Before the plenary presentation, a team of methodologists, clinical epidemiologists, and clinicians systematically reviewed the literature. A search of Medline, Embase, and the Cochrane Library was done. The search strategy included the term āneuroendocrine tumoursā AND/OR search strings connected to the topics of interestāeg, incidence. Results were restricted to human trials and those published between January, 1940, and October, 2012. 293ā3050 records were identified at each session.
ReviewRecommendations for management of patients with neuroendocrine liver metastases
Introduction
Gastroenteropancreatic (GEP) neuroendocrine neoplasms (GEP NENs), also called GEP neuroendocrine tumours (NETs) or carcinoids, were previously regarded as rare, but in fact are increasing in incidence (3Ā·65 per 100ā000 individuals per year1) and occur as frequently as testicular tumours, Hodgkin's disease, gliomas, and multiple myeloma.2 They represent an important clinical issue for two reasons: first, 40ā95% are metastatic at diagnosis, and second, evidence-based best practice strategies are scarce. Most present management is based on a synthesis of experience, local practice patterns, or archaic concepts.2 The central management issue is that at diagnosis about 65ā95% of GEP NENs (excluding appendiceal, gastric, and rectal NETs, about 85ā90% of which are local) show hepatic metastasis.3, 4 Indeed, liver metastases represent the most crucial prognostic factor, irrespective of the primary NET site. In historical series, 5 year survival is 13ā54% compared with 75ā99% for patients without hepatic metastases.5, 6 Experience indicates 5 year overall survival of 56ā83% for metastatic intestinal NETs and 40ā60% for pancreatic NETs,7 which is indicative of earlier diagnosis, more advanced imaging techniques, amplified surveillance, and the implementation of new treatment approaches.2 Despite various complex management strategies for neuroendocrine liver metastases, surgery is the only treatment that offers potential for cure.7 For unresectable lesions, optimum selection of palliative treatment options (timing and method) is crucial to maintain or improve quality of life and prolong survival. A key need is for the development of strategies that identify patient subgroups that would benefit from specific treatment and personalise management of neuroendocrine liver metastases.
To this end, the European-African Hepato-Pancreato-Biliary Association (E-AHPBA) initiated a consensus conference to address optimisation of management of neuroendocrine liver metastases. The aims were to: critically review the existing approaches to neuroendocrine liver metastases, assess the evidence on which management decisions were based, develop internationally acceptable recommendations for clinical practice (when evidence was available), and make recommendations for clinical and research endeavours.
Section snippets
Methods
The conference was organised by the E-AHPBA and seven national and international societies focused on liver diseases or NETs, and was held on Dec 12ā13, 2012, in London, UK. The Danish Consensus Conference model8 was used. 15 key questions about diagnosis and management were defined for assessment by 15 groups selected by the scientific committee on the basis of their expertise in the specific area. The recommendations presented by the working groups were debated in plenary sessions and
What are the incidence, prevalence, and prognosis of NETs and neuroendocrine liver metastases?
Despite decades of mandatory cancer registration, the precise incidence and prevalence and survival from NETs remain difficult to define. Nevertheless, the incidence of NETs recorded since 2000 is between 1Ā·9 and 5Ā·7 per 100ā000 people per year, about 60% of which are GEP NETs.1 The common GEP NET primary sites are small intestine (about 30%), rectum (about 15%), colon (about 13%), pancreas (about 16%), stomach (9%), and appendix (about 20%).1 Distant metastases at diagnosis are found in about
Should patients with a low Ki-67 index be followed up after resection of the primary tumour for the detection of liver metastases?
No published data specifically address this question. The Ki-67 index is known to strongly correlate with patient survival (particularly for pancreatic NETs) and is prognostic for pancreatic lesions.13 However, NETs from other sites (eg, ileum) have a high probability of developing liver metastases despite a low (<2%) Ki-67 index.14, 15, 16 Additionally, tumours from specific organsāeg, stomach, appendix, rectumārarely metastasise.1 Gastric and rectal NETs are usually diagnosed serendipitously
Should genetic signature and the presence of circulating tumour cells be used to predict liver metastases and to inform treatment decisions?
