Current status of interventional radiology in China

Cardiovasc Intervent Radiol DOI 10.1007/s00270-014-1005-z REVIEW Current Status of Interventional Radiology in the Management of Gastro-Entero-Pancreatic Neuroendocrine Tumours (GEP-NETs) Gianluigi Orgera • Miltiadis Krokidis • Matteo Cappucci Sofia Gourtsoyianni • Marcello Andrea Tipaldi • Adam Hatzidakis • Alberto Rebonato • Michele Rossi • Received: 21 April 2014 / Accepted: 18 August 2014 Ó Springer Science+Business Media New York and the Cardiovascular and Interventional Radiological Society of Europe (CIRSE) 2014 Abstract Within the group of Gastro-Entero-Pancreatic Neuroendocrine tumours (GEP-NETs), several heterogeneous malignancies are included with a variety of clinical manifestations and imaging characteristics. Often these cases are inoperable and minimal invasive treatment offered by image-guided procedures appears to be the only option. Interventional radiology offers a valid solution in the management of primary and metastatic GEP-NETs. The purpose of this review article is to describe the current status of the role of Interventional Radiology in the management of GEP-NETs. Keywords Interventional oncology
Embolization
Ablation
Neuroendocrine tumours G. Orgera
M. Cappucci
M. A. Tipaldi
M. Rossi Unit of Interventional Radiology, S. Andrea University Hospital ‘‘Sapienza’’, Rome, Italy M. Krokidis (&) Department of Radiology, Cambridge University Hospitals NHS Trust, Hills Road, Cambridge CB2 0QQ, UK e-mail: mkrokidis@hotmail.com S. Gourtsoyianni Department of Radiology, Guy’s and St Thomas’ NHS Foundation Trust, London, UK A. Hatzidakis Department of Radiology, Medical School of Crete, University Hospital of Heraklion, Heraklion, Greece A. Rebonato Department of Surgical and Biomedical Sciences, Santa Maria della Misericordia University Hospital, University of Perugia, Perugia, Italy Introduction Gastro-Entero-Pancreatic Neuroendocrine tumours (GEP-NETs) are a vast and heterogeneous group of neoplasms that may involve organs of the foregut (oesophagus, stomach, proximal duodenum, liver, pancreas), the midgut (distal duodenum, ileum, jejunum, ascending colon and proximal 2/3 of the transverse colon) and the hindgut (distal 1/3 of the transverse colon, descending and sigmoid colon and rectum). Their incidence is raised from 1.09/ 100.000 to 5.25/100.000 in the last 30 years; the highest prevalence is in the fourth to sixth decades of life [1]. GEP-NETs may appear either as a single sporadic lesion or in the context of a more generalized genetic endocrine disorder such as Multiple Endocrine Neoplasia (MEN) type 1, Von Hippel–Lindau (VHL) disease, Neurofibromatosis (NF) type 1 or tuberous sclerosis [2]. Their clinical presentation is related to the secretion or not of ‘‘active peptides’’ that act distantly as hormones. The most common GEP-NETs are the carcinoids (apudomas that may be located in the small or large bowel with mainly serotonin secretion) that are usually presented with diarrhoea and heart disease. Other GEP-NETs are the insulinomas (islets of Langerhans cells of the pancreas with insulin secretion) that are presented with severe hypoglicemia, the gastrinomas (mainly pancreatic location and gastrin secretion) that are presented with acid hypersecretion and duodenal ulceration and the glucagonomas (alpha cells of the pancreas and glucagon secretion) that are presented with diabetes and necrolytic migratory erythema. More rare GEP-NETs are the Vasoactive Intestinal Peptide tumours (or VIPomas, known as Verner Morrison syndrome) and the somatostatinomas that appear with a frequency of 1/10,000,000 cases) [3]. The most frequent GEP-NETs and their clinical presentation characteristics 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs Table 1 The most common GEP-NETs and their clinical presentation characteristics Type of NET Clinical presentation Special exams (other than CT or MRI) Neuroendocrine tumours of the midgut Carcinoid syndrome characterized by flushing, diarrhoea, bronchoconstriction and cardiac abnormalities CT/MR enterography, octeotride-scan, DOTATOC-PET and colonoscopy Gastrinoma Typical peptic ulcer disease and severe diarrhoea Endoscopic ultrasound, octeotride-scan or DOTATOC-PET (only for patients with normal gastrin blood level) Insulinoma Episodes of Hypoglycaemia and weight loss Endoscopic ultrasound VIPoma Watery diarrhoea, hypokalaemia, achlorhydria (Verner Morrison Syndrome) Octeotride-scan, DOTATOC-PET Glucagonoma Diabetes and Necrolytic migratory erythema Mass symptoms such as liver failure, obstruction (bowel and biliary), ischaemia and bleeding Octeotride-scan, DOTATOC-PET Non-functioning NET, with or without liver metastases Octeotride-scan, DOTATOC-PET are shown in Table 1. GEP-NETs metastasize to the liver with a frequency between 25 and 90 % [3] (Table 2). Accurate grading and staging permits the evaluation of clinical, pathological and prognostic features. GEP-NETs are classified according to the World Health Organization (WHO) classification, the European Neuroendocrine Tumor Society (ENETS) grading system and the Union for International Cancer Control (UICC) system. In 2010, the latest version of the WHO classification was published amending the two previously published versions of 2000 and 2004 that distinguished between well-differentiated neuroendocrine tumours (NETs) and poorly differentiated neuroendocrine carcinomas (NECs). The GEP-NETs are now separated according to the mitotic activity found in 10 high-power fields (HPF), into G1:\2/10 HPF, G2: 2–10/10 HPF and G3: [10 HPF and the proliferative activity as expressed by the antigen KI-67 as low (\2 %), intermediate (3–20 %) and high ([20 %) [4, 5]. The American Joint Committee on Cancer (AJCC), in the 7th edition AJCC Cancer Staging Manual, and the European Neuroendocrine Tumor Society (ENETS) have published two systems of TNM staging manual; there are 123 some differences between these systems, particularly for primary tumours of the pancreas and the appendix, but there is also a considerable overlap. Additionally, the staging criteria for both systems rely predominantly on the size of the tumour and the extent of invasion into similar landmarks as used for the staging of non-neuroendocrine carcinomas of the same sites [6]. Grading and TNM directly correlate with patients’ survival. Localized, regional and metastatic disease is linked to a progressively lower survival rate. In addition, the primary tumour site, the histological type, the age, sex and race appear to influence overall survival [7]. Therapeutic approaches may either involve total surgical excision where applicable or local and symptomatic control in the cases of