Risk of Treatment-Related Mortality with Sorafenib in Cancer Patients: a Meta-Analysis of 20 Randomly Controlled Trials

Xiongwen Yang,Xiong Pan,Xiaoshu Cheng,Yingzhang Cheng,Yukang Kuang
DOI: https://doi.org/10.1007/s11096-015-0151-y
IF: 2.305
2015-01-01
International Journal of Clinical Pharmacy
Abstract:Background Sorafenib is a relatively new multi-kinase inhibitor used to treat a wide range of cancers. As fatal adverse events from sorafenib therapy are rare, their investigation requires a meta-analysis. Aim of the review To provide a meta-analysis of sorafenib-associated fatal adverse events with the most expansive and current data. Method We searched Medline, EMBASE, Web of Science and Cochrane Library databases. We also searched abstracts from a number of conferences. Twenty trials of sorafenib were found in 9434 cancer patients, tested against placebos and against other drugs. We calculated relative risks and incidences for sorafenib-associated mortality. Results Overall incidence of sorafenib-associated mortality was 3.3 %. Patients with renal cell carcinoma (RCC) and thyroid cancer had treatment-related mortality ≥5 %. Patients treated with sorafenib had a significantly greater risk of mortality than those in placebo/control groups, with an RR of 1.75. Subgroup analyses also showed significant differences in sorafenib versus placebo (RR 1.87, 95 % CI 1.23–2.86; I 2 = 0.0 %, P = 0.865); and sorafenib + platinum-based chemotherapy (RR 2.03, 95 % CI 1.15–3.59; I 2 = 0.0 %, P = 0.654). However, sorafenib had lower risk than other multi-targeted antiangiogenic tyrosine kinase inhibitors. Patients with RCC and non-small-cell lung carcinoma were significantly more vulnerable. Conclusion Sorafenib presents a significant risk of fatal adverse events (FAEs) in patients with cancer, especially for RCC or non-small-cell lung carcinoma, and in patients treated with sorafenib + platinum-based chemotherapy. However, compared with other multi-targeted antiangiogenic tyrosine kinase inhibitors, sorafenib has a lower risk of FAEs.
What problem does this paper attempt to address?