If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Pancreatic neuroendocrine tumours (pNETs) are rare and the majority of patients present with advanced disease. Such patients have limited treatment options. We conducted a systematic review of published clinical trials of non-surgical interventions in pNET, to understand the efficacy, safety and health related quality of life (HRQoL) outcomes from the current evidence base.
Methods
Electronic databases and manual bibliographic searches were conducted to identify relevant studies. Data were extracted by two independent reviewers.
Results
Forty seven clinical studies met the predefined inclusion criteria. The following interventions were included: targeted therapies (two RCTs and six single-arm studies), chemotherapy (two RCTs, one prospective nonrandomised, comparative study and 14 single-arm studies);somatostatin analogues (SSA) and radiolabeled SSA therapies (nine single-arm studies), liver-directed therapies (six single-arm studies), mixed treatment regimens (one RCT, four single-arm studies) and other interventions such as interferon and recombinant human endostatin (one single-arm study for each). The paucity of RCT data and lack of consistency in reporting validated study outcomes and differing patient inclusion criteria between studies made it difficult to compare the relative efficacy of therapies.
Discussion
The majority of published studies assessing treatment regimens for the management of pNET are single arm, non-randomised studies, often enrolling a small number of patients and not reporting clinically meaningful outcomes. However data from recently conducted studies assessing targeted therapies indicate that it is possible to conduct adequately powered RCTs reporting standardised oncological endpoints in this rare cancer. Further, similarly robust studies should be conducted to define the optimal treatment algorithm.
Pancreatic neuroendocrine tumours (pNETs) are rare. Worldwide, the annual incidence of pNET is estimated to range from 0.2 to 0.4 per 100,000 population, although due to the relatively indolent nature of these tumours the true prevalence may be much higher [
]. At presentation, 65% of patients have unresectable or metastatic disease. The 5-year survival rate of patients with metastatic disease is 30–40% and has not changed significantly over the last 30 years [
Clinically, pNETs are divided into two groups: functional (10–30%) or non-functional (70–90%). Functional pNETs secrete biologically active peptides, or hormones producing one of nine recognised specific hormonal syndromes. These tumours are associated with a reduced quality of life (QoL) in patients [
Fluorouracil, doxorubicin, and streptozocin in the treatment of patients with locally advanced and metastatic pancreatic endocrine carcinomas. [Erratum appears in J Clin Oncol 2005 Jan 1;23(1):248].
Fluorouracil, doxorubicin, and streptozocin in the treatment of patients with locally advanced and metastatic pancreatic endocrine carcinomas. [Erratum appears in J Clin Oncol 2005 Jan 1;23(1):248].
]. The hormonal syndromes are associated with diverse clinical features with regard to both metastatic potential and survival. For example 10% of patients who present with an insulinoma will develop metastases, compared with 50% of those with somatostatinoma and up to 70% of patients with VIPoma [
Surgery, where possible, is considered the first-line treatment for pNET patients. Due to the presence of distant metastatic disease or local extension of the tumour, surgery is often non-curative, but even in advanced cases surgical debulking of disease can reduce symptoms related to tumour burden and hormone production [
]. For patients who are not candidates for surgery, the choice of treatment depends on the stage of the disease, symptoms and histological features of the tumour [
]. Treatment options include SSA and liver-directed therapies (for example, chemoembolisation, radioembolisation, arterial embolisation and radiofrequency ablation, which are palliative options for liver-dominant disease) [
ENETS Consensus Guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system: well-differentiated pancreatic non-functioning tumors.
]. In clinical practice, systemic chemotherapy is commonly used in the treatment of pNET, but with modest efficacy (responses are rarely complete) and the attendant toxicity profiles. Such chemotherapy agents include streptozocin, doxorubicin, 5-fluorouracil, dacarbazine, capecitabine and temozolomide [
ENETS Consensus Guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system: well-differentiated pancreatic non-functioning tumors.
]. However, an improved understanding of the molecular mechanisms underlying pNET has led to more recent treatment options that include agents directed at inhibiting growth factors or their receptors that are produced by these tumours [
]. Several of these agents are still investigational and to date, only the tyrosine kinase inhibitor sunitinib and the mTOR inhibitor everolimus have been licensed by the European Medicines Agency and the FDA for the treatment of unresectable or metastatic, well-differentiated pNETs with disease progression in adults.
A number of reviews of treatments specifically for pNET have been previously published [
]. Several evidence-based guidelines on the management of pNET are available which include recommendations for the treatment of pNET (e.g. guidelines from the UK and Ireland Neuroendocrine Tumour Society (UKINETS) [
ENETS Consensus Guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system: well-differentiated pancreatic non-functioning tumors.
More recently, key recommendations from the NET Clinical Trials Planning Meeting included the separate examination of carcinoid tumours and pancreatic NETS in clinical trials and the avoidance of SSA washout periods when evaluating novel agents for the control of hormonal syndromes [
Future directions in the treatment of neuroendocrine tumors: consensus report of the National Cancer Institute Neuroendocrine Tumor clinical trials planning meeting.