advanced and metastatic disease [8]. Management is specific for each type of tumour and is described in the international guidelines of the National Comprehensive Cancer Network (NCCN) with the latest version published in 2014 (NCCN 2.2014 guidelines version). Surgery is applied to approximately 10 % of the cases; the remaining of the cases are considered inoperable either due to the extensive location of the primary tumour with involvement of surrounding anatomical structures or due to the presence of extensive metastatic disease. Surgical excision of the primary GEP-NET may be curative in 75–98 % of patients [9]. Debulking resection may be also offered for primary locally advanced lesions. However, the morbidity is high and complications such as bleeding, fistula formation or tumour seeding may occur. Laparoscopic treatment of pancreatic GEP-NETs demonstrated to be correlated to slightly lower morbidity than open resection; the main advantage was the reduction of intra-abdominal collections and pleural effusion formation, postoperative fever and infection, postoperative haemorrhage, necrotizing pancreatitis and duration of hospitalization [10]. Many alternatives to surgery have been proposed for unresectable GEP-NETs with or without metastatic disease aiming to improve the median progression-free survival and to control symptoms in functional disease. Of particular interest are biological therapies (everolimus, sunitinib), Peptide Receptor Radio nucleotide Therapy (PRRT) and cytotoxic chemotherapy. However, chemotherapy does not appear to be effective for the reduction of the tumour burden whereas somatostatine analogues like octreotide do not appear to be effective for the control of symptoms [8]. Image-guided treatment in the form of percutaneous embolization; chemoembolization, trans-arterial radioembolization (TARE) or thermal ablative treatment has also offered a valid option in a variety of cases of patients with GEP-NETs and unresectable primary or metastatic disease. The role of palliation is to offer a good quality of life (QoL) and to preserve the patient as symptom-free for as long as possible [11]. Article Technique Patients 1 year OS 2 years OS 3 years OS 5 years OS Median OS Notes 33.8 months – Embolic TAE 32 nd nd nd nd TACE 22 nd nd nd nd Strosberg et al. [45] TAE 84 nd nd nd 45 % 36 months – Ruutiainen et al. [51] TAE 23 68 nd 46 33 nd – TACE 44 86 nd 67 50 nd Pitt et al. [52] TAE TACE 51 49 70 69 54 52 nd nd 13 13 25.7 months 25.5 months – Maire et al. [53] TAE 14 100 100 nd nd nd – TACE 12 91 80 nd nd nd De Baere et al. [55] DEBs-TACE 18 100 nd nd nd nd – Kennedy et al. [59] Y90-Radioembolization 148 nd nd nd nd 70 months – Gupta et al. [44] King et al. [60] Y90-Radioembolization 34 nd nd nd nd 27.6 months (mean) – Cao et al. [61] Y90-Radioembolization 58 86 58 47 nd 36 months – Kalinowski et al. [63] Y90-Radioembolization 9 100 57 57 nd nd – Limouris et al. [68] In-111-DTPA-Phe1-Octreotide 17 nd nd nd nd 32 months – Ablative Akyildiz et al. [70] VLS-RFA 89 nd nd nd 57 72 months – Mazzaglia et al. [71] VLS-RFA 63 91 77 nd 48 3.9 years Not only neuroendocrine Martin et al. [72] MW 11 nd nd nd nd 18 months – Shapiro et al. [74] Cryoablation 5 60 40 nd nd nd – Bilchik et al. [76] Cryoablation 19 nd nd nd nd 49 months – OS overall survival, TAE trans-arterial embolization, TACE trans-arterial chemioembolization, DEBs-TACE drug-eluting beads TACE, VLS-RFA videolaparoscopic radiofrequency ablation, MW microwaves, nd not defined G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs Table 2 Overall survival as described in the literature post-ablation end embolization of metastatic lesions 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs In the multidisciplinary tumour board, the choice of which image-guided procedure to use depends on the local expertise and availability; there are no randomized clinical trials comparing the efficacy loco-regional therapies and palliative liver surgery [12, 13]. The purpose of this review article is to describe the current status on the use of image-guided treatment in the management of Gastro-Entero-Pancreatic GEP-NETs. Image-guided treatment The first-line treatment for primary GEP-NETs is surgical resection according to the International Guidelines [12]. However, image-guided interventions play a significant role in complicated cases or in cases that cannot undergo surgical resection. The percutaneous treatment of primary GEP-NETs either with embolization or thermal ablation is described by several authors in the literature [12, 13, 19]. Primary GEP-NETS Embolization Detection and Imaging Characteristics Percutaneous embolization is mainly used for the control of inoperable functional insulinomas or gastrinomas. Both appear to be suitable for treatment with embolization, as they are nourished by a rich blood supply, receiving about 10–15 % of the pancreatic blood flow [20]. A blush of contrast is visible in the angiographic pictures. The lesion is usually receiving supply from the dorsal pancreatic artery, however, other arteries like branches of the colic artery may also supply the area. The first reported case was described by Moore et al., where embolization was performed with microfibrillar collagen, in a young patient with recurrent insulinoma 1 year after surgical resection [21]. The authors reported control in the blood glucose levels in the 11 months followup. Uflacker et al., in 1992, reported two other cases with insulinoma, one in a patient who denied surgery and the other who underwent emergency embolization. PVA particles (150–300 nm) were used in both cases [22]. Rott et al. presented the case of an 84-year-old woman with a symptomatic insulinoma, who refused surgery and underwent successful embolization with 300–500 nm trisacryl gelatin microspheres [23]. Peppa et al. presented a 30-year-old patient with MEN1 and recurrent insulinoma with severe hypoglycaemic episodes who could not be surgically treated due to the adherence of the tumour to large blood vessels and to prior multiple surgical operations. He was treated by repeated embolization using non-spherical polyvinyl alcohol particles (100–300 nm), resulting in shrinkage of the tumour, decrease of the severity and the frequency of the hypoglycaemic episodes, and improvement of the quality of life [20]. Embolization may also be used in some cases for the reduction of blood supply prior to surgical resection. Ben-Ishay et al. [24] described preoperative embolization of an extremely hypervascular NET located at the head of the pancreas, which was successfully embolized and then excised, and Whipple operation followed. A relevant technical question, regarding the