Gene signatures derived from transcriptome studies18 and the detection or quantification of circulating tumour cells, either with capture-based approaches or real-time PCR (so-called liquid biopsies), have been done in NETs. In four studies in small intestinal NETs, different sets of genes were identified to be differentially expressed between primaries and metastases, including NAP1L1 and MTA1,19 CXCL14 and NKX2-3,20 REG3A and TGFBR2,21 and CD302.22 MTA1 has been confirmed to be over-expressed
Which biochemical markers should be used for detection and post-treatment follow-up of liver metastases?
Chromogranin A is the most widely used biomarker. Plasma chromogranin A is increased in neuroendocrine liver metastases, and concentrations generally correlate with hepatic NET burden. Chromagranin A concentration correlated with hepatic burden (when assessed as <25%, 25ā50%, >50%) and survival.31 Increases in chromogranin A were associated with tumour progression and shorter survival. Chromogranin A concentrations are reduced after hepatic resection or transplantation.32 Few studies have
Which morphological imaging method should be used to assess resectability of liver metastases with a curative intent?
The morphological imaging methods used for assessment of hepatic metastases include conventional ultrasonography or a contrast-enhanced ultrasonography technique, CT, and MRI. A mixed hyperechoic and hypoechoic pattern with central cystic appearance and hypervascularity on colour Doppler imaging are characteristic ultrasonography features of neuroendocrine liver metastases.35
Contrast-enhanced ultrasonography identifies significantly more hepatic metastases with a higher specificity than does
Which functional imaging method should be used to assess resectability of hepatic metastases with a curative intent?
NETs variably express somatostatin receptors (60ā100% of tumours, about 85% are somatostatin subtype receptor 2) thereby providing a target for functional imaging with labelled somatostatin analogues. Indium-111 (111In)-octreotide scintigraphy has a lower sensitivity (69ā86%) and a higher cost for detection of GEP NETs than do PET or CT using gallium-68 (68Ga)-labelled somatostatin analogues (DOTATOC, DOTATATE or DOTANOC).41
A novel isotope, copper-64 DOTATE could be more sensitive than
Is a biopsy of both the primary and liver metastases needed for the treatment decision on liver metastases?
Therapeutic GEP NET decision making is based predominantly on the grade of the tumour. About 50% of GEP NETs show distant metastasis, including liver metastases, at diagnosis.45 The National Comprehensive Cancer Network46 and the European Society for Medical Oncology47 guidelines do not address liver metastases sampling or consider multiple biopsies, and suggest that Ki-67 assessment is optional. No difference in Ki-67 expression between primaries and metastases has been reported.48 However,
When should a liver resection be done?
Resection of neuroendocrine liver metastases, primary tumour, and locoregional lymph node metastases is thought to positively benefit long-term survival and quality of life (figure 3). The overall survival after hepatic resection is 46ā86% at 5 years and 35ā79% at 10 years.51, 52, 53 These results should be viewed with caution (complete resection in only 20ā57% and local recurrence evident in up to 94% at 5 years).54 Candidates for resection include: grade 1 or 2 tumours; when no evidence of
Should the primary tumour be resected in the presence of non-resectable liver metastases?
Resection of an asymptomatic primary NET in the presence of unresectable hepatic metastases is controversial. No randomised controlled trials exist. Unresectable liver metastases are present in about 15ā80% of GEP NETs.45 Positive aspects of resection are the prevention of local symptoms induced by tumour mass (pain, bleeding, perforation, obstruction), amelioration of hormonal symptoms, and a positive effect on survival.56 In a single retrospective series of small intestinal NETs,15 survival
When should a liver transplantation be done?
Neuroendocrine liver metastases are an accepted indication for liver transplantation, because most show low biological aggressiveness and grow slowly. Experience is scarce because liver transplantation for NET disease represents 0Ā·3% and 0Ā·2% of transplants (European Liver Transplant Registry and United Network for Organ Sharing database, respectively).58 Assessment is hampered because disease-free survival is not routinely reported, follow-up is not uniform, and no studies directly examine
Should neoadjuvant and adjuvant treatment strategies be used?
