Therapeutic Options for Neuroendocrine Tumors: A Systematic Review and Network Meta-analysis

Abstract

Importance: Multiple therapies are currently available for patients with neuroendocrine tumors (NETs), yet many therapies have not been compared head-to-head within randomized clinical trials (RCTs).

Objective: To assess the relative safety and efficacy of therapies for NETs.

Data sources: PubMed, Embase, the Cochrane Central Register of Controlled Trials, trial registries, meeting abstracts, and reference lists from January 1, 1947, to March 2, 2018, were searched. Key search terms included neuroendocrine tumors, gastrointestinal neoplasms, therapy, and randomized controlled trial.

Study selection: Randomized clinical trials comparing 2 or more therapies in patients with NETs (primarily gastrointestinal and pancreatic) were evaluated. Thirty RCTs met the selection criteria.

Data extraction and synthesis: Pairs of independent reviewers screened studies, extracted data, and assessed the risk of bias. A network meta-analysis with a frequentist approach was used to compare the efficacy of therapies; the Preferred Reporting Items for Systematic Reviews and Meta-analyses guideline was used.

Main outcomes and measures: Disease control, progression-free survival, overall survival, adverse events, and quality of life.

Results: The systematic review identified 30 relevant RCTs comprising 3895 patients (48.4% women) assigned to 22 different therapies for NETs. These therapies showed a broad range of risk for serious and nonserious adverse events. The network meta-analyses included 16 RCTs with predominantly a low risk of bias; nevertheless, precision-of-treatment estimates and estimated heterogeneity were limited. The network meta-analysis found 7 therapies for pancreatic NETs: everolimus (hazard ratio [HR], 0.35 [95% CI, 0.28-0.45]), everolimus plus somatostatin analogue (HR, 0.35 [95% CI, 0.25-0.51]), everolimus plus bevacizumab plus somatostatin analogue (HR, 0.44 [95% CI, 0.26-0.75]), interferon (HR, 0.37 [95% CI, 0.16-0.83]), interferon plus somatostatin analogue (HR, 0.31 [95% CI, 0.13-0.71]), somatostatin analogue (HR, 0.46 [95% CI, 0.33-0.66]), and sunitinib (HR, 0.42 [95% CI, 0.26-0.67]), and 5 therapies for gastrointestinal NETs: bevacizumab plus somatostatin analogue (HR, 0.22 [95% CI, 0.05-0.99]), everolimus plus somatostatin analogue (HR, 0.31 [95% CI, 0.11-0.90]), interferon plus somatostatin analogue (HR, 0.27 [95% CI, 0.07-0.96]), Lu 177-dotatate plus somatostatin analogue (HR, 0.08 [95% CI, 0.03-0.26], and somatostatin analogues (HR, 0.40 [95% CI, 0.21-0.78]) with higher efficacy than placebo and suggests an overall superiority of combination therapies.

Conclusions and relevance: The findings from this study suggest that a range of efficient therapies with different safety profiles is available for patients with NETs.

Figure 1.. Treatment Efficacy in Pancreatic Neuroendocrine Tumors (pNETs)

Network plots for disease control (A) and progression-free survival (PFS) (B) in pNETs. The thickness of the edges is proportional to the inverse SEs of the pairwise comparisons, and the numbers indicate the number of studies. One 3-arm study is marked by shading. C and D, Forest plots for disease control and PFS, respectively, in pNETs. The forest plots include only comparisons vs placebo. An odds ratio (OR) larger than 1 indicates increased disease control of the active treatment and a hazard ratio (HR) smaller than 1 indicates a reduced risk for progression for the active treatment. An HR smaller than 1 indicates a reduced risk for progression. All therapies are listed in order of their P scores, with the most effective therapy at the top. Heterogeneity was assessed by the between-study variance τ² value, Cochran Q with a P value, and I². Total No. refers to the total number of patients, and No. to the number of patients with disease control. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was used to rate the quality of evidence of estimates from pairwise and network meta-analysis, where 1 is very low; 2, low; 3, moderate; and 4, high. SSA indicates somatostatin analogue.

Figure 2.. Treatment Efficacy in Gastrointestinal Neuroendocrine Tumors (GI-NETs)

Network plots for disease control (A) and progression-free survival (PFS) (B) in GI-NETs. The thickness of the edges is proportional to the inverse SEs of the pairwise comparisons, and the numbers indicate the number of studies. One 3-arm study is marked by shading. C and D, Forest plots for disease control and PFS, respectively, in GI-NET. The forest plots include only comparisons vs placebo. An odds ratio (OR) larger than 1 indicates increased disease control of the active treatment and a hazard ratio (HR) smaller than 1 indicates a reduced risk for progression for the active treatment. An HR smaller than 1 indicates a reduced risk for progression. All therapies are listed in order of their P scores, with the most effective therapy at the top. Heterogeneity was assessed by the between-study variance τ² value, Cochran Q with a P value, and I². Total No. refers to the total number of patients, and No. to the number of patients with disease control. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was used to rate the quality of evidence of estimates from pairwise and network meta-analysis, where 1 is very low; 2, low; 3, moderate; and 4, high. ¹⁷⁷Lu-dotatate indicates radiolabeled lutetium Lu 177-dotatate; SSA, somatostatin analogue.