As new targeted therapies emerge and become more widely used in the management of pNET, this review was undertaken to understand the current evidence base in terms of efficacy and safety of non-surgical treatments and to assess the trial methodology supporting the use of chemotherapies and new agents in this setting.
Methods
Inclusion criteria
Randomised controlled trials (RCTs), non-RCTs and prospective single-arm studies were included if they enrolled adult patients with a confirmed diagnosis of pNET (as defined by recognised clinical guidelines). Studies enrolling patients with NETs of any aetiology (including pancreas) were included as long as relevant efficacy/safety outcomes were reported for the pNET subset of patients. Only studies with at least 10 pNET patients were included in this review as the robustness of results from smaller studies was felt to be questionable (Table 1).
Table 1Inclusion and exclusion criteria.
Criterion
Included
Excluded
Population
•
Age: ⩾18 years
•
Race: any
•
Qualifying disease: pancreatic neuroendocrine tumours (pNET)
Studies enrolling patients with neuroendocrine tumours of any aetiology (including pancreas) were be included as long as relevant efficacy/safety outcomes were reported for the pNET subset of patients.
•
No restriction on previous treatment/surgery (ie treatment naïve & refractory patients)
No restriction on dose, formulation, or mode of delivery
• Surgery
Control intervention/treatments
• Any of the interventions listed abovePlacebo/usual careNo treatment
-
Included trial outcomes
Efficacy, including but not restricted to overall survival, progression free survival, objective overall response rate (PR + SD), Time to progression (TTP)/duration of response Safety, including withdrawals due to: • Any reason • Lack of efficacy • Adverse events • Health-related quality of life
Studies only reporting symptomatic relief outcomes for functioning tumours
† Studies enrolling patients with neuroendocrine tumours of any aetiology (including pancreas) were be included as long as relevant efficacy/safety outcomes were reported for the pNET subset of patients.
A systematic review of electronic databases and conference proceedings was conducted to identify relevant studies. Medline, Embase, The Cochrane Library, BIOSIS and ISI Web of Knowledge were accessed in April 2012. The search combined pNET terms ‘neuroendocrine tumor’ and ‘pancreas’ with intervention terms for somatostatin, chemotherapy and pharmacotherapy agents, interferon, radionuclide therapies and liver-directed therapies. Conference proceedings from the American Society of Clinical Oncology (ASCO), European Neuroendocrine Tumor Society (ENETS) and the European Society for Medical Oncology (ESMO) were hand-searched (2006–2012 inclusive).
Identified studies were independently assessed by two researchers in order to ascertain whether they met a set of pre-defined inclusion/exclusion criteria for inclusion in the systematic review (Table 1).
Data extraction
A pre-determined data extraction table was designed in an Excel® spreadsheet. The primary efficacy outcomes extracted included overall survival (OS), progression free survival (PFS), objective overall response rate (ORR), tumour response and time to progression (TTP)/duration of response. The safety outcomes of interest included incidence of death, incidence of withdrawal, and incidence of serious adverse events (AE). Health related quality of life (HRQoL) data (reported using generic- or disease-specific questionnaires) were also extracted.
Quality assessment
Study quality was assessed independently by two reviewers using the methods recommended in the Cochrane Reviewer’s handbook [
Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011. http://www.cochrane-handbook.org.
]. Any differences of opinion were resolved by discussion and consensus.
Data analysis
As discussed in more detail in the results section, there was considerable variation between the studies meeting eligibility criteria for inclusion in the review in terms of study design (majority of studies enrolled single-arm cohorts), reported outcomes and included populations. Therefore as a robust meta-analysis comparing the efficacy and safety of all available treatments was not feasible, results were summarised qualitatively.
Results
Study flow
In total, 5,100 studies were identified from the literature search (following removal of duplicates). A further fourteen relevant publications were identified from searching reference lists and conference proceedings. After excluding papers not meeting the inclusion criteria, 52 publications of 47 studies were included in the systematic review (Fig. 1).
Systemic chemotherapy was the most common intervention (two RCTs, 15 non-randomised studies) with many single-arm studies evaluating a wide range of agents (Table 2). The second most frequently reported treatments were SSAs and radiolabeled SSA therapies (nine non-randomised studies), followed by novel targeted agents (two RCTs and six non-randomised studies), liver-directed therapies (six non-randomised studies), mixed treatment regimens (one RCT, four non-randomised studies) and other interventions such as interferon and recombinant human endostatin (one non-randomised study each).
Table 2Study characteristics of included studies. Primary outcome shown in bold.
Pavel M, Heuck F, Plockinger U, Pape UF, Tiling N, Rinke A, et al. Prospective randomized trial: Biotherapy versus chemotherapy in malignant nonfunctional neuroendocrine tumors of the pancreas and brochial tract (ENET-1). ASCO GI conference. 2008.
