**2. Liver Metastases from Neuroendocrine Tumors**

2 Recent Advances in Liver Diseases and Surgery

Cancer metastasization is a highly selective, sequential, interdependent, nonrandom process which causes transient or permanent changes in different genes at the DNA and/or mRNA, creating a complex phenotype which favors the survival of a population of tumor cells within an organ environment, distant from the primary tumor [1, 2]. This kind of complex process usually requires many years in order to complete the great series of cumulative DNA changes which consent neoplastic cells to metastasize. As the primary tumor growth may be very slow and the metastasization process does not directly depend on primary tumor size, we can observe two kinds of metastases based on the timing of their diagnosis: synchronous metastases are defined as secondary lesions diagnosed within the first year after primary tumor diagnosis, although this time, interval may vary in the literature between six months and one year, while metachronous metastases are usually diagnosed after one year from the

Taking into consideration all primary tumors, the liver represents the most common site of the distant metastases. This particular affinity for distant metastases may have different explanations. First, it may be attributable to the site specificity patterns of neoplastic cells from various primary cancers, which found the liver as the adequate soil where to seed their circulating neoplastic cells [3, 4]. Furthermore, the liver receives 30% of the whole cardiac output, second only to the kidney for the quantity of blood which every day will perfuse its parenchyma, and is consequently more susceptible to neoplastic cell attachment. Moreover, anatomical or mechanical considerations, such as the efferent venous blood stream or the loco-regional lymphatic drainage, may be strongly responsible for this preference in the metastasization site [5]. In fact, most intra-abdominal cancers, and in particular those which originated in the digestive tract, result in having a great affinity for liver metastasization. In spite of the great progresses of surgical techniques against the primary tumors, as well as the improvement of adjuvant therapies, metastatic disease continues to be the greatest challenge for the medical and the surgical oncologists. In fact, metastases are well recognized as being the major cause of death among neoplastic patients, and the prognosis of patients affected by unresectable liver metastases is very poor. However, although once metastatic malignancies were commonly considered as a terminal neoplastic stage, nowadays, many different therapeutical options have been introduced in order to provide a safe and efficient treatment for these kinds of patients and improve both their quantity and quality of life [6–9]. Despite the abundant literature about colorectal liver metastases and the existence of a great number of guidelines about this argument, there is still great debate about the treatment strategy in the case of non-colorectal ones and controversies especially about the management of rare liver secondaries. On the other hand, a recent review about non-colorectal non-neuroendocrine liver metastases demonstrated surgery to be a benefit for these kinds of metastatic patients, especially for those affected by primary testicular, ovarian, and renal cell cancers but also for women with isolated breast cancer metastases to the liver

Obviously, patients affected by rare hepatic metastases should be conveyed into bigger and more experienced centers, which could be able to more appropriately treat this kind of disease. However, metastatic patients are always more numerous – thanks to the improvement of chemotherapy and the introduction of targeted biological drugs – and

**1. Introduction**

primary tumor diagnosis.

[10].

Neuroendocrine tumors (NETs) include a heterogeneous group of neoplasms with different origins and biological behaviors [11–13]. They are commonly distinguished into two classes based on their primary origin: pancreatic neuroendocrine tumors (also known as islet cell tumors) and gastrointestinal neuroendocrine tumors (also known as carcinoids) [14–16].

The widely promulgated benignity of these neoplasms has been more times brought into question, due to the high prevalence of distant metastases and recurrences, and the current literature recognizes nowadays their malignant potential [12, 17–24]. In particular, the recent classifications of the 7th American Joint Committee on Cancer/Union Internationale Contre le Cancer (AJCC/UICC) 2009 and of the European Neuroendocrine Tumor Society (ENETS) 2006, associated with the WHO classification 2010, classified NETs into well-differentiated neuroendocrine tumors (low and intermediate grade based on the Ki67 labeling index, also named NET-G1 or carcinoid and NET-G2, respectively) and into the group of neuroendocrine carcinomas (high grade, poorly differentiated, also named NEC) [21].

Despite their rarity, liver metastases from NETs are not an infrequent finding, because they are usually characterized by a very slow growth pattern and diagnosed in advanced stages [9, 15, 25–29]. In particular, NETs present with liver metastases in even the 50–75% of cases [30–32]. In an analysis of Modlin and colleagues on 13,715 patients, synchronous distant metastases were already evident in 12.9% of patients with gastrointestinal NETs, whose 5-year overall survival rate resulted 67.2% [13].

