the tissue was exposed.

Inclusion Criteria:

1. A diagnosis of a pancreatic cystic tumour based on multidisciplinary review of imaging, for which further surveillance with non-invasive imaging is indicated.
2. Pancreatic cystic tumour between 0.5 and 3cm in size. Cysts greater than 3cm or with mural nodules can be included only if patients are unsuitable for surgical resection.
3. ECOG performance status 0, 1 or 2.
4. Estimated life expectancy of at least 12 weeks.
5. Age >18 years.
6. Capable of giving written informed consent.
7. Women of child-bearing potential must have a negative pregnancy test (qualitative serum hCG) in the week before treatment, AND be using an adequate contraception method, which must be continued for at least 1 week after RF.

Exclusion Criteria:

1. A diagnosis of a pancreatic cystic tumour where surgical resection is indicated.
2. Pancreatic cysts greater than 3cm or less than 0.5cm in size.
3. Benign pancreatic cysts (e.g. pseudocyst).
4. Serous cystadenomas.
5. Pancreatic cysts with malignant transformation.
6. Cysts involving or in close proximity to vessels or the biliary tree where the zone of ablation is likely to compromise these structures.
7. Cysts arising from the main pancreatic duct.
8. History of active or prior malignancy that will interfere with the response evaluation (exceptions include in-situ carcinoma of the cervix treated by cone-biopsy/resection, nonmetastatic basal and/or squamous cell carcinomas of the skin, any early stage (stage l) malignancy adequately resected for cure greater than 5 years previously).
9. Acute pancreatitis within the previous 4 weeks.
10. Any evidence of severe or uncontrolled systemic diseases or laboratory finding that in the view of the investigator makes it undesirable for the patient to participate in the trial.
11. Any psychiatric disorder making reliable informed consent impossible.
12. Pregnancy or breast-feeding.
13. ECOG performance status 3 or 4