Hobday T, Rubin J, Holen K, Picus J, Donehower R, Marschke RF, et al. MC044h, a phase II trial of sorafenib in patients with metastatic neuroendocrine tumors (NET): A phase II Consortium (P2C) study. ASCO GI conference. 2007.
Hobday T, Holen K, Donehower R, Camoriano J, Kim G, Picus J, et al. A phase II trial of gefitinib in patients with progressive metastatic neuroendocrine tumors (NET): A Phase II Consortium (P2C) study. ASCO GI conference. 2006.
Progression-free survival (PFS) by blinded independent central review (BICR) and updated overall survival (OS) of sunitinib versus placebo for patients with progressive, unresectable, well differentiated pancreatic neuroendocrine tumor (NET).
Kulke M, Blaszkowsky A, Zhu AX, Florio S, Regan E, Ryan DP, et al. Phase I/II study of everolimus (RAD001) in combination with temozolomide (TMZ) in patients with advanced pancreatic neuroendocrine tumors (NET). ASCO GI conference. 2010.
Good-risk. Chlorozotocin (CTZ) 175 mg/m2 IV day 1 + 5-FU 800 mg/m2/d IV days 1-4; maintenance dose of 100 mg CTZ and 600 mg of 5-FU. Poor-risk pts given lower dosage.Maximum of 18 months
Biopsy-proven islet-cell carcinoma, not amenable to surgery
Ramanathan RK, Cnaan A, Hahn RG, Carbone PP, Haller DG. Phase II trial of dacarbazine (DTIC) in advanced pancreatic islet cell carcinoma. Study of the Eastern Cooperative Oncology Group-E6282. Ann Oncol. 2001;12(8):1139–1143.
Doxorubicin 60 mg/m2 IV, repeated at 3 weeks, 6 weeks and then every 4 weeks.Until evidence of disease progression or until a total of doxorubicin dose of 500 mg/m2 had been administered
Histopathologic confirmation of locally unresectable or metastatic islet cell carcinoma
O CMJ, Rubin J. Treatment of Neuroendocrine Carcinomas with Combined Etoposide and Cisplatin Evidence of Major Therapeutic Activity in the Anaplastic Variants of These Neoplasms.
Sprenger A, Wied M, Mueller HH, Klose K, Arnold R. Effect of dacarbacine (DTIC) on tumor growth in patients with metastatic neuroendocrine gastroenteropancreatic (GEP) tumors. Gastroenterology. 2000;118(4 Suppl. 2 Part 1):AGA A647.
Strosberg J, Choi J, Gardner N, Kvols L. First-line treatment of metastatic pancreatic endocrine carcinomas with capecitabine and temozolomide. ASCO GI conference. 2008.
FOLFIRI chemotherapy: (irinotecan 180 mg/m2 infusion combined with simplifiedLV5FU2) every 14 days. Cycles repeated every 14 days using a chemotherapy free-interval scheme
Metastatic or advanced well-differentiated pNET and progressive disease.
SSA given s.c. 50ug BID day 1, 100ug BID day 2 then 150ug TID (n = 12).Dose increased to 250ug, 500ug and 500ug TID (n = 10).Median 5 months(range 1–15)
Long-term clinical outcome of somatostatin analogues for treatment of progressive, metastatic, well-differentiated entero-pancreatic endocrine carcinoma.
Chemotherapy: IV Streptozotocin0.5 g/m2 for 5 days followed by 1 gm/m2 every third week + 5-FU IV 400 mg/m2 for 3 days and then every third week. Interferon: 5 MU 3 times a week. Octreotide: 100ug bd
Analysis of progression-free survival (PFS) by prior chemotherapy use and updated safety in radiant-3: A randomized, double-blind, placebo-controlled, multicenter, phase III trial of everolimus in patients with advanced low-or intermediate-grade pancreatic neuroendocrine tumors (PNET).
Kim YH, Ajani JA, Humberto Carrasco C, Dumas P, Richli W, Lawrence D, et al. Selective hepatic arterial chemoembolization for liver metastases in patients with carcinoid tumor or islet cell carcinoma. Cancer Invest. 1999;17(7):474–8.
Capitanio V, Papa M, Zerbi A, Vitali G, De Cobelli F, Venturino M, et al. Transarterial chemoembolization (TACE) of liver metastasis from pancreatic well-differentiated endocrine carcinoma. Neuroendocrinology. 2010:4.
A summary of the included studies is reported in Table 2. The included studies were highly varied in terms of study design; duration and outcomes; sample size; doses and schedules of the interventions employed (particularly in systemic chemotherapy studies) and features of the enrolled patient populations. A significant proportion (15%) of the included studies were published over 15 years ago including two of the pivotal RCTs [
]. These early studies often used criteria (e.g. physical examination) which are no longer considered appropriate to measure response to treatment. More recent studies have reported response rates using validated radiological criteria such as WHO [
New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada.