The occurrence of hepatic secondaries is one of the most important prognostic factor for survival [9, 21, 33, 34]. In fact, although for patients with unresectable liver disease, biotherapy with somatostatin analogues, peptide-mediated radioreceptor therapy, transarterial chemoembolization, selective intra-arterial radiotherapy, or new molecular target-directed therapy can be employed [16, 35, 36], these therapies are considered as palliative.

The role of liver surgery for patients with liver metastases from gastrointestinal NETs remains to be an argument of great debate. In particular, the very small number of patients affected by NETs explains the lack of randomized control trials in order to better define the role of surgery in these rare cases. Moreover, the usually inert growth of NETs and their long-term natural history make even more difficult to assess the real effectiveness of hepatic surgical approach on their overall survival. Furthermore, it is well known that in most patients, neuroendocrine liver metastases recur after hepatic resection in up to 70–94% of cases at 5 years [21, 24, 25, 33, 37–39], the liver is the most common site of progression of disease (69%) [35], and data on repeat liver directed surgery for recurrent disease have been extremely limited and controversial [36]. In particular, in a study of Saxena and colleagues, most patients with hepatic metastases from NETs experienced treatment failure after liver resection. In particular, 57 patients (79%) developed disease progression at a median time of 23 months, and the liver represented the most common site of progression of disease (69%) [23]. In another multi-institutional study on 339 patients, Mayo and colleagues demonstrated that the majority of liver metastases from NETs came from carcinoid (53%) and, at 5 years after surgery, their recurrence rate was 94% [38]. Therefore, the true curative role of liver-directed surgery results is still very questionable.

A recent meta-analysis performed on five studies, considering 374 patients affected by NETs liver metastases treated in a conservative manner and 161 treated with liver surgery, demonstrated a significant increased survival in the group of patients treated with surgical hepatic resections HR 0.45 (CI.95 0.34–0.60) in comparison to conservative treatments and to embolization HR 0.34 (CI.95 0.21–0.55) [40]. All considered studies showed an increased survival in the groups treated with complete surgical resection of liver metastases, but none of the included studies were randomized so that the clinical evidence was low [21, 41, 42].

A systematic review considering 29 studies (between 1980 and 2009) found 5 years OS of 70.5% (range 31–100%) and a 5-year progression-free survival of 29% (range 6–66%) [43]. Histological grade, extrahepatic disease, and macroscopically incomplete resection of liver metastases were associated with poor prognosis. In another multi-institutional study evaluating 339 patients, Mayo et al. demonstrated at multivariate analysis that synchronous disease, nonfunctional NET hormonal status, and extrahepatic disease were the most important predictive factors of worse survival [36]. Concerning other prognostic parameters for primary NETs and liver metastases, Katz et al. demonstrated that the robust presence of tumor-infiltrating lymphocytes is a significant predictor of outcome [44]. Other recent articles confirmed surgical therapy to be the most efficient approach against solitary hepatic metastases [7, 8, 14, 22, 23, 33, 34, 40, 45, 46], with a potential curative resection of liver secondaries can be undertaken in 13.7 to 24.5% of patients with metastatic NETs [47–49], and a significant reduction of carcinoid symptoms [43], but the majority of studies focused exclusively on resection rather than combined-modality approaches with ablation or chemotherapy.

Taking into consideration our institutional cases of liver metastases from NETs, including 52% of synchronous metastases and 48% of metachronous ones, the overall survival of patients who underwent hepatic resections and OLT resulted, respectively, in 44.9% (95% CI 26.0–77.7%) and 50% (95% CI 12.5–100.0%) at 5 years [45]. The median number of resected hepatic metastases was 3. Surgical radicalness (R0) was reached in 84% of cases. Recurrences happened in 60% of patients, among which, 66.7% were intrahepatic and 33.3% extrahepatic. Postoperative complications affected 12% of patients but required reintervention in a single case.

In our opinion, surgery should be always considered for patients with resectable hepatic disease, as this treatment results more likely to offer the best long-term outcome. All patients should be considered for curative surgical treatment, but also palliative resection of liver metastases can be suggested. The advantage that can be potentially achieved with surgery is that of removing all gross diseases. In the future, new clinical and biological prognostic factors could be of help for the better identification of those patients who might benefit from hepatic surgical therapy.
