Systemic chemotherapy may be indicated for metastatic or rapidly progressive disease, although it is of limited utility. The most common chemotherapeutic agents used in NETs are alkylating agents including streptozocin and temozolide; however, they have significant renal and hematologic toxicity. They may be most effective in pancreatic primaries and in combination with fluorouracil (5-FU), where they have a high response rate and improved progression-free survival. In one study of patients with advanced pancreatic primaries, streptozocin plus 5-FU had a 33% complete response rate and 26-month median survival compared to 12% and 16.5 months in streptozocin alone.23 However, patients with metastases involving >75% of the liver had worse outcomes than those with <75% involvement.24 Similarly, temozolamide with capecitabine had a 70% response rate and progression-free survival of 18 months with good tolerance in patients with pancreatic primaries.25
Gastrointestinal or midgut NETs appear to be very chemoresistant. High-grade (NEC G3) cancers respond best to platinum-based therapy, with carboplatin plus etoposide and/or paclitaxel being the most common regimens, and may be more effective with a Ki67 of greater than 55%.26,27 Interferon alpha also may have some efficacy in both relief of symptoms and survival in patients with NEC G3 disease, but it has been limited by drug toxicity.24,28 A progressive randomized trial comparing interferon to lanetoide and combination therapy in 80 patients found low response rates for all arms, with only one patient each in the single therapy groups and two in the combination groups having a partial response. Fifty percent to 55% of patients in each arm had progression of disease, and side effects were more frequent with combination therapy.29
For less aggressive disease, everolimus and sunitinib are targeted therapies for advanced NETs. Everolimus is a mammalian target of rapamycin inhibitor, and is used alone or in addition to octreotide. The well-known RADIANT trials compared everolimus plus octreotide to placebo in pNETs. The second trial found a progression-free survival of 16.4 months compared to 11.4 in the control, with improved survival in pNETs. Radiant 3 looked at NET G1 and G2 pNETs and compared everolimus alone versus a placebo and also demonstrated an increase in progression-free survival.30,31 The use of everolimus in non-pNETs is less well studied. Sunitinib is a tyrosine kinase inhibitor that targets vascular endothelial growth factor (VEGF) and acts as an angiogenesis inhibitor. A phase III trial in low-to-intermediate grade pNETs showed a PFS of 11.4 months compared to 5.5 months for placebo.32 Pazopanib, a similar drug, has similar results in GEP NETs.33
Somatostatin analogues (SSAs) were initially used to treat symptoms in patients with functional tumors, but have some antiproliferative benefits in patients with octreotide avid disease.34 In a study of 33 patients with carcinoid syndrome, lanreotide was compared to octreotide with crossover at 1 month. Flushing and diarrhea improved equally in both groups (53.8% and 45.4% in lanreotide compared to 68% and 50% in octreotide); however, the patients preferred the easy delivery of lanreotide because of easier delivery.35 A randomized controlled trial in patients with metastatic midgut and pancreatic NETs (the CLARINET trial) demonstrated a significant increase in progression-free survival in patients with grade I and II tumors using lanreotide compared to placebo. Regardless of hepatic tumor burden36 there was an increase in progression-free survival in pancreatic primaries as well. SSAs appear to be indicated in patients with hormonally active midgut or pancreatic SSAs, and have efficacy in preventing disease progression in patients with metastases.
Radiolabeled SSAs (also referred to as peptide receptor radionuclide therapy (PRRT)) including DOTA, DOTATOC, and DOTATATE labeled with 177 LU and 90 Y have been shown to offer response (10% to 33%) in patients with hormonally active tumors and metastatic disease.37,38 Similar to other treatments, they are less effective with a high hepatic tumor volume and poorly differentiated tumors (NEC G3).39,40 In 310 patients, 89% with hepatic metastases, treated with 177-LU-DOTATATE, 2% had a complete response and 28% a partial response with median overall survival of 46 months. Interestingly, a combination of 90-Y-DOTATOC and 177-LU-DOTATOC in patients with metastatic disease leads to a survival of 5.51 years compared to 3.96 years alone, possibly due to synergy between the two isotopes.38 90-Y-DOTALAN (DOTA lanreotide) and 90-Y or 177-LU DOTAOTC injected intra-arterially has shown promise in treating large burden hepatic metastases.41,42
And 90-Y-DOTATOC has been successfully used to downstage hepatic metastases for resection.43 PRRT appears to have the most efficacy in patients with receptor avid disease, and grade I or II hepatic metastases.