] had an open-label study design and reported ambiguous allocation methods. Both of these criteria have been reported to be important determinants of trial bias and studies with inadequate reporting tend to overestimate the treatment effect [
Pavel M, Heuck F, Plockinger U, Pape UF, Tiling N, Rinke A, et al. Prospective randomized trial: Biotherapy versus chemotherapy in malignant nonfunctional neuroendocrine tumors of the pancreas and brochial tract (ENET-1). ASCO GI conference. 2008.
] was reported as a conference abstract only and therefore limited information on determinants of bias was reported. Two further placebo-controlled RCTs examining the targeted therapies sunitinib [
] were powered to show standard oncological end-points (PFS and OS) and employed the standardised RECIST criteria to measure response.
Many of the non-randomised studies enrolled small numbers of patients and are therefore underpowered to detect significant treatment effects. Forty of the 42 non-randomised studies enrolled 50 patients or less (median 19, [range 11–50] patients). The older studies tended to lack standardised inclusion criteria (e.g. including heterogenous tumour subtypes, patients with different exposure to prior therapy, severity of disease or evidence of disease progression at study entry).
The majority of non-randomised studies reported partial information making it difficult to appraise study quality using the Chambers checklist [
A summary of the survival and response data reported in RCT and non-randomised studies is reported in Table 3, Table 4 respectively; given the higher level of evidence they provide we will focus on the RCTs.
Table 3Survival/response data reported in RCT studies. Additional HRs are reported in footnotes.
Analysis conducted June 2010. The result is confounded by crossover; 69% of patients crossed from placebo to sunitinib. Additional HRs (95% CI) reported using four models to adjust for crossover (median follow-up 34.1months): (i) censoring at crossover, 0.43 (0.24–0.77); p=0.004; (ii) time-dependent Cox model, 0.49 (0.29–0.85); p=0.01; (iii) RPSFT model, 0.43 (0.17–1.20); p=0.12; (iv) RPSFT model adjusted for crossover time, 0.57 (0.18–1.09); p=0.12 [48].
(i) Investigator assessed PFS: revised HR based on excluding patients who had WHO performance status of 2 from the RADIANT-3 trial (to match inclusion criteria of the Raymond, 2011 study: 0.38 (0.29–0.49) [42]; (ii) Analysis of data for a subgroup of 84 patients for whom MRI/CT scans were available to compare the results with central review assessment: sunitinib, 19.8months vs. placebo, 5.8months: investigator assessed HR 0.45 HR 0.45 (0.22–0.92) [39]; (iii) Retrospective blinded independent central review of the tumour imaging scans: sunitinib 12.6months vs. placebo 5.8months; HR 0.32 (0.18–0.55) [49].
OS Analysis cut-off Feb 2011 (146 events: 68 everolimus; 78 placebo). The result is confounded by crossover, 85% of patients crossed from placebo to everolimus. Median OS not reached for everolimus treatment arm, so data immature [80]. No crossover adjusted OS estimates have currently been published.
Response/regression definition used : Favourable objective response if tumour mass reduced by>50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
NR
NR
NR
30
Response/regression definition used : Favourable objective response if tumour mass reduced by>50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
NR
21
Streptozocin + Fluorouracil
1.4
p<0.004 vs. Streptozocin+Doxorubicin.
4
Response/regression definition used : Favourable objective response if tumour mass reduced by>50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
45
Response/regression definition used : Favourable objective response if tumour mass reduced by>50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
13
Streptozocin + Doxorubicin
2.2
14
Response/regression definition used : Favourable objective response if tumour mass reduced by>50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
69
Response/regression definition used : Favourable objective response if tumour mass reduced by>50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
Pavel M, Heuck F, Plockinger U, Pape UF, Tiling N, Rinke A, et al. Prospective randomized trial: Biotherapy versus chemotherapy in malignant nonfunctional neuroendocrine tumors of the pancreas and brochial tract (ENET-1). ASCO GI conference. 2008.
‡‡ Response/regression definition used : Favourable objective response if tumour mass reduced by > 50% or 30% if malignant hepatomegaly was used, lab assay reduced by 50% tumour regression.
§§ Median duration of response: ‘all responses’, 17 months; ‘complete responses’, 24 months.
¶¶ OS Analysis cut-off Feb 2011 (146 events: 68 everolimus; 78 placebo). The result is confounded by crossover, 85% of patients crossed from placebo to everolimus. Median OS not reached for everolimus treatment arm, so data immature [80]. No crossover adjusted OS estimates have currently been published.
††† (i) Investigator assessed PFS: revised HR based on excluding patients who had WHO performance status of 2 from the RADIANT-3 trial (to match inclusion criteria of the Raymond, 2011 study: 0.38 (0.29–0.49)
Signorovitch J, Swallow E, Kantor E, Wang X, Metrakos P. Overall survival with everolimus, sunitinib, and placebo for advnaced pancreatic neuroendocrine tumors: A matching-adjusted indirect comparison of randomized trials. J Clin Oncol. 2012;30(Suppl 4):Abstract 237.
