Analysis of Bevacizumab-Induced Gastrointestinal Perforation Using the Japanese Adverse Drug Event Report Database

Bevacizumab is a recombinant humanized monoclonal antibody against human vascular endothelial growth factor (VEGF), which inhibits angiogenesis in tumor tissues by blocking VEGF activity. Although rare, bevacizumab-induced gastrointestinal perforation (BIGP) can reduce patients' quality of life and even lead to death. We aimed to evaluate the time to BIGP onset according to the indication, its outcome, and the effect of bevacizumab in combination with various anticancer agents.

Adverse events in the Japanese Adverse Drug Reaction Reports (JADER) database were defined according to the Medical Dictionary for Regulatory Activities, and "gastrointestinal perforation (SMQ 'standardized MedDRA inquiry' 20000107)" was extracted. Reasons for use were categorized by seven indications, and other diseases were classified as "other" and were evaluated for time-to-onset analysis and outcomes. Association rule mining was used to assess the risk of BIGP associated with the administration of bevacizumab combined with various anticancer agents.

The JADER database includes 887,704 reports submitted between April 2004 and March 2024, including 2,112 reports of BIGP. The times to BIGP onset (quartile range) for non-small cell lung, colorectal, and ovarian cancers were 46.0 (11.0-122.0), 77.0 (29.0-196.0), and 67.0 (23.0-203.0) days, respectively. The log-rank test demonstrated that BIGP occurred earlier in patients with non-small cell lung cancer than in patients with colorectal (P < 0.0001) or ovarian (P = 0.0033) cancer. Association rule mining results showed that for the consequent (right-hand side) "1-100 days onset," drugs used to treat non-small cell lung cancer were at the top. However, for "101-200 days onset," irinotecan and drugs for colorectal cancer were at the top of the association rule. BIGP outcomes in the JADER report were 64.7% "improvement" and 35.3% "no improvement."

In non-small cell lung cancer, the time to BIGP onset is earlier than that in colorectal and ovarian cancers. This finding suggests that healthcare providers can detect and intervene BIGP at early stages.