Surgery is the treatment of choice for neuroendocrine metastases to the liver, as it is the best opportunity to control symptoms and for long-term survival. In addition, unresectable patients with functional tumors or large tumor burden may benefit from debulking. Patients with type I or II metastatic disease are most likely to be candidates for curative resection, while only 20% to 30% of patients with type III disease are resectable.5 Similar to extended liver resections for other disease (see Chapter 115, "Metastatic Colorectal Cancer"), both portal vein embolization to stimulate remnant liver lobe hypertrophy as well as two-stage liver resections may be occasionally indicated for the resection of extensive bilobar disease, though the majority of cases that require extensive hepatectomy are done for multifocal disease that is not amenable to “curative” therapy but is being done as a debulking operation. In addition, liver resection combined with ablative therapy is also indicated in some patients. A general rule of thumb is that resection or resection with ablation should be the first line of treatment if greater than 90% of the disease can be treated.44 In select cases, synchronous resection of the primary and liver metastases may be indicated.45
It is difficult to assess the outcomes of liver resection for metastatic NETs, as there are no prospective randomized trials, and most of the series are single center with a heterogeneous group of disease and presentations. However, most of these series demonstrate a survival benefit to resection of NET metastases, especially in patients with type I and II diseases.46-49 In addition, patients with functional tumors can have resolution or improvement of symptoms, which leads to improved quality of life even if complete resection is not feasible.44 A retrospective multi-institutional analysis of 339 patients from eight centers offers a comprehensive review of outcomes of liver resection for NET metastases. Survival at 5 and 10 years was 75% and 51% in this series; however, recurrence-free survival was 57% at 1 year and only 6% at 5 years.50 Similar results were found in a recent systematic review of 29 series on liver resection for metastatic NET. The median survival at 1 year, 5 year, and 10 years was 94%, 70.5%, and 42% and R0 and R1 resection was completed in 63% of the resections. However, the recurrence-free survival rates were much lower. At 1, 5, and 10 years, median recurrence-free survival was 63%, 29%, and 1%.51
Liver-directed therapies include radiofrequency ablation (RFA),52,53 microwave ablation (MA),54 alcohol ablation, cryoablation,55 transarterial embolization (TAE) or chemoembolization (TACE),56 or yttrium-90 radioablation. They can be used alone or in combination with surgical resection57 for relief of symptoms, and in some cases may prolong survival.58 In general, local-regional therapies are reserved for symptomatic patients with unresectable disease, often multiple diffuse small bilobar metastases (type III disease) or as adjunct to surgical resection. However, some ablative therapies have been used as a stand-alone procedure.
Similar to hepatic resection, ablative therapies with curative intention work best in patients with low tumor burden, dominant liver lesions <5 cm, and ablation margins of > 1 cm.5 In a series of 80 patients who underwent laparoscopic RFA for liver metastases, 57% were symptomatic, and 70% had complete symptom relief for a duration of 11 months, with a 3.9-year median survival after RFA.53 Another study of 119 patients who underwent RF ablation found a 97% rate of symptomatic relief, but 22% developed hepatic recurrence, 6% developed new liver lesions, and 59% developed extraheptic disease. Repeat treatment was completed in 34% of patients, and overall survival was 6 years and disease-free survival 1.3 years. Morbidity and mortality were low, at 6% and 1%.59 In general, ablative techniques have a role as an adjunct to surgery, in patients with a low burden of disease, primarily for symptomatic relief and in general have a low risk of complications.
