Risk of oral and gastrointestinal mucosal injury among patients receiving selected targeted agents: a meta-analysis.

Purpose: The purpose of this study was to estimate the risk and severity of oral and gastrointestinal mucosal toxicities associated with selected targeted agents.
Methods: We searched the English-language literature in February 2011 for reports of randomized clinical trials comparing a FDA-approved targeted agent to a standard of care regimens. Long-term follow-up and secondary reports of trials were excluded, leaving 85 studies for analysis. Using meta-analytic methods, we calculated the relative risks of oral and gastrointestinal toxicities, adjusting for sample size using the inverse variance technique. For each targeted agent and each side effect, we calculated the number needed to harm, the number of patients that, if treated with the more toxic regimen, would produce one additional episode of the toxicity.
Results: Oral mucositis was significantly more frequent among patients treated with bevacizumab, erlotinib, sorafenib, or sunitinib, although this difference was confined to low-grade mucositis. The clinical significance of these findings is unclear given its low incidence and mild severity. In contrast, diarrhea was significantly more frequent with most of the targeted agents studied, with adjusted relative risks between 1.5 and 4.5. An additional patient with diarrhea will be observed for every three to five patients treated with these targeted agents, compared with conventional regimens.
Conclusions: Oral mucosal toxicities occasionally complicate treatment with these targeted agents, but the clinical significance of this finding is not clear. Diarrhea is a hallmark of treatment with these targeted agents; this side effect should be carefully ascertained to permit early intervention and control.