embolization of pancreatic NETs, is what might be the most suitable embolic material. Considering the size, there are no data regarding the precise diameter of the arteries that supply the NETs, so the choice of particle size is purely empirical. The detection of primary GEP-NETs usually requires the multimodality approach according to recent published guidelines [14], combining Computed Tomography (CT), Magnetic Resonance Imaging (MRI) and somatostatin receptor scintigraphy (SSRS). Endoscopic ultrasound (EUS) is also utilized in case of suspected GEP-NETs, while Gallium-68 Positron Emission Tomography (PET)/ CT (DOTATOC, DOTATATE) is recommended for detection of unknown primaries. In their majority, GEPNETs are small, thus the sensitivity is rather low for primary detection, with primary tumour revealed through the presence of mesenteric disease, enlarged lymph nodes and liver metastases. CT or MR Enteroclysis, allowing for multiplanar reformats certainly increase reported sensitivity and specificity for small bowel NETs. Functional, usually smaller, pancreatic NETs demonstrate avid contrast enhancement during arterial phase CT/ MRI. Combined with EUS, multidetector CT may reach 100 % sensitivity [15]. Larger, malignant, non-functioning pancreatic tumours demonstrate necrosis, calcification and invasion of adjacent structures. Pancreatic NETs show high signal intensity on T2 weighted and low signal intensity on T1-weighted sequences. On post-intravenous contrast administration, MR images lesions may be isointense to surrounding pancreatic parenchyma. Imaging plays a key role also during follow-up. The ENETS recommends the use of all available imaging techniques. However, the technique of choice depends on the local expertise; CT or MR is recognized as the gold standard during follow-up, allowing the evaluation of progression and/or spread of the tumour. Also octreoscan is widely used, however, DOTATOC-PET demonstrated high diagnostic accuracy and probably is going to replace octreoscan in the future [16]. Chromogranin A (CgA) samples are useful for both diagnosis and follow-up with a sensitivity and specificity of 75.3 and 84.2 %, respectively [17]. The CgA levels at diagnosis are also linked with the overall survival of the patient [18]. 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs Most of the operators used microparticles (100–300 nm) and there are no reports in the literature of failed embolization. Particles are usually used because an acute ischaemic effect is desirable. Potential complications of pancreatic NET embolization include exudative pancreatitis and a form of type 2 diabetes in insulinomas [23]. Thermal ablations Radiofrequency ablation of the pancreas is nowadays a feasible method of tumour control [25]. It was initially applied to animal models. Goldberg et al. studied the safety and efficacy of RFA in experimental models and concluded that RFA can be used in small neuroendocrine tumours [26]. Date et al. [27] demonstrated the safety and efficacy of RFA in the normal pancreas of a porcine model. Matsui et al. [28] published the first clinical study of 20 patients in the year 2000. In the literature, the use of this technique for the treatment of NETs was described in just few cases and was mainly indicated in patients with functional tumours, not candidate to surgery and where the medical therapy failed. The first case report was published in 2009 by Limmera et al. and describes the case of CT-guided RFA of an insulinoma localized at the tail of the pancreas in an elderly patient with high surgical risk and where the medical therapy was not satisfactory [29]. After the procedure, the patient was free of symptoms and no complications occurred. Akhlaghpoor et al. described a case of a 48-yearold man with a functional insulinoma of the tail of the pancreas that after surgery and medical therapy continued to be symptomatic due to the presence of residual functional tumour [30]. The patient underwent a CT Fluoroscopy–guided transcaval radiofrequency ablation, which was clinically successful and no complication occurred. Pei-Hong et al. reported a case of a percutaneous radiofrequency ablation approach through the spleen for a pancreatic gastrinoma [31]. Hlavsa et al. presented the first case report of a patient with locally advanced pancreatic neuroendocrine tumour successfully treated with intraoperative RFA [32]. However, due to the delicate structure of the organ and the central location, pancreatic RFA may lead to severe complication [29]. Rates of overall complications, in patients with locally advanced pancreatic tumours, ranged from 10 to 43 % [33]. The types of complication reported varied widely and included pancreatic fistulae, portal vein thrombosis, gastrointestinal bleeding, acute pancreatitis, pneumonia, peritoneal cavity abscess, acute renal failure, transient ascites, hepatic insufficiency, pseudo-membrane colitis, haemoperitoneum, abdominal fluid collection, gastric bypass fistula, gastric ulcer, choledocholithiasis and sepsis [33]. These complications were reported after the treatment of all types of pancreatic tumours and not specifically NETs. High Intensity Focused Ultrasound (HIFU) is another innovative ablation technique that has been recently introduced for the treatment of non-resectable tumours. HIFU achieves ablation by way of focused US energy from an external source that is targeted within the body, resulting in thermally induced necrosis. Acoustic energy is absorbed and delivering high-acoustic intensity to the tissue generates heat. Because it is focused, the acoustic intensity is high only within the focal region; however, outside the focal region, the intensity is substantially lower, thus minimizing the risk of unintended injury to the surrounding structures. Orgera et al. reported the treatment of two cases of pancreatic insulinomas treated with HIFU in two inoperable young female patients [9]. Both suffered from episodes of severe nightly hypoglycaemia that was not efficiently controlled by medical treatment. After HIFU ablation, local disease control and symptom relief were achieved without complications. Chen et al. described the treatment of an unresectable giant pancreatic neuroendocrine tumour successfully treated with HIFU with no significant complication detected [34]. Irreversible electroporation (IRE or non-thermal irreversible electroporation) is yet another novel technique of tissue ablation that uses short pulses of high voltage and leads to cellular death. IRE demonstrated to be a feasible technique for the treatment of pancreatic lesions in the swine model; recent studies demonstrated the feasibility of this novel approach on