The neuropeptide somatostatin is one of the major regulatory peptides in the central nervous system and digestive tract. Somatostatin acts by inhibiting the secretion of various hormones and peptides. For this reason, SSAs remain the mainstay of treatment of symptomatic GEP-NETs. Due to their short half-life (<2 min), native SSAs are not currently used. Nevertheless, the SSAs long-acting octreotide and lanreotide autogel are characterized by long-acting formulations which allow monthly administration. Long-acting octreotide and lanreotide autogel bind mainly to the somastostatin receptor subtypes 2 (SSTR2) and 5 (SSTR5) and provide symptomatic as well as biochemical responses in ≤75% of patients [18]. Recently, pasireotide (a SSA with high affinity for all types of SSTRs) was approved for acromegaly treatment. Although early data suggest that this drug is effective in patients affected by NETs not responsive to currently available SSAs, the treatment of NETs with pasireotide is restricted in clinical trials [19]. SSAs are safe, well-tolerated and easy to use. In retrospective experiences, SSAs provided a 5% partial response (PR) and 50% stable disease (SD).

Rinke et al. randomized 85 treatment-naive patients with metastatic well differentiated tumors in the midgut to receive placebo or octreotide LAR 30 mg monthly until tumor progression. The primary endpoints were time to progression (TTP) and response rate (RR). The study was designed for the enrollment of 162 patients in 18 academic centers in Germany, but it was stopped early after the enrollment of 85 patients (43 received octreotide LAR 30 mg and 42 received placebo) due to the results of the interim analysis. In fact, median TTP was 15.6 months in the octreotide LAR group vs 5.9 months in the placebo group (hazard ratio (HR) 0.34; 95% confidence interval (CI), 20–0.59; p=0.000072) [20]. The authors demonstrated, in a preplanned subgroup analysis, that patients with low (0% to 10%) liver involvement and resected primary tumor benefited most from the treatment (HR, 0.17; 95% CI, 0.08–0.40; and HR, 0.16; 95% CI, 0.07–0.36, respectively). This prospective trial suggested an anti-tumor activity of SSAs in patients with well-differentiated midgut tumors. There is no prospective evidence supporting the anti-tumor activity of SSAs in pNETs but, in our opinion, the first-line treatment of patients affected by wellor moderately differentiated, low proliferating pNETs should consist of SSAs rather then chemotherapy. The Study of Lanreotide Autogel in Non-functioning Entero-pancreatic Endocrine Tumors (CLARINET) is an ongoing randomized trial which aims to define this issue. In this study, patients with G1 and G2 nonfunctioning pNETs and midgut NETs are randomized to lanreotide autogel (120 mg) monthly or to placebo. The primary endpoint is TTP. The enrolment is complete and data are expected within the next months.

INFα monotherapy has revealed similar efficacy to SSAs for controlling symptoms and inducing biochemical responses (80% of patients). Nevertheless, INFα has a different tolerability profile, with more frequent high-grade toxicity, such as flu-like syndrome, fatigue, weight loss, polyneuropathy, myositis, thrombocytopenia, anemia, leukopenia and hepatotoxicity as well as other side effects [21]. There is evidence that combination therapy (interferon alpha plus SSAs) does not increase RR but leads to a significantly lower risk of progression compared with SSAs alone [22].

The role of chemotherapy in the treatment of pNET is controversial. Based on retrospective studies, systemic chemotherapy with cisplatin and etoposide is the standard of care for poorly differentiated NECs with Ki-67 >20% (Fig. 9.1). This platinum-based combination provides a high RR, but the prognosis of this group of patients is poor (median survival, ≈20 months) [23, 24]. The recently published NORDIC NEC study retrospectively analyzed 305 patients affected by metastatic GI-NECs or unknown primary tumor with gastrointestinal metastases [25]. In this large retrospective study, patients with Ki-67 <55% were less responsive to platinum-based chemotherapy but survived longer than patients with Ki-67 >55%. The authors concluded that Ki-67 <55% represents a favorable prognostic factor and a negative predictive factor for the efficacy of platinum- based chemotherapy. Therefore, different from the present WHO classification, we should not consider GI-NECs as a single disease.

Several studies suggest that pNETs are responsive to chemotherapy (Tables 9.4 and 9.5).

Streptozotocin is an alkylating agent that has been studied in combination with 5-Fluorouracil (5FU) and doxorubicin in well-differentiated NETs. Notably, randomized trials comparing first-line chemotherapy vs best supportive care (BSC) are lacking. In a randomized trial, streptozotocin plus doxorubicin compared with streptozotocin plus 5FU significantly improved the RR (69% vs 45%), progression-free survival (PFS) (20 months vs 6.9 months) and OS (26 months vs 18 months) [26]. On the basis of these results, streptozotocin was approved by the US Food and Drug Administration (FDA) for the treatment of pNETs. Nevertheless, the authors of this trial used non-standard criteria for response evaluation, and assessed biochemical response rather than objective tumor response as defined by Response Evaluation Criteria in Solid Tumors (RECIST) criteria. Moreover, in two retrospective analyses, streptozotocin plus doxorubicin showed mild anti-tumor activity with a RR of 10% [27, 28]. Dacarbazine monotherapy has been evaluated in a Phase II study enrolling 50 patients with pNETs. In this trial, dacarbazine provided a RR of 34% and OS of 19.3 months [29]. Nevertheless, due to its toxicity profile, dacarbazine is not frequently used in the treatment of pNETs. Temozolomide is an alkylating agent developed as an oral and less toxic alternative to dacarbazine [30]. In a retrospective analysis, Ekeblad et al. evaluated the efficacy and toxicity of temozolamide monotherapy in advanced NETs. The RR was 14% and SD was noted in 53% with a global disease control rate (DCR) of 67% [31]. The median TTP was 7 months. With respect to RR, 1 patient among the 14 affected by GI-NETs and 5 patients among the 11 affected by pNETs showed a radiological response. This observation supports the fact that well-differentiated pNETs are more responsive to systemic chemotherapy than NETs arising from the stomach, small intestine and large intestine. Temozolamide has been studied in patients affected by advanced NETs in combination with different anti-cancer drugs such as thalidomide, capecitabine and bevacizumab. In a Phase II trial, 41 patients affected by well-differentiated NETs were treated with a temozolamide plus thalidomide combination. In this trial, the authors observed a 45% RR (including 1 complete response (CR)) among the 11 patients affected by pNET [32]. The capecitabine and temozolamide combination is also effective in pNETs. In a single-institution retrospective experience, 30 patients with unresectable/metastatic pNETs were treated with capecitabine and temozolamide. The authors observed a RR of 59% and a CR in 6% [33].