In the era of somatostatin analogues and targeted therapies, the role of chemotherapy in NET remains largely undefined. This systematic review aimed to assess the effect of chemotherapy on response rates (RR), progression-free survival (PFS), overall survival (OS) and toxicity compared to other chemotherapies/systemic therapies or best supportive care in patients with advanced or metastatic NET.
Randomised controlled trials (RCTs) from 1946 to 2015 were identified from MEDLINE, EMBASE, other databases and conference proceedings. Review of abstracts, quality assessment and data abstraction were performed independently by two investigators. Meta-analyses were conducted using Mantel-Haenszel analysis with random-effects modelling.
Six RCTs comparing standard streptozotocin plus 5-fluorouacil (STZ/5FU) chemotherapy to other chemotherapy regimens, and 2 comparing this to interferon (IFN) were included. Only 1 study was considered at low risk of bias. STZ/5-FU was no different to other chemotherapies in response rate [RR 0.96; 95% confidence interval (CI) 0.72-1.27], PFS (RR 0.95; CI 0.81-1.13), or OS (RR 1.03; CI 0.77-1.39). IFN may produce higher response than STZ/5FU (RR 0.20; CI 0.04-1.13), but event rates were small and survival was no different. Interferon was associated with higher overall haematological (RR 0.47; CI 0.27-0.82) and lower overall renal toxicity (RR 3.61; CI 1.24-10.51).
Strong evidence is lacking in the area of chemotherapy in neuroendocrine tumors. There is currently no evidence that one chemotherapeutic regimen is significantly better than the other, nor is interferon better than chemotherapy. There is an urgent need to design RCTs comparing modern chemotherapy to other agents in NET.