pancreatic masses, mainly an intraoperative setting [35]. However, IRE is technically demanding and may lead to significant complication and the best results are offered for the treatment of metastatic liver disease whereas there is no experience yet with primary GEP-NETs [36]. Metastatic Lesions Liver is the predominant site for GEP-NET metastases. Dissemination from a primary GEP-NET to the liver parenchyma may occur in 25–90 %, and it is the major factor altering both quality of life and prognosis regardless the primary site considering that patients with metastatic GEP-NETs appear to show worst prognosis compared to patients with only primary tumour localization [37–42]. Imaging Characteristics Hepatic metastases and the degree of liver involvement are considered the major prognostic factors for survival in patients with neuroendocrine tumours. GEP-NET-derived liver metastases are in their majority hypervascular, 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs Fig. 1 CT scan in arterial phase of a 61-year-old female with carcinoid located in the small bowel. The primary was removed surgically, however, the patient remained symptomatic due to the presence of two metastatic lesions in the segments VII (A) and V (B) of the liver (arrows) demonstrating avid contrast uptake during arterial phase imaging regardless of the used imaging technique (CT or MRI). However, liver metastases may also present with atypical hypovascular or delayed enhancement, in a reported 16 %, and a peripheral enhancement with progressive fill-in, mimicking haemangiomas, in 11 %, respectively [43]. It is also important to keep in mind that chemotherapy may alter liver metastases vascularity, lowering the performance of hepatic arterial phase imaging. The combination of arterial phase and fat-suppressed fast spin-echo T2-weighted images depicts at least 80 % of hepatic metastases derived from neuroendocrine tumours, rendering these cardinal sequences for their detection. Image-Guided Treatment The choice of the type of treatment depends on the local expertise, the disease location and extension (number and size of lesions). Minimal invasive treatment may be applied to both functional and non-functional tumours. Loco-regional therapies may also be used in combination with somatostatin analogues (SSA), particularly in functional GEP-NETs. If bulky disease is present, loco-regional therapy is indicated as an early measure even in the case of non-functional tumours, and may be useful for downstaging. In patients with functioning tumours, loco-regional therapies may be considered in the presence of extrahepatic disease if liver is the major site of the disease. If extrahepatic tumour load is higher than hepatic tumour burden, 123 and if pancreas is the primary tumour site, systemic medical therapies or peptide receptor radionuclide therapy (PRRT) may be combined to the loco-regional approach [13]. Patients with G2 disease and diffuse liver involvement may be treated with SSA/Interferon-alfa, or chemotherapy, molecular-targeted therapy (sunitinib or everolimus), PRRT or TACE/TAE depending on primary tumour site and the individual conditions. In highly selected candidates with diffuse metastases, liver transplantation may be also an option [13]. Embolization Embolization techniques may be used to treat liver metastases in patients where surgery is not feasible regardless of the origin of the primary tumour. These techniques appear to be effective in the control of symptoms and tumour growth and result in significant decrease in biochemical tumour activity [13]. Embolization is based on the principle that the lesions are hypervascular and therefore may be visible in the angiographic pictures (Fig. 1). Trans-arterial Embolization (TAE) may be performed with a large variety of embolizing materials like polyvinyl alcohol particles, gelfoam and microspheres aiming for tumour ischaemia [44, 45]. Embolization therapies can be subdivided in various sessions, according to the extension of the lesion and the tumour response, and may be repeated when there is progression of disease. In general, when there is an G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs Fig. 2 Selective angiogram from the right hepatic artery confirmed the presence of two hypervascular lesions (A, arrows). Embolization of the lesions with microparticles (75 micron) was performed; no contrast blush is further detected in the post-embolization angiogram (B) Fig. 3 CT scan in arterial phase at 3 weeks post-embolization confirmed lack of enhancement of the previously treated lesions. The patient’s symptoms were controlled at this stage involvement of more than 50 % of the liver parenchyma, embolization needs to be performed as a staged procedure to reduce cytotoxic effects [46, 47]. Other factors, like the requirement for urgent treatment or the presence of additional extra hepatic metastases are predictors of shorter survival after embolization of hepatic neuroendocrine metastases [46]. In addition, octreotide may be injected subcutaneously before TAE as prophylaxis against a potential exacerbation of a hormonal crisis [44]. An increase of angiogenetic factors was demonstrated after TAE in GEP-NET liver metastasis [48]. It is supposed that these factors affect the success of TAE; in fact, after TAE, lesions do not show a complete necrosis, and the development of new peripheral neo-vessels was demonstrated [49]. The association of TAE with new anti-angiogenetic drugs (as sunitinib) showed promising results in liver metastasis treatment [50] (Figs. 2, 3, 4). Embolization may also be combined with drug delivery in the case of Trans-arterial Chemoembolization (TACE), most commonly with doxorubicin, streptozotocin, cisplatin, mitomycin C or gemcitabine that may be injected in the arterial tree and then embolization with conventional material may follow or that can be delivered in the form of drug-eluting beads. The advantage over systemic delivery is the higher drug concentration in hepatic circulation with lower systemic side effects. 