Comparative safety and effectiveness of serotonin receptor antagonists in patients undergoing chemotherapy: a systematic review and network meta-analysis

Andrea C. Tricco, Erik Blondal, Areti Angeliki Veroniki, Charlene Soobiah, Afshin Vafaei, John Ivory, Lisa Strifler, Roberta Cardoso, Emily Reynen, Vera Nincic, Huda Ashoor, Joanne Ho, Carmen Ng, Christy Johnson, Erin Lillie, Jesmin Antony, Derek J. Roberts, Brenda R. Hemmelgarn, Sharon E. Straus

Although serotonin (5-HT3) receptor antagonists are effective in reducing nausea and vomiting, they may be associated with increased cardiac risk. Our objective was to examine the comparative safety and effectiveness of 5-HT3 receptor antagonists (e.g., dolasetron, granisetron, ondansetron, palonosetron, tropisetron) alone or combined with steroids for patients undergoing chemotherapy. We searched MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials from inception until December 2015 for studies comparing 5-HT3 receptor antagonists with each other or placebo in chemotherapy patients. The search results were screened, data were abstracted, and risk of bias was appraised by pairs of reviewers, independently. Random-effects meta-analyses and network meta-analyses (NMAs) were conducted. After screening 9226 citations and 970 full-text articles, we included 299 studies (n = 58,412 patients). None of the included studies reported harms for active treatment versus placebo. For NMAs on the risk of arrhythmia (primary outcome; three randomized controlled trials [RCTs], 627 adults) and mortality (secondary outcome; eight RCTs, 4823 adults), no statistically significant differences were observed between agents. A NMA on the risk of QTc prolongation showed a significantly greater risk for dolasetron + dexamethasone versus ondansetron + dexamethasone (four RCTs, 3358 children and adults, odds ratio 2.94, 95% confidence interval 2.13-4.17). For NMAs on the number of patients without nausea (44 RCTs, 11,664 adults, 12 treatments), number of patients without vomiting (63 RCTs, 15,460 adults, 12 treatments), and number of patients without chemotherapy-induced nausea or vomiting (27 RCTs, 10,924 adults, nine treatments), all agents were significantly superior to placebo. For a NMA on severe vomiting (10 RCTs, 917 adults), all treatments decreased the risk, but only ondansetron and ramosetron were significantly superior to placebo. According to a rank-heat plot with the surface under the cumulative ranking curve results, palonosetron + steroid was ranked the safest and most effective agent overall. Most 5-HT3 receptor antagonists were relatively safe when compared with each other, yet none of the studies compared active treatment with placebo for harms. However, dolasetron + dexamethasone may prolong the QTc compared to ondansetron + dexamethasone. All agents were effective for reducing risk of nausea, vomiting, and chemotherapy-induced nausea or vomiting. This study was registered at PROSPERO: ( CRD42013003564 ).