Histological tumor classification is a cornerstone in NEN treatment. This classification has been recently modified with the last update in 2017 (Table 1). The main modifications were the ki67 cut off levels, which changed from 2 to 3%, and the concept of MANEC (mixed adeno-neuroendocrine carcinoma) that developed into MENEN/MINEN (Mixed endocrine non-endocrine neoplasms). The other novelty was the inclusion of high grade well-differentiated (WD) neuroendocrine tumor (NET G3). Previously, all tumors with a high proliferation index (Ki67 > 20%) were considered neuroendocrine carcinomas (NEC) regardless of the level of differentiation [6].

NETs can also be classified in two major categories: functioning and non-functioning. Functioning NETs produce a wide spectrum of clinical syndromes depending on which hormone is released. Non-functioning tumors, which correspond to 50–85%, may secrete some substances but they do not cause a hormonal syndrome [7].

On one side, functioning tumors cause different symptoms depending on the amine or peptide hormone released; insulinomas cause hypoglycaemias, which result in unusual behaviour, confusion, trembling, and diaphoresis. Gastrinomas are related to Zollinger-Ellison syndrome, causing peptic ulcer disease and diarrhea. Glucagonoma is associated to necrolytic migratory erythema, diabetes, and diarrhea. The clinical syndrome of VIPoma is watery diarrhea with hypokalaemia and hypochlorhydria [4].

On the other side, non-functioning (NF) tumors can be subdivided into three types: NF-P-NETs that do not produce hormones; NF-P-NETs that secrete hormones at a low enough level not to cause symptoms; and NF-P-NETs that produce hormones such as chromogranin, neuron-specific enolase, pancreatic polypeptide or ghrelin, which do not produce a clinical syndrome. The presence of these three subtypes may lead to a delayed diagnosis [8].

For unresectable or metastatic NETs, the treatment is multidisciplinary and includes systemic and loco-regional therapies. SSAs are nowadays the front-line therapy in NETs (Table 2): They are effective in palliating hormone-related symptoms and have demonstrated, in phase III trials, their role as antiproliferative agents, delaying disease progression in patients with NETs. Current practice guidelines are based on the results of the PROMID and CLARINET trials. The PROMID study randomized 85 patients with WD midgut NETs to receive placebo or octreotide LAR 30 mg intramuscularly in monthly intervals until progression or death. The study closed early due to slow accrual, but achieved the benefit in terms of time to progression (14.3 months in the octreotide arm vs. 6.0 months in the placebo arm; HR0.34, p < 0.001) [9]. This study was the first trial to demonstrate the antitumoral effect of SSA in patients with WD midgut G1 NET (Ki67 1–2%). In this sense and to assess the benefit of SSA in a wider population, the CLARINET trial evaluated the role of lanreotide depot 120 mg every 28 days compared with placebo in patients with advanced WD GEP-NET and a Ki67 < 10%. The study was positive for the experimental arm, with a primary endpoint of median PFS not reached (65.1% vs. 33% in the placebo group at 2 years) [10].

Even though SSAs remain the first line treatment of most patients with unresectable WD NETs, the majority of patients will eventually experience disease progression. Ongoing research about the use of peptide receptor radionuclide therapy (PRRT) with radiolabelled SSAs, such as octreotide or octreotate in P-NETs, are showing promising results in somatostatin receptor (SSTR)-expressing tumors. This therapy is composed by a carrier molecule (SSA), a radionuclide isotope (¹¹¹In, ⁹⁰Y and ¹⁷⁷Lu), and a chelator [Tetra-azacyclododecane-tetra-acetic acid (DOTA) and diethylenetriamine penta-acetic acid (DTPA)] that binds and stabilizes the complex. So far, the most satisfactory results have been obtained with ¹⁷⁷Lu-based radiolabelled SSAs, based on the phase III trial NETTER 1 [11]. However, research is going further, looking for radionuclides that offer higher affinity, greater efficacy, and less toxicity [12].

Chemotherapy also has a role in the treatment of patients with advanced WD NETs. The use of streptozocin (SZT) alone or in combination with doxorubicin or fluoropirimidines has been the standard chemotherapy approach for many years [5]. Temozolomide has also shown activity among P-NETs in combination with other drugs, such as thalidomide and capecitabine. In a study where all subtypes of NETs were included, the efficacy of the combination of capecitabine and temozolomide (CAPTEM) in metastatic P-NETs was demonstrated [13]. There is an ongoing phase II clinical trial using the above-mentioned combination, with a median PFS in the P-NETs cohort of 18.2 months [14].

The role of chemotherapy in the treatment sequence of patients with P-NETs is currently under research in the SEQTOR trial (NCT02246127) that randomizes patients to receive everolimus followed by STZ-fluorouracil (5-FU) vs. STZ-5FU followed by everolimus. In addition, the preliminary results of an ongoing phase II trial comparing CAPTEM versus temozolomide alone in advanced P-NETs, reported a prolonged PFS (22.7 months) [15]. However, these results may be influenced by an imbalance between the two arms, since P-NETs in the CAPTEM subgroup had a lower grade [16].

Targeted agents have also broken in the therapeutic landscape of patients with P-NETs. Considering the expression of vascular endothelial growth factor receptors (VEGFR), platelet-derived growth factor receptor (PDGFR) α and β, and stem cell factor receptor (c-kit) in P-NETs, antiangiogenic agents such as sunitinib, pazopanib, cabozantinib, or lenvatinib are or have been under research in this group of patients showing considerable antitumoral activity. Sunitinib is the only one approved for P-NETs treatment at the moment (Table 3) [5]. The mTOR pathway is proven to be involved in tumor development and progression of P-NETs by over-activating mTOR. Because of this reason, everolimus, a rapamycin analogue mTOR inhibitor, has been investigated in this group of patients based on preclinical data. Indeed, the phase III randomized clinical trial RADIANT 3 showed that everolimus had an impact in median PFS (11.4 vs. 5.4 months; HR 0.35, p < 0.0001) when compared to placebo in the treatment of advanced P-NETs [17].

Waiting for definitive biomarkers to guide therapeutics decisions, the treatment algorithm in P-NETs depends on histological classification, radiological images, and clinical symptoms, as well as patient comorbidities. In this sense, treatment selection at each moment of the patient disease is crucial in order to obtain the maximal benefit with every agent administered. Future and current translational research may help in deciding which treatment strategy may be the most effective in a more accurate manner. [24].

Nevertheless, initial approaches are aimed to define molecular pathways involved in tumor development and resistant mechanisms to previous therapies that may guide therapeutic decisions. As we have previously mentioned, angiogenesis is a key path in P-NETs progression, and antiangiogenic agents, such as sunitinib, have obtained a significant survival benefit. In this sense, the improvement in the biological knowledge of resistance mechanisms to antiangiogenics would be relevant for novel therapies development that may improve patients’ survival.