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs Fig. 4 CT scan in arterial phase at 6 months post-embolization shows minor rim enhancement of the lesions that, however, appear significantly smaller. The patient’s symptoms were still controlled at this follow-up scan The effectiveness of TACE over TAE is still debatable for hepatic neuroendocrine metastases. The retrospective study of Ruutiainen et al. which included 67 patients who received either TAE (n = 23) with polyvinyl alcohol particles or TACE (n = 44) with cisplatin, doxorubicin or mitomycin C, reported a small advantage of TACE regarding rates of freedom from progression at 1, 2 and 3 years, symptom relief and survival rates at 1, 3 and 5 years, however, without statistical significance [51]. The results were not confirmed in the study of Pitt et al. that performed also a retrospective review of 100 patients with hepatic neuroendocrine metastases who underwent either TAE (n = 51) or TACE (n = 49) and reported that morbidity, mortality, symptom improvement and overall survival are similar between the two groups [52]. The authors used polyvinyl alcohol particles, gelfoam or embospheres for the embolization, and the drugs administered for the TACE were cisplatin, adriamycin and mitomycin C. In addition, the authors performed a significant higher number of procedures in the TACE group (123 TACE vs. 106 TAE; p \ 0.02). Maire et al. in a smaller prospective randomized study of 26 patients who underwent TAE (n = 14) or TACE (n = 12) also reported that progression-free survival and overall survival were similar in the two groups without statistically significant difference [53]. Fiore et al. in another retrospective study of 30 patients with a histologically confirmed NET and liver metastases who underwent either TAE (n = 17) or TACE (n = 13) reported similar results, with no significant difference in 123 the progression-free survival between the two groups [54]. In addition, patients treated with TAE in this study developed post-embolization syndrome in 41 versus 61 % of the patients who were treated with TACE, suggesting that TAE offers not only similar results to TACE but also a slightly better toxicity profile. With the use of drug-eluting beads (DEB), the drugs— usually doxorubicin—are released slowly within a time frame of 7–14 days after the embolization. De Baere et al. in a small study of 20 patients used TACE with DEBs (500–700 mum) loaded with doxorubicin and reported a partial response of 80 % at 3 months and 45 % disease control at 15 months with no liver toxicity [55]. Gaur et al. in a similar study of 18 patients used 100–300 and 300–500 lm doxorubicin-eluting beads and reported 65 % of response at intermediate-term follow-up (mean, 445 days; range, 163–1247) [56]. Bhagat et al. [57] in a phase II trial of 13 patients with hepatic metastases from neuroendocrine tumours who underwent DEB-TACE with 100–300 lm beads loaded with B100 mg doxorubicin reported a mean of 12 % decrease in tumour size (p \ 0.0003) and a 56 % decrease in tumour enhancement (p \ 0.0001) at 1 month postprocedure. However, 54 % of the patients developed bilomas, and 30 % underwent a drainage procedure which was unexpected by the authors and led to forced interruption of the trial. The explanation was mainly based on the potential deposition of chemotherapeutic agents into the peribiliary arterial plexus. The authors observed also that the presence of large ([4 cm) multiple lesions was G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs protective against the formation of bilomas, which occurred in patients with small (\4 cm) multiple lesions. In addition to the small tumour burden, the smaller bead size may be another predisposing factor for biloma formation due to more profound ischaemic changes, which ultimately affect the peribiliary plexus. Guiu et al. compared DEBsTACE with TACE both in Hepatocellular Carcinoma (HCC) and GEP-NET liver metastases; DEBs-TACE demonstrated an higher risk (OR: 6.6) of liver or biliary injury than TACE; moreover, biloma and liver infarct were independently correlated with both DEBs-TACE (OR: 9.78) and NET (OR: 8.13), so they suggest to carefully consider this procedure in a non-cirrhotic liver, as well as in NET metastases [58]. Trans-arterial Radio-embolization (TARE) is another embolization technique that allows to apply high radiation doses in a selective area of the liver using microspheres of glass (diameter 20–30 lm) or resin (diameter 20–60 lm) loaded with a radioisotope, conventionally Yttrium-90. There are various studies that analysed the effects of TARE in NET hepatic metastases. Kennedy et al. in a retrospective review of 148 cases, treated with a median radiation dose delivery of 1.14 GBq per session, reported that imaging response to the treatment was stable in 22.7 %, partial in 60.5 %, complete in 2.7 and 4.9 % of patients presented progressive disease [59]. The median survival was 70 months. The range of treatment as described by other authors varies between studies 39 and 66 % [60–63]. Paprottka et al. reported a partial response of only 22.5 %, however, in 97.5 % of patients, the liver lesions appeared either hypovascular or partially necrotic [64]. Rhee et al. reported no significant difference in terms of tumour response and median survival between patients treated with glass or resin microspheres [65]. Saxena et al. have identified three factors associated with a complete/partial response: female gender (p = 0.040), well-differentiated tumour (p \ 0.001) and low hepatic tumour burden (p = 0.041) [62]. The control of symptoms varies from 50 to 100 % of cases [60, 66]. The most common complications after Yttrium-90 radio-embolization are fatigue, abdominal pain, nausea and fever [59]. Radiation-induced gastritis, liver dysfunction and pneumonia may also occur. King et al. reported 2 cases of gastritis, 1 of duodenal ulcer and 1 case of liver dysfunction and pneumonia among 34 treated patients [60]. An advantage of TARE over TAE and TACE is that it generally needs no more than one session of treatment and that the majority of patients may be treated on an outpatient basis [59, 67]. Infusion of In-111-DTPA-Phe1-octreotide after selective catheterization of hepatic arteries allows delivering the labelled octreotide directly to the lesions avoiding drug spread in other organs and especially in the kidneys. Repeated infusions are usually performed achieving stabilization or regression of the disease in up to the 70 % of patients [68]. Also Y-90 and Lu-177 are available to label DOTA and to be used in selective arterial catheterization. The number and the dimension of the lesions to embolize limit the performance and feasibility of this technique. Thermal Ablative Techniques Radiofrequency ablation (RFA) for NET hepatic metastases may be performed either percutaneously under Computed Tomography (CT) or Ultrasound (US) guidance or intraoperatively with or without concomitant liver resection. RFA conventionally is adopted for liver lesions smaller than 5 cm in size [69] (moreover other ablative techniques are not considered a suitable single therapy for lesions larger than 5 cm) [13]. The intent of the ablation is not only tumour debulking but also control of hormonal symptoms, and sometimes treatment of several lesions