Granulocyte colony-stimulating factor-associated aortitis in the Japanese Adverse Drug Event Report database

• Published in: Cytokine (Philadelphia, Pa.) Vol. 119; pp. 47 - 51
• Main Authors: Oshima, Yasuo, Takahashi, Satoshi, Tani, Kenzaburo, Tojo, Arinobu
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.07.2019
• Subjects: Adverse event; Aortitis; Big data; Granulocyte colony-stimulating factor (G-CSF); Japanese Adverse Drug Event Report; Real world evidence (RWE); RWE; AE; EPO; Adverse event; JADER; G-CSF; BCPNN; ADR; EPPV; Granulocyte colony-stimulating factor (G-CSF); RMP; MAH; Japanese Adverse Drug Event Report; DUI; Aortitis; Big data; Real world evidence (RWE); PMDA
• ISSN: 1043-4666; 1096-0023; 1096-0023
• DOI: 10.1016/j.cyto.2019.02.013

•Anecdotal case reports have suggested association between G-CSF and aortitis.•Relationship between G-CSF and aortitis had remained unclear.•Japanese pharmacovigilance database JADER harbors 102,014 subjects with malignancies.•The adjusted odds ratio for aortitis in the malignancy cases with G-CSF was 45.87 [19.16, 109.8].•The difference was statistically significant. This supports the link between G-CSF and aortitis. Granulocyte colony-stimulating factor (G-CSF) is the standard-of-care therapy for chemotherapy-associated neutropenia in patients with malignancies. Recent case reports have implied that G-CSF treatment may be associated with the development of aortitis, but the precise nature of the relationship is unclear. We investigated the association between G-CSF and risk for aortitis in patients with various malignancies. We performed an observational study of 102,014 subjects with malignant neoplasms documented in the Japanese Adverse Drug Event Report (JADER) database between April 2004 and February 2018. The adjusted odds ratio (OR) and 95% confidence interval (CI) for aortitis in patients treated and not treated with G-CSF were estimated using multivariate logistic regression with R software. Among the 102,014 subjects, 25 developed aortitis. Of the 3409 and 98,630 subjects treated and not treated with G-CSF, 16 (0.47% [95% CI; 0.27, 0.76]) and 9 (0.01% [0.00, 0.02]) developed aortitis, respectively. Multivariate logistic regression indicated an association between G-CSF and aortitis (adjusted OR 45.87 [19.16, 109.8], p < 0.001). The values for filgrastim, pegfilgrastim, and lenograstim were 0.25% (0.07, 0.63), 1.58% (0.79, 2.81), and 0.24% (0.05, 0.69), respectively. G-CSF treatment was associated with a signal of increased risk for aortitis among patients with malignant neoplasms. Three different G-CSF agents were associated with such risk, implying that it is a class effect. However, we do not recommend changing G-CSF prescriptions, because our results may have been influenced by the limitations of the JADER database and because the benefit of G-CSF treatment outweighs the potential risk.