; (ii) Analysis of data for a subgroup of 84 patients for whom MRI/CT scans were available to compare the results with central review assessment: sunitinib, 19.8 months vs. placebo, 5.8 months: investigator assessed HR 0.45 HR 0.45 (0.22–0.92)
Evaluation of progression-free survival by blinded independent central review in patients with progressive, well-differentiated pancreatic neuroendocrine tumors treated with sunitinib or placebo.
Progression-free survival (PFS) by blinded independent central review (BICR) and updated overall survival (OS) of sunitinib versus placebo for patients with progressive, unresectable, well differentiated pancreatic neuroendocrine tumor (NET).
‡‡‡ Analysis conducted June 2010. The result is confounded by crossover; 69% of patients crossed from placebo to sunitinib. Additional HRs (95% CI) reported using four models to adjust for crossover (median follow-up 34.1 months): (i) censoring at crossover, 0.43 (0.24–0.77); p = 0.004; (ii) time-dependent Cox model, 0.49 (0.29–0.85); p = 0.01; (iii) RPSFT model, 0.43 (0.17–1.20); p = 0.12; (iv) RPSFT model adjusted for crossover time, 0.57 (0.18–1.09); p = 0.12
Faivre S, Niccoli P, Raoul JL, Bang Y-J, Borbath I, Valle JW, et al., editors. Updated Overall Survival (OS) Analysis From A Phase Iii Study Of Sunitinib Vs Placebo In Patients (Pts) With Advanced, Unresectable Pancreatic Neuroendocrine Tumor (Net). ESMO; 2012 28th September-2nd October; Vienna, Austria.
]. The first study established that doublet chemotherapy (streptozocin and 5-FU) resulted in a superior response rate (63% vs. 36%) and OS (26 vs. 16.5 months) compared with streptozocin monotherapy in 82 randomised patients [
]. In the follow-on study comparing chlorozotocin vs. streptozocin/doxorubicin vs. streptozocin/5-FU, patients receiving combination chemotherapy once again fared better, with a response rate of 69% and OS of 2.2 years reported for the streptozocin/doxorubicin combination which appeared to be the most active regimen, although associated with significant toxicity [
The two most recent RCTs (assessing the targeted agents sunitinib and everolimus), are the first and only published studies to documented an improvement in PFS compared with best supportive care (BSC)/placebo. Both studies had a similar design: selection of a homogeneous patient subgroup (pathologically-confirmed well-differentiated pNET with evidence of disease progression), use of a double-blind placebo-controlled design using a robust oncological end-point (PFS, by RECIST) and ability of patients to cross over from placebo to active drug on disease progression. Sunitinib demonstrated an improved PFS of 11.4 vs. 5.5 months (vs. placebo, HR 0.42 (0.26–0.66); p < 0.0001 vs. placebo) [
], this endpoint could not be reliably assessed due to extensive crossover from placebo to active treatment (69% of patients in the sunitinib trial and 74% of patients in the everolimus trial) [
Faivre S, Niccoli P, Raoul JL, Bang Y-J, Borbath I, Valle JW, et al., editors. Updated Overall Survival (OS) Analysis From A Phase Iii Study Of Sunitinib Vs Placebo In Patients (Pts) With Advanced, Unresectable Pancreatic Neuroendocrine Tumor (Net). ESMO; 2012 28th September-2nd October; Vienna, Austria.
The final RCT (comparing streptozocin / 5-FU vs. octreotide) will not be discussed further given its small size (n = 16 patients) and the limited abstract-only available information.
It was not possible to draw conclusions on the comparable efficacy of different interventions due to the significant heterogeneity between studies. There was significant variation in the duration of follow up between studies, which limits the assessment of survival outcomes. The comparability of efficacy between studies was also hampered by differences in eligibility criteria, baseline characteristics, and response criteria employed in these studies. For example, differences in PFS between studies could be attributed to studies not requiring disease progression or prior chemotherapy prior to enrolment. The importance of response criteria is illustrated by two RCTs which established streptozocin combined with either fluorouracil or doxorubicin as the treatment of choice in pNET based on “response rates” of 45–69%; however, these “responses” included reduction in clinical hepatomegaly, biochemical improvement and/or radiological regression [
]. Over the last 30 years, significant advances in imaging techniques have resulted in more accurate determination of tumour bulk and validated criteria to describe tumour response such as the WHO criteria or, more recently, RECIST. Several subsequent studies, using these stricter radiological definitions of response rather than clinical ones, have, however failed to confirm the high response rates seen in the early RCTs [
Despite the increasing importance of patient reported outcomes (PRO) in assessing the effectiveness of treatments, HRQoL has been rarely assessed in this patient population. Indeed, only two studies, both investigating the targeted therapy sunitinib, assessed QoL using validated PRO measures (EORTC Quality of Life Questionnaire [
], in the absence of a NET-specific quality of life tool). In both studies, despite its inherent toxicities, active treatment did not result in significant detriment from baseline in overall QoL measures compared with best supportive care.