Neoadjuvant and adjuvant treatment is not considered for hepatic resection. In a retrospective cohort study (59% pancreatic NETs), no difference in survival was noted between the treatment (streptozotocin and fluorouracil) and non-treatment groups after resection of liver metastases (or transplantation).59 Findings from smaller series and case reports indicate that downstaging of neuroendocrine liver metastases with immunochemotherapy or peptide receptor radionuclide therapy (PRRT), or both,
When should locally ablative techniques be used?
Locally ablative methods have been used extensively in treatment of secondary hepatic malignancies alone or as an adjunct to surgical resection. The methods include radiofrequency ablation, microwave ablation, laser ablation, and cryotherapy, and can be done with percutaneous, open, or laparoscopic approaches. Microwave ablation is thought to be more efficacious than radiofrequency ablation because a shorter time is needed for each ablation and higher intratumour temperatures can be reached.63
When should angiographic liver-directed techniques be used?
Liver-directed intra-arterial therapies for the treatment of unresectable neuroendocrine liver metastases include transarterial embolisation, transarterial chemoembolisation, and selective internal radiotherapy with yttrium-90 (90Y)-microspheres. Transarterial embolisation or chemoembolisation produces symptomatic responses in 53ā100% of patients (10ā55 months) and morphological responses in 35ā74% (6ā63 months), with progression-free survival of about 18 months and 5 year survival of 40ā83%.63
When should peptide-receptor radionuclide therapy be used?
Most GEP NETs express somatostatin receptors, and treatment with 90Y or lutetium-177 (177Lu) somatostatin analogues is therefore feasible. PRRT has been extensively used since 1999.44 It is effective with about 75% stable disease and outcomes including progression-free survival (17ā40 months) and overall survival (22ā46 months) better than those with other methods. In a clinical phase 2 single-centre trial of 1109 patients, morphological response was evident in 34Ā·1%, biochemical response in
When should chemotherapy, targeted therapy, or biotherapy be used?
The type of therapy used is dependent on the grade and proliferation of the tumour. High-grade lesions, especially from the pancreas (neuroendocrine carcinoma [grade 3]), are amenable to chemotherapy (fluorouracil, doxorubicin, and streptozotocin). Targeted therapiesāeg, everolimus or sunitinibāand biotherapyāsomatostatin analogues or interferonāare used in slow-growing lesions (NET grade 1 or grade 2). Objective response rates (35ā40%) in pancreatic neuroendocrine tumours79, 80 are higher with
Conclusions
The recommendations (panel 1) and research proposals (panel 2) from this consensus meeting represent current knowledge of the management of neuroendocrine liver metastases. They present a rationale for therapeutic strategy and serve as a basis for the development of clinical and research programmes necessary to advance the specialty. A crucial need identified was delineation of cellular and molecular indicators of metastatic growth and blood, urinary, and tissue biomarkers of neuroendocrine
Search strategy and selection criteria
References (85)
- et al.
The epidemiology of gastroenteropancreatic neuroendocrine tumors
Endocrinol Metab Clin North Am
(2011) - et al.
Gastroenteropancreatic neuroendocrine tumours
Lancet Oncol
(2008) - et al.
Progression and survival results after radical hepatic metastasectomy of indolent advanced neuroendocrine neoplasms (NENs) supports an aggressive surgical approach
Surgery
(2011) - et al.
GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables
J Clin Epidemiol
(2011) - et al.
Novel markers for enterochromaffin cells and gastrointestinal neuroendocrine carcinomas
Mod Pathol
(2009) - et al.
MicroRNA expression in ileal carcinoid tumors: downregulation of microRNA-133a with tumor progression
Mod Pathol
(2010) - et al.
Plasma chromogranin A as marker for survival in patients with metastatic endocrine gastroenteropancreatic tumors
Clin Gastroenterol Hepatol
(2008) - et al.
Neuroendocrine gastro-entero-pancreatic tumors: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up
Ann Oncol
(2012) - et al.
Ki-67 proliferative index predicts progression-free survival of patients with well-differentiated ileal neuroendocrine tumors
Hum Pathol
(2012) - et al.
Ki-67 proliferative index predicts progression-free survival of patients with well-differentiated ileal neuroendocrine tumors
Hum Pathol
(2012)