may be required in the same session, particularly in an intraoperative setting. In a study from Akyildiz et al. 89 patients with metastatic lesions mainly from carcinoid and pancreatic islet cell underwent RFA under laparoscopic guidance; no liver resection was performed [70]. The mean number of treated lesions was 6 ± 1, and the mean tumour size was 3.6 ± 0.2 cm. Symptom control was successful in 97 % of patients 1 week after the procedure, however, perioperative morbidity was 5.6 % due to post-operative haemorrhage, and the 30-day mortality was 1 % due to intra-abdominal sepsis. Median follow-up was 30 ± 3 months with 15 months of disease-free survival. Local liver recurrence occurred in 22 %, new liver lesions occurred in 63 % and extra hepatic metastatic disease occurred in 59 % of the cases. In case of single lesions, repeated RFA was performed in 27 % of the cases, whereas if the recurrence was multifocal, chemoembolization (7 %) was performed. Median disease-free survival was 1.3 years, and the overall survival was 6 years. The effect of RFA in such patients is mainly local control and symptom relief, which is achievable, however, the multifocal lesions may be difficult to be treated with RFA and extensive ablation may lead to generalized sepsis. Mazzaglia et al. in a study of 63 patients reported similar results in terms of local control and symptomatic improvement and also noted a correlation between lesion size and median survival that was less than 3 years for lesions bigger than 3 cm. RFA complications as liver abscess, pain, bile leakage and haemorrhage were not correlated with the histology of the tumour treated [71]. Microwave ablation (MWA) is performed with electromagnetic devices usually with a frequency higher than 900 MHz. The basic principle of this technique is not different from RFA but there are some differences that 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs theoretically result in a better control of the disease: in fact, in MWA, we do not observe ‘‘heat sink’’ effect (it should be particularly indicated to treat lesions close to vessels), and MWA permits a higher temperature within the lesion treated with a corresponding higher cytotoxic effect. Some studies on MWA in liver lesions include metastases from NET. Martin et al. treated 11 patients with a mean number of 4 lesions per case, the majority of the procedures were performed intraoperatively in combination with hepatic resection or with extra hepatic tumour resection [72]. Complete ablation of 90 % was reported. Complications occurred in three patients and there was no recurrence at the ablation site. Groeschl et al. [73] treated in a large multicentre analysis of 450 patients treated with MWA included 61 patients with neuroendocrine liver metastasis. Complete ablation was confirmed for 97.0 % of the subgroup patients. They noted that patients with 3 cm or more tumours showed a propensity for early recurrence, regardless of histology. No significant differences in complication rates or survival based on the surgical approach were reported; however, local recurrence rates were highest for percutaneously treated lesions. Cryotherapy is another thermal ablation method that is based on the concept that low temperature (\-50 °C degrees) causes cell death and may reduce the tumour burden and lead to symptomatic relief. Some authors in the treatment of NET liver metastases used cryotherapy. Shapiro et al. [74] in 1998 reported that in 5 patients with symptomatic liver metastases from carcinoid tumour underwent intraoperative, ultrasound-guided cryotherapy. Immediate symptomatic relief occurred in all patients and in 4 out of 5 lasted more than 3 months. However, survival was limited in this small series, therefore immediate symptomatic relief was the only benefit. In another study from the same year, Seifert et al. [75] treated 13 patients with cryotherapy. At the follow-up of 13.5 months, 12 of them were alive with satisfactory control of the disease. Bilchik et al. [76] in a study published 1 year earlier of 19 patients with metastatic lesions from NETs who underwent intraoperative US-guided cryotherapy with resection of the primary tumour also reported relief from symptoms with reduction in tumour markers. However, in both studies cryotherapy led to post-procedural coagulopathy that required transfusions [75, 76]. Jansen et al. have shown in a rat liver model that cryoablation induces greater inflammatory and coagulative response compared to radiofrequency ablation and that coagulation anomalies are related with the volume of the tissue treated and not with the number of performed sessions [66]. Seifert et al. suggested also that cryotherapy in metastatic NETs could cause release of substances that may influence the coagulation system [77]. Sheffer et al. in a recent systematic review of IRE safety and efficacy reported a low incidence of complication in 123 the treatment of liver metastasis, where only minor complication were reported; otherwise in the primitive pancreas tumour ablation they reported major complications (bile leak and portal vein thrombosis). Conclusion Neuroendocrine lesions and the metastatic disease that derives from them are a very vast and heterogeneous group of diseases that are not all approached with the same strategy. The role of Interventional Radiology in the management of these tumours is to provide symptom control and local control in the cases that are considered inoperable. Embolization appears to be the most diffuse method in the management of both primary and metastatic lesions. Ablation techniques have also shown promising results. IR is an effective treatment option even if there is the necessity of assessing the impact on the patient’s quality of life. Conflict of interest Gianluigi Orgera, Miltiadis Krokidis, Matteo Cappucci, Sofia Gourtsoyianni, Marcello Andrea Tipaldi, Adam Hatzidakis, Alberto Rebonato and Michele Rossi have no conflicts of interest. References 1. Amodio A, Crosetti S (2010) Tumori Neuroendocrini. In: Lopez M, Gebbia N, Cascinu S, Marchetti P (eds) Oncologia Medica Pratica. Società Editrice Universo, Roma, pp 2173–2194 2. Muniraj T, Vignesh S, Shetty S et al (2013) Pancreatic neuroendocrine tumor. Dis Mon 59:5–19 3. Colao A, Faggiano A, Milone F (2011) Tumori Neuroendocrini. In: Lenzi A, Lom-bardi G, Martino E, Vigneri R (eds) Endocrinologia Clinica. Edizioni Minerva Medica, Torino, pp 217–224 4. Klöppel G (2011) Classification and pathology of gastroenteropancreatic neuroendocrine neoplasms. Endocrine Related Cancer 18:S1–S16 5. Bosman FT, Carneiro F, Hruban R et al (2010) WHO classification of tumours of the digestive system. IARC Press, Lyon 