Risk of thromboembolism with thrombopoietin receptor agonists in adult patients with thrombocytopenia: systematic review and meta-analysis of randomized controlled trials

Abstract

Background and objective: Thrombopoietin receptor (TPOr) agonists (romiplostim and eltrombopag) are a new approach for the treatment of thrombocytopenia-associated conditions. They promote megakaryocyte differentiation, proliferation and platelet production. In the European Union, both are orphan drugs with an indication restricted to splenectomized immune thrombocytopenic purpura patients who are refractory to other treatments. Due to increasing platelet counts, these drugs may represent a risk for thromboembolic complications. We analyzed whether TPOr agonists affect thromboembolisms occurrence in adult thrombocytopenic patients.

Materials and methods: We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs). Searches were carried out in PubMed, SCOPUS, Cochrane Central Register, regulatory agencies websites and publicly available registries of manufacturers. RCTs using romiplostim or eltrombopag in at least one group were included. Absolute risk ratios (ARR), number needed to harm (NNH) and relative risks (RR) were provided. Heterogeneity was analyzed using Cochran's Q test and I(2) statistic.

Results: Of 373 publications identified, 8 studies met the inclusion criteria (n=1180 patients). The quality of reporting amongst studies was variable. Estimated frequency of thromboembolisms was 3.1% (95% CI, 1.8-4.4%) for TPOr agonists and 1.7% (95% CI, 0.3-3.1%) for controls. Summary analyses produced overall meta-ARR for thromboembolisms of 1.8% (95% CI, -0.1-3.6%), and meta-RR of 1.5 (95% CI, 0.7-3.3), meaning a NNH of 55 (1 additional thromboembolism for each 55 patients treated with TPOr agonists). All pooled estimates were homogeneous.

Conclusions: TPOr agonists show a numerically but non-statistically significant trend to increase the occurrence of thromboembolisms compared to controls, but analyses were underpowered and in some studies information on outcomes was incomplete and of poor quality.