Transarterial embolization and TACE have been used extensively in the treatment of NET metastases to the liver, especially in patients with unresectable disease for symptom relief. A comparison of TACE, TAE, and RFA found that TACE and TAE were similar with symptomatic response rates of 65% to 95%, progression-free survival between 18 and 88 months, and survival rate of 40% to 67%. RFA had a symptomatic response of 71% to 95% and survival mean of 1.6 years after ablation, compared to surgical resection, which had a mean survival of 4.26 years. The authors of this study concluded that for patients with oligonodular disease, local resection or RFA or both are recommended, but for patients with high tumor load TACE or TAE is preferred.60
There are a few studies that compare the outcomes of surgical resection to local regional therapies.61,62 A multi-institutional study looking at 753 patients from nine institutions looked at surgical resection compared to transarterial embolization. The comparison groups were not equal, with the patients in the embolization group having a greater burden of disease and more hormonally active tumors. Not surprisingly, median survival in the resection group was 123 months compared to 34 months in the embolization group. However, for asymptomatic patients with >25% tumor burden there was no difference in long-term outcome. They concluded that resection is indicated for symptomatic patients and patients with a tumor burden less than 25%, and asymptomatic patients with a high tumor burden would be better off with transarterial embolization.63 A similar study looked at 120 patients in which 59 underwent embolization and 61 underwent operative cytoreduction, including 23 patients for palliation only. In the operative cytoreduction group, 69% had complete symptom relief and 32% partial, compared to 59% and 23% in the embolization group (p = 0.08); however, the mean duration of relief was significantly loner in the cytoreduction group (35 months vs. 24 months, p = <0.01). Survival was significantly higher in patients who underwent curative resection.64
The combination of resection, ablation, and embolization (referred to as aggressive treatment) has been used successfully in select patients. The combination of resection, ablation, and transarterial chemoembolization was studied in 60 patients with NET metastases to the liver, for which 23 received no treatment (nonaggressive) and 37 received treatment (all with resection/ablation, while 18 also had transarterial chemoembolization).65 Overall, survival was significantly better in the aggressive group; however, patients who had >50% of their liver involved with metastases had a poor outcome regardless of treatment. In general, liver resection is the gold standard, but ablative therapies have a role in type I or II tumors (especially in poor surgical candidates) or as an adjunct to surgery. TAE, TACE, and Y-90 TheraSpheres have a role for unresectable type III disease to both treat symptoms and slow progression, and may be a less invasive option with the same or better outcome as resection.
Patients with unresectable hepatic metastases (type III) may be considered candidates for liver transplantation if they meet the following criteria: prior resection of the primary tumor, no evidence of extrahepatic disease, no evidence of significant cardiac valvular disease, and a well-differentiated tumor (Ki67 less than 10%).66 With careful patient selection, many series have shown reasonable long-term disease-free survival.67-70 Patients with carcinoid tumor have increased long-term and disease-free survival when compared to patients with pNET.66,71 Some authors discourage liver transplantation for pancreatic and duodenal primaries with hepatomegaly as a marker of extensive hepatic disease.72
Although there are no prospective randomized controlled trials, there are several large multicenter or database studies that examine the outcomes of liver transplantation for metastatic NET. An analysis from the United Network for Organ Sharing (UNOS) database looked at 150 patients who underwent liver transplantation for NET metastases, including 13 patients who underwent a multivisceral transplant. For the group who had an isolated liver transplant, 1-, 3-, and 5-year survival was 81%, 65%, and 49% and there was no difference between carcinoid and noncarcinoid tumors.69 In addition, patients who waited longer on the wait-list had better outcomes. A similar retrospective analysis of 213 patients who underwent liver transplantation over a 25-year period found 1-, 3-, and 5-year survival of 81%, 65%, and 52% with a disease-free survival rate of 30% at 5 years. Seventeen patients died from transplant-related complications while recurrence free.73 In the United States these outcomes are comparable to liver transplantation for hepatocellular carcinoma, which is well established as an indication for liver transplantation.68,69
In patients with unresected primaries, several groups have investigated the combination of liver transplantation with primary resection and multivisceral transplantation in limited series. The combination of liver transplantation and pancreatic resection has poor outcomes.74 Indications for multivisceral transplantation include unresected or unresectable primary with extrahepatic metastatic disease. The results of multivisceral transplantation for metastatic NET are mixed, with some groups reporting outcomes equivalent to liver transplant alone69,75 and others reporting high morbidity, mortality, and recurrence.67,76