Safety
A summary of selected AEs are reported in Table 5 (RCTs) and Table 6 (non-randomised studies). In general, the reported incidence of grade 3–4 AEs was similar to the previously characterised profile for the different classes of agents. Chemotherapy agents have high levels of reversible myelosuppression, emesis (especially prior to the use of modern anti-emetics) and agent-specific toxicities (e.g. doxorubicin-induced cardiotoxicity; fluoropyrimidine-induced diarrhea). The targeted therapies (sunitinib and everolimus) have fewer grade 3–4 toxicities compared with chemotherapeutic agents in keeping with experience of these drugs in other tumour types.
Table 5Summary of safety data reported in RCT studies.
Extended follow up (20.1months) [everolimus vs. placebo]: Diarrhoea, all grade: 34.3 vs. 10.3%; stomatitis, all grade: 52.9 vs. 12.3%; stomatitis, grade 3/4, 4.9% vs. 0%.
Pavel M, Heuck F, Plockinger U, Pape UF, Tiling N, Rinke A, et al. Prospective randomized trial: Biotherapy versus chemotherapy in malignant nonfunctional neuroendocrine tumors of the pancreas and brochial tract (ENET-1). ASCO GI conference. 2008.
Main AEs nausea, emesis, mucositis, electrolyte disturbance and thromboembolism
Octreotide
Main AEs: abdominal pain and meteroism
† Thrombocytopenia: any,<100 × 109 cells/litre.
‡ Thrombocytopenia: severe:<50 × 109 cells/litre.
§ Leukopenia: any,<4 × 109 cells/litre.
¶ Leukopenia: severe:<2 × 109 cells/litre.
†† Treatment group not reported.
‡‡ <4000 to ⩾2000/mm3.
§§ <2000 to ⩾1000/mm3.
¶¶ <1000/mm3.
††† <150,000 to ⩾100,000/mm3.
‡‡‡ <100,000 to ⩾50,000/mm3.
§§§ <50,000/mm3.
¶¶¶ Extended follow up (20.1 months) [everolimus vs. placebo]: Diarrhoea, all grade: 34.3 vs. 10.3%; stomatitis, all grade: 52.9 vs. 12.3%; stomatitis, grade 3/4, 4.9% vs. 0%.
This systematic review was undertaken to assess non-surgical treatments for pNET. Robust methods were used to identify all relevant studies and assess the methodological quality of the included studies.
RCTs represent the gold standard study design for assessment of treatment efficacy. However, there are few RCTs performed to date to assess the efficacy and tolerability of treatment options in the management of pNET. This is partly due to the complexity and clinical heterogeneity of pNETs and a perception that these tumours were “too rare” to undertake adequately-powered RCTs. As a result, most of the published evidence base comes from single-arm non-randomised studies. While recently conducted non-randomised studies have employed validated criteria to measure patient response [
Hobday T, Holen K, Donehower R, Camoriano J, Kim G, Picus J, et al. A phase II trial of gefitinib in patients with progressive metastatic neuroendocrine tumors (NET): A Phase II Consortium (P2C) study. ASCO GI conference. 2006.
Hobday T, Rubin J, Holen K, Picus J, Donehower R, Marschke RF, et al. MC044h, a phase II trial of sorafenib in patients with metastatic neuroendocrine tumors (NET): A phase II Consortium (P2C) study. ASCO GI conference. 2007.
], many older ‘historical’ studies lacked standardised inclusion criteria (e.g. including heterogenous tumour subtypes, and patients with varied severity of disease, extent of prior therapy, evidence of disease progression at study entry and other prognostic factors) and standardised outcomes [
Future directions in the treatment of neuroendocrine tumors: consensus report of the National Cancer Institute Neuroendocrine Tumor clinical trials planning meeting.
]. Fig. 2 indicates that over the last three decades, while studies examining chemotherapy regimens have continued to enrol a consistently small number of patients (<50 patients), the last five years has seen the publication of adequately powered studies (both observational and RCTs) examining targeted therapies enrolling a larger number of patients. Such studies are both sufficiently powered and report clinically meaningful endpoints such as PFS and OS and have indicated the feasibility of conducting robust RCTs.
Fig. 2Relationship between date of publication and number of patients enrolled in clinical studies.
There are currently no head-to-head RCTs comparing the efficacy of licensed treatments in this indication in the first-line setting or studies evaluating the optimal sequencing of available therapies.