6. Klimstra DS, Modlin IR, Coppola D et al (2010) The pathologic classification of neuroendocrine tumors: a review of nomenclature, grading, and staging systems. Pancreas 39:707–712 7. Yao JC, Hassan M, Phan A et al (2008) One hundred years after ‘‘carcinoid’’: epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States. J Clin Oncol 26(18):3063–3072 8. Burns WR, Edil BH (2011) Neuroendocrine pancreatic tumors: guidelines for management and update. Curr Treat Options Oncol 13:24–34 9. Orgera G, Krokidis M, Monfardini L et al (2011) High intensity focused ultrasound ablation of pancreatic neuroendocrine tumours: report of two cases. Cardiovasc Intervent Radiol 34:419–423 10. Drymousis P, Raptis DA, Spalding D et al (2014) Laparoscopic versus open pancreas resection for pancreatic neuroendocrine tumours: a systematic review and meta-analysis. HPB (Oxford) 16(5):397–406 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs 11. Rossi RE, Massironi S, Spampatti MP, Conte D, Ciafardini C, Cavalcoli F, Peracchi M (2012) Treatment of liver metastases in patients with digestive neuroendocrine tumors. J Gastrointest Surg. 16(10):1981–1992 12. Oberg K, Knigge U, Kwekkeboom D, ESMO Guidelines Working Group et al (2012) Neuroendocrine gastro-entero-pancreatic tumors: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 23(Suppl 7):vii124–vii130 13. Pavel M, Baudin E, Couvelard A et al (2012) ENETS Consensus Guidelines for the management of patients with liver and other distant metastases from neuroendocrine neoplasms of foregut, midgut, hindgut, and unknown primary. Neuroendocrinology 95:157–176 14. Ramage JK, Ahmed A, Ardill J, UK and Ireland Neuroendocrine Tumour Society et al (2012) Guidelines for the management of gastroenteropancreatic neuroendocrine (including carcinoid) tumours (NET’s). Gut 61(1):6–32 15. Gouya H, Vignaux O, Augui J et al (2003) CT, endoscopic sonography, and a combined protocol for preoperative evaluation of pancreatic insulinomas. AJR Am J Roentgenol 181(4):987–992 16. Arnold R, Chen YJ, Costa F, Mallorca Consensus Conference participants, European Neuroendocrine Tumor Society et al (2009) ENETS Consensus Guidelines for the Standards of Care in Neuroendocrine Tumors: follow-up and documentation. Neuroendocrinology 90(2):227–233 17. Campana D, Nori F, Piscitelli L et al (2007) Chromogranin A: is it a useful marker of neuroendocrine tumors? J Clin Oncol 25(15):1967–1973 18. Arnold R, Wilke A, Rinke A et al (2008) Plasma chromogranin A as marker for survival in patients with metastatic endocrine gastroenteropancreatic tumors. Clin Gastroenterol Hepatol 6(7): 820–827 19. Kulke MH, Benson AB III, Bergsland E et al (2012) Neuroendocrine tumors. J Natl Compr Cancer Netw 10:724–764 20. Peppa M, Brountzos E, Economopoulos N et al (2009) Embolization as an alternative treatment of insulinoma in a patient with multiple endocrine neoplasia type 1 syndrome. Cardiovasc Intervent Radiol 32:807–811 21. Moore T, Peterson L, Harington D (1982) Successful arterial embolization of an insulinoma. JAMA 248:1953–1955 22. Uflacker R (1992) Arterial embolization as definitive treatment for benign insulinoma of the pancreas. J Vasc Interv Radiol 3:441–639 discussion 644–666 23. Rott G, Biggemann M, Pfohl M (2008) Embolization of an insulinoma of the pancreas with trisacryl gelatin microspheres as definitive treatment. Cardiovasc Intervent Radiol 31:659–662 24. Ben-Ishay O, Linder R, Ofer A et al (2012) Hypervascular lesion in the head of the pancreas. Preoperative angiography and selective embolization results in bloodless operation. JOP 13:671–673 25. Rossi M, Orgera G, Hatzidakis A et al (2014) Minimally invasive ablation treatment for locally advanced pancreatic adenocarcinoma. Cardiovasc Intervent Radiol 37:203–210 26. Goldberg SN, Mallery S, Gazelle GS et al (1999) EUS-guided radiofrequency ablation in the pancreas: results in a porcine model. Gastrointest Endosc 50:392–401 27. Date RS, Biggins J, Paterson I et al (2005) Development and validation of an experimental model for the assessment of radiofrequency ablation of pancreatic parenchyma. Pancreas 30:266–271 28. Matsui Y, Nakagawa A, Kamiyama Y et al (2000) Selective thermocoagulation of unresectable pancreatic cancers by using radiofrequency capacitive heating. Pancreas 20:14–20 29. Limmer S, Huppertb P, Juettea V et al (2009) Radiofrequency ablation of solitary pancreatic insulinoma in a patient with episodes of severe hypoglycemia. Eur J Gastroenterol Hepatol 21:1097–1101 30. Akhlaghpoor S, Dahi F, Alinaghizadeh M et al (2011) CT fluoroscopy-guided transcaval radiofrequency ablation of insulinoma. J Vasc Interv Radiol 22:409–410 31. Wu P, Pan C, Huang Z et al (2010) Percutaneous radiofrequency ablation approach through the spleen: initial case report for pancreatic tail gastrinoma. Chin J Cancer 29:836–841 32. Hlavsa J, Procházka V, Kala Z et al (2011) Radiofrequency ablation of pancreatic neuroendocrine tumor. Klin Onkol 24: 209–215 33. Fegrachi S, Besselink MG, van Santvoort HC et al (2014) Radiofrequency ablation for unresectable locally advanced pancreatic cancer: a systematic review. HPB (Oxford) 16:119–123 34. Chen Q, Zhu X, Chen Q et al (2013) Unresectable giant pancreatic neuro-endocrine tumor effectively treated by HIFU: a case report and a review of the literature. Pancreatology 13:634–638 35. Dunki-Jacobs EM, Philips P, Martin RC 2nd (2014) Evaluation of resistance as a measure of successful tumor ablation during irreversible electroporation of the pancreas. J Am Coll Surg 218(2):179–187 36. Scheffer HJ, Nielsen K, de Jong MC et al (2014) Irreversible electroporation for nonthermal tumor ablation in the clinical setting: a systematic review of safety and efficacy. J Vasc Interv Radiol 25(7):997–1011 37. Mayo SC, De Jong MC, Pulitano C et al (2010) Surgical management of hepatic neuroendocrine tumor metastasis: results from an international multi-institutional analysis. Ann Surg Oncol 17:3129–3136 38. John BJ, Davidson BR (2012) Treatment options for unresectable neuroendocrine liver metastases. Expert Rev Gastroenterol Hepatol 6:357–369 39. Frilling A, Sotiropoulos GC, Li J et al (2010) Multimodal management of neuroendocrine liver metastases. HPB (Oxford) 12:361–379 40. McDermott EW, Guduric B, Brennan MF (1994) Prognostic variables in patients with gastrointestinal carcinoid tumours. Br J Surg 81:1007–1009 41. Zeitels J, Naunheim K, Kaplan EL et al (1982) Carcinoid tumors: a 37-year experience. Arch Surg 117:732–737 42. Panzuto F, Merola E, Rinzivillo M et al (2014) Advanced digestive neuroendocrine tumors: metastatic pattern is an independent factor affecting clinical outcome. Pancreas 43:212–218 43. Dromain C, de Baere T, Baudin