Results from the present review indicate that over the last three decades, the criteria used to address response in pNET have progressed from clinical / biological responses to the use of validated WHO and RECIST criteria. However, these criteria have been validated for cytotoxic chemotherapy. As the use of targeted agents increases, patients may benefit from novel criteria (e.g. Choi) which assess both tumour density/necrosis in addition to change in tumour size. Targeted therapies can induce extensive regions of necrosis within a tumour. Although the induction of tumour necrosis may be associated with clinical benefit, RECIST criteria are restricted to measuring tumour size and so unlike the Choi criteria, they are insensitive to changes in tumour density and may therefore underestimate the anti-tumour effect of targeted agents in pNET, which are generally insidious in nature. While the utility of the Choi criteria have been established in the management of gastrointestinal stromal tumours (GIST) [
Grande E, Jose Diez J, Pachon V, Angeles Vaz M, Longo F, Guillen C, et al. Response by Choi criteria to sunitinib plus octreotide LAR in a functional heavily pretreated advanced pancreatic neuroendocrine tumor. Anticancer Drugs. 2011;22(5):477–479.
ClinicalTrials.gov. NCT01525550. A study of the efficacy and safety of sunitinib in patients with advanced well-differentiated pancreatic neuroendocrine tumors.
Hentic O, Dreyer C, Zappa M, Hammel P, Mateescu C, Bouattour M, et al. Response evaluation usinf RECIST and CHOI criteria in patients with well-differentiated pancreatic neuroendocrine tumors (pNET) treated with sunitinib or everolimus. J Clin Oncol. 2012;30:Suppl; abstr e14660.
Despite the limitations in the robustness of its data, systemic cytotoxic chemotherapy has a role in the management of pNET; indeed it is recommended as a first-line treatment option by the European Neuroendocrine Tumour Society [
ENETS Consensus Guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system: well-differentiated pancreatic non-functioning tumors.
]. However, there is scope for improved, robust evidence of its efficacy from future RCTs.
Several new targeted agents have been developed and investigated in recent studies although again the majority of these have been assessed in non-randomised studies (Table 2). This review identified two RCTs assessing targeted therapies: one examining sunitinib [
]. Based on these results, these agents are valid and applicable treatment options for patients with advanced pNET in daily practice. However, there remain as yet a number of unexplored issues such as the optimal positioning of these drugs with respect to each other and, indeed, chemotherapy; the feasibility, safety and efficacy of combining these treatments with other modalities (e.g. liver-directed therapy, or radionuclide therapy); as well as their role in other stages of the disease pathway (e.g. neoadjuvant or as an adjuvant therapy to surgery) [
Furthermore, the concept of personalised patient-specific treatment depending on tumour histology, comorbid conditions and objectives of treatment as part of an optimised treatment algorithm is becoming increasingly important [
]. For example, patients with a high tumour burden or mitotically active tumour (as measured by Ki67 index) may benefit from an antiproliferative treatment such as chemotherapy. However, for those patients with low-grade tumours or lower tumour burden, consideration of the patient’s quality of life is of utmost importance. In such cases, results from the phase III sunitinib RCT indicate that clinical benefits were gained without adversely affecting quality of life [
] assessing the targeted therapies have indicated the feasibility of conducting robust trials which report clinically meaningful endpoints such as PFS and OS. Following publication of the sunitinib trial, the United States Food and Drug Administration (FDA) has issued a recommendation that interim efficacy analyses should be carefully planned in order to reduce the risk of overestimating treatment effects [
]. Further research will be required to confirm the efficacy of these novel therapies and to define the ideal treatment algorithm for the management of pNET. These trials will require international collaboration and should ideally be designed following multidisciplinary clinical input and include patients classified according to histological guidelines specifically developed for pNETs to ensure homogenous enrolment of patients. In addition, future trials should assess alternative treatment strategies incorporating effective agents, used concurrently or sequentially dependent on tumour characteristics, which may replace the concept of a single therapy per patient.
Conflict of interest statement
Stephen Mitchell is an employee of Abacus International and was a paid consultant to Pfizer in connection with the development of this manuscript. Juan Valle and Martin Eatock received an honorarium in conjunction with development of the manuscript. Zahava Gabriel, Roxanne Ferdinand and Ben Clueit are paid employees of Pfizer.
References
Ramage J.K.
Davies A.H.G.
Ardill J.
Bax N.
Caplin M.
Grossman A.
et al.
Guidelines for the management of gastroenteropancreatic neuroendocrine (including carcinoid) tumours.
Fluorouracil, doxorubicin, and streptozocin in the treatment of patients with locally advanced and metastatic pancreatic endocrine carcinomas. [Erratum appears in J Clin Oncol 2005 Jan 1;23(1):248].
ENETS Consensus Guidelines for the management of patients with digestive neuroendocrine neoplasms of the digestive system: well-differentiated pancreatic non-functioning tumors.
Future directions in the treatment of neuroendocrine tumors: consensus report of the National Cancer Institute Neuroendocrine Tumor clinical trials planning meeting.
Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011. http://www.cochrane-handbook.org.
New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada.
Pavel M, Heuck F, Plockinger U, Pape UF, Tiling N, Rinke A, et al. Prospective randomized trial: Biotherapy versus chemotherapy in malignant nonfunctional neuroendocrine tumors of the pancreas and brochial tract (ENET-1). ASCO GI conference. 2008.