E et al (2003) MR imaging of hepatic metastases caused by neuroendocrine tumors: comparing four techniques. AJR Am J Roentgenol 180(1):121–128 44. Gupta S, Johnson MM, Murthy R et al (2005) Hepatic arterial embolization and chemoembolization for the treatment of patients with metastatic neuroendocrine tumors: variables affecting response rates and survival. Cancer 104:1590–1602 45. Strosberg JR, Choi J, Cantor AB et al (2006) Selective hepatic artery embolization for treatment of patients with meta-static carcinoid and pancreatic endocrine tumors. Cancer Control 13:72–78 46. Sofocleous CT, Petre EN, Gonen M et al (2013) Factors affecting periprocedural morbidity and mortality and long-term patient survival after arterial embolization of hepatic neuroendocrine metastases. J Vasc Interv Radiol 25:22–30 47. Kress O, Wagner HJ, Wied M et al (2003) Transarterial chemoembolization of advanced liver metastases of neuroendocrine tumors—a retrospective single-center analysis. Digestion 68:94–101 48. Korse CM, Bonfrer JM, Prevoo W et al (2011) Increase of angiogenic growth factors after hepatic artery embolization in patients with neuroendocrine tumours. Tumour Biol 32(4):647–652 49. Liao XF, Yi JL, Li XR et al (2004) Angiogenesis in rabbit hepatic tumor after transcatheter arterial embolization. World J Gastroenterol 10(13):1885–1889 123 G. Orgera et al.: Interventional Radiology in the Management of GEP-NETs 50. Strosberg JR, Weber JM, Choi J et al (2012) A phase II clinical trial of sunitinib following hepatic transarterial embolization for metastatic neuroendocrine tumors. Ann Oncol 23(9):2335–2341 51. Ruutiainen AT, Soulen MC, Tuite CM et al (2007) Chemoembolization and bland embolization of neuroendocrine tumor metastases to the liver. J Vasc Interv Radiol 18:847–855 52. Pitt SC, Knuth J, Keily JM et al (2008) Hepatic neuroendocrine metastases: chemo- or bland embolization? J Gastrointest Surg 12:1951–1960 53. Maire F, Lombard-Bohas C, O’Toole D et al (2012) Hepatic arterial embolization versus chemoembolization in the treatment of liver metastases of well-differentiated midgut endocrine tumors. A prospective randomized study. Neuroendocrinology 96:294–300 54. Fiore F, Del Prete M, Franco R et al (2014) Transarterial embolization (TAE) is equally effective and slightly safer than transarterial chemoembolization (TACE) to manage liver metastases in neuroendocrine tumors. Endocrine 47(1):177–182 55. De Baere T, Deschamps F, Teriitheau C et al (2008) Transarterial chemoembolization of liver metastases from well differentiated gastroenteropancreatic endocrine tumors with doxorubicin-eluting beads: preliminary results. J Vasc Interv Radiol 19:855–861 56. Gaur SK, Friese JL, Sadow CA et al (2011) Hepatic arterial chemoembolization using drug-eluting beads in gastrointestinal neuroendocrine tumor metastatic to the liver. Cardiovasc Intervent Radiol 34:566–572 57. Bhagat N, Reyes DK, Lin M et al (2013) Phase II study of chemoembolization with drug-elutingbeads in patients with hepatic neuroendocrine metastases: high incidence of biliary injury. Cardiovasc Intervent Radiol 36:449–459 58. Guiu B, Deschamps F, Aho S et al (2012) Liver/biliary injuries following chemoembolisation of endocrine tumours and hepatocellular carcinoma: lipiodol vs. drug-eluting beads. J Hepatol 56(3):609–617 59. Kennedy AS, Dezarn WA, McNeillie P et al (2008) Radioembolization for unresectable neuroendocrine hepatic metastases using resin 90Y-microspheres: early results in 148 patients. Am J Clin Oncol 31:271–279 60. King J, Quinn R, Glenn DM et al (2008) Radioembolization with selective internal radiation microspheres for neuroendocrine liver metastases. Cancer 113:921–929 61. Cao CQ, Yan TD, Bester L et al (2010) Radioembolization with yttrium microspheres for neuroendocrine tumour liver metastases. Br J Surg 97:537–543 62. Saxena A, Chua TC, Bester L et al (2010) Factors predicting response and survival after yttrium-90 radioembolization of unresectable neuroendocrine tumor liver metastases: a critical appraisal of 48 cases. Ann Surg 251:910–916 63. Kalinowski M, Dressler M, Konig A et al (2009) Selective internal radiotherapy with yttrium-90 microspheres for hepatic metastatic neuroendocrine tumors: a prospective single center study. Digestion 79:137–142 123 64. Paprottka PM, Hoffmann RT, Haug A et al (2012) Radioembolization of symptomatic, unresectable neuroendocrine hepatic metastases using yttrium-90 microspheres. Cardiovasc Intervent Radiol 35:334–342 65. Rhee TK, Lewandowski RJ, Liu DM et al (2008) 90Y radioembolization for metastatic neuroendocrine liver tumors: preliminary results from a multi-institutional experience. Ann Surg 247:1029–1035 66. Jansen MC, Van Hillegersberg R, Schoots IG et al (2010) Cryoablation induces greater inflammatory and coagulative responses than radiofrequency ablation or laser induced thermotherapy in a rat liver model. Surgery 147:686–695 67. Vyleta M, Coldwell D (2011) Radioembolization in the treatment of neuroendocrine tumor metastases to the liver. Int J Hepatol 2011:785315 68. Limouris GS, Chatziioannou A, Kontogeorgakos D et al (2008) Selective hepatic arterial infusion of In-111-DTPA-Phe1octreotide in neuroendocrine liver metastases. Eur J Nucl Med Mol Imaging 35(10):1827–1837 69. Auernhammer CJ, Goke B (2011) Therapeutic strategies for advanced neuroendocrine carcinomas of jejunum/ileum and pancreatic origin. Gut 60:1009–1021 70. Akyildiz HY, Mitchell J, Milas M et al (2010) Laparoscopic radiofrequency thermal ablation of neuroendocrine hepatic metastases: long-term follow-up. Surgery 148:1288–1293 71. Mazzaglia PJ, Berber E, Milas M et al (2007) Laparoscopic radiofrequency ablation of neuroendocrine liver metastases: a 10-year experience evaluating predictors of survival. Surgery 142:10–19 72. Martin RC, Scoggins CR, McMasters KM (2010) Safety and efficacy of microwave ablation of hepatic tumors: a prospective review of a 5-year experience. Ann Surg Oncol 17:171–178 73. Groeschl RT, Pilgrim CH, Hanna EM et al (2013) Microwave ablation for hepatic malignancies: a multiinstitutional analysis. Ann Surg 259(2):1195–1200 74. Shapiro RS, Shafir M, Sung M et al (1998) Cryotherapy of metastatic carcinoid tumors. Abdom Imaging 23:314–317 75. Seifert JK, Cozzi PJ, Morris DL (1998) Cryotherapy for neuroendocrine liver metastases. Semin Surg Oncol 14:175–183 76. Bilchik AJ, Sarantou T, Foshag LJ et al (1997) Cryosurgical palliation of metastatic neuroendocrine tumors resistant to conventional therapy. Surgery 122:1040–1047 discussion 1047–1048 77. Seifert JK, Morris DL (1999) World survey on the complications of hepatic and prostate cryotherapy. World J Surg 1923:109–113 discussion 113–114