Hobday T, Holen K, Donehower R, Camoriano J, Kim G, Picus J, et al. A phase II trial of gefitinib in patients with progressive metastatic neuroendocrine tumors (NET): A Phase II Consortium (P2C) study. ASCO GI conference. 2006.
Hobday T, Rubin J, Holen K, Picus J, Donehower R, Marschke RF, et al. MC044h, a phase II trial of sorafenib in patients with metastatic neuroendocrine tumors (NET): A phase II Consortium (P2C) study. ASCO GI conference. 2007.
Evaluation of progression-free survival by blinded independent central review in patients with progressive, well-differentiated pancreatic neuroendocrine tumors treated with sunitinib or placebo.
Signorovitch J, Swallow E, Kantor E, Wang X, Metrakos P. Overall survival with everolimus, sunitinib, and placebo for advnaced pancreatic neuroendocrine tumors: A matching-adjusted indirect comparison of randomized trials. J Clin Oncol. 2012;30(Suppl 4):Abstract 237.
Grande E, Jose Diez J, Pachon V, Angeles Vaz M, Longo F, Guillen C, et al. Response by Choi criteria to sunitinib plus octreotide LAR in a functional heavily pretreated advanced pancreatic neuroendocrine tumor. Anticancer Drugs. 2011;22(5):477–479.
ClinicalTrials.gov. NCT01525550. A study of the efficacy and safety of sunitinib in patients with advanced well-differentiated pancreatic neuroendocrine tumors.
Hentic O, Dreyer C, Zappa M, Hammel P, Mateescu C, Bouattour M, et al. Response evaluation usinf RECIST and CHOI criteria in patients with well-differentiated pancreatic neuroendocrine tumors (pNET) treated with sunitinib or everolimus. J Clin Oncol. 2012;30:Suppl; abstr e14660.
Faivre S, Niccoli P, Raoul JL, Bang Y-J, Borbath I, Valle JW, et al., editors. Updated Overall Survival (OS) Analysis From A Phase Iii Study Of Sunitinib Vs Placebo In Patients (Pts) With Advanced, Unresectable Pancreatic Neuroendocrine Tumor (Net). ESMO; 2012 28th September-2nd October; Vienna, Austria.
Progression-free survival (PFS) by blinded independent central review (BICR) and updated overall survival (OS) of sunitinib versus placebo for patients with progressive, unresectable, well differentiated pancreatic neuroendocrine tumor (NET).
Kulke M, Blaszkowsky A, Zhu AX, Florio S, Regan E, Ryan DP, et al. Phase I/II study of everolimus (RAD001) in combination with temozolomide (TMZ) in patients with advanced pancreatic neuroendocrine tumors (NET). ASCO GI conference. 2010.
Ramanathan RK, Cnaan A, Hahn RG, Carbone PP, Haller DG. Phase II trial of dacarbazine (DTIC) in advanced pancreatic islet cell carcinoma. Study of the Eastern Cooperative Oncology Group-E6282. Ann Oncol. 2001;12(8):1139–1143.
O CMJ, Rubin J. Treatment of Neuroendocrine Carcinomas with Combined Etoposide and Cisplatin Evidence of Major Therapeutic Activity in the Anaplastic Variants of These Neoplasms.
Sprenger A, Wied M, Mueller HH, Klose K, Arnold R. Effect of dacarbacine (DTIC) on tumor growth in patients with metastatic neuroendocrine gastroenteropancreatic (GEP) tumors. Gastroenterology. 2000;118(4 Suppl. 2 Part 1):AGA A647.
Strosberg J, Choi J, Gardner N, Kvols L. First-line treatment of metastatic pancreatic endocrine carcinomas with capecitabine and temozolomide. ASCO GI conference. 2008.
Long-term clinical outcome of somatostatin analogues for treatment of progressive, metastatic, well-differentiated entero-pancreatic endocrine carcinoma.
Analysis of progression-free survival (PFS) by prior chemotherapy use and updated safety in radiant-3: A randomized, double-blind, placebo-controlled, multicenter, phase III trial of everolimus in patients with advanced low-or intermediate-grade pancreatic neuroendocrine tumors (PNET).
Kim YH, Ajani JA, Humberto Carrasco C, Dumas P, Richli W, Lawrence D, et al. Selective hepatic arterial chemoembolization for liver metastases in patients with carcinoid tumor or islet cell carcinoma. Cancer Invest. 1999;17(7):474–8.
Capitanio V, Papa M, Zerbi A, Vitali G, De Cobelli F, Venturino M, et al. Transarterial chemoembolization (TACE) of liver metastasis from pancreatic well-differentiated endocrine carcinoma. Neuroendocrinology. 2010:4.
It was discovered that the original online version of the above article contained errors that were not the fault of the author(s). The Publisher decided to ‘resupply’ (repost and replace) the XML and online PDF of the article. Unfortunately the discovery of the errors was too late to correct the printed issue.
00188-6&id=gr1.jpg){kind=link}
00188-6&id=gr2.jpg){kind=link}