Increased risk for aseptic meningitis after amoxicillin or amoxicillin‐clavulanic acid in males: A signal revealed by subset disproportionality analysis within a global database of suspected adverse drug reactions

Abstract Purpose Drug‐induced aseptic meningitis (DIAM) is an inflammation of the membranes of the central nervous system caused by certain medications. It is a rare clinical entity whose risk factors are not yet fully elucidated. A local pattern of disproportionality within a global database of suspected adverse drug reactions (ADRs) revealed an increased reporting of aseptic meningitis and amoxicillin‐clavulanic acid (AC) in males. The aim of this report is to explore the clinical probability of a higher risk in males to support the use of statistical methods to identify subgroups at risk for adverse drug reactions. Methods Disproportionality analysis was performed for all drug‐adverse event (AE) pairs in the entire database and for the subsets of males and females. AC‐aseptic meningitis was highlighted for an increased disproportionality in the male subgroup in the absence of an elevated disproportionality measure for the database overall. A clinical review was undertaken. Results Clinical review revealed a similar statistical pattern of gender difference observed for amoxicillin, evidence to suggest a delayed type 4 hypersensitivity reaction with Th1 cells as a mechanism for amoxicillin‐aseptic meningitis, the existence of sex differences in immune responses (Th1/Th2), and an analogous increased risk of drug‐induced liver injury by AC in males. Conclusions Subgroup disproportionality analysis has revealed a larger than expected number of reports of aseptic meningitis after amoxicillin and AC in males. Evidence synthesis supports the statistical finding. Further exploration of spontaneous databases with more extensive analyses could usher in a new era of “precision pharmacovigilance.”

pairs in the entire database and for the subsets of males and females using methodology for local pattern discovery. 3 AC-aseptic meningitis was highlighted for an increased disproportionality in the male subgroup in the absence of an elevated disproportionality measure for the database overall. IC 0005 is the lower endpoint of a 99.9% credibility interval for the information component (IC), and as such was used to support analysis of subgroup specific associations between substances and AE. Initial manual assessment expanded the review to include amoxicillin, which included both an increased disproportionality in males as well as in the entire database. A thorough clinical review of the drug-ADR-risk group triplet was undertaken, which included further review of spontaneously reported ADRs, the reported case series, as well as the medical literature for evidence of biological plausibility of an increased risk in males.

| RESULTS
As of September 9, 2018, there was a total of 2695 individual case safety reports in VigiBase, which included the preferred term (PT) "meningitis aseptic"; 56.7% of the reports described females, 37.3% males, and in 6.0% of the reports, gender was not stated. The most common drugs reported in association with aseptic meningitis were human immunoglobulins (35.3%) and ibuprofen (6.6%). There was a total of 47 reports with amoxicillin (26 males, 21 females) and 24 for AC (18 males, 6 females).
For the combination of amoxicillin and the MedDRA PT, meningitis aseptic, 26 cases were observed compared with seven expected in males (IC 0005 0.80), and 21 cases were observed and nine expected in females (IC 0005 -0.14). Further investigation revealed a similar pattern of gender difference for AC; 18 cases were observed compared with seven cases expected in males (IC 0005 0.04); six cases were observed and nine expected in females (IC 0005 -3.14).
After removing suspected duplicate reports, there were a total of 36 unique cases reported in males (Table 1) • Sex differences in immune responses, such as delayed type hypersensitivity reactions, which can be a mechanism for drug induced aseptic meningitis, have been described.
• More extensive analyses of databases of spontaneous adverse drug reaction (ADR) reports could be used to assist in the identification of potential risk factors for ADRs.    is a key Th1-type cytokine released from activated T cells in druginduced hypersensitivity. 9 There is growing recognition of the existence of sex differences in immune responses. Contributing factors to the differential development and function of the immune system between males and females include sex chromosome genes and sex hormones as well as nutritional status and composition of the microbiome in the gastrointestinal tract. The results of these differences in immune responses is evidence of the differential susceptibility of males and females to autoimmune diseases, malignancies and infectious diseases, and of responses to vaccination. More specifically, there is evidence that females have higher CD4+ T cells and greater T cell activation and proliferation while males have higher CD8+ T cell frequencies.
Females tend to be polarised towards Th2-type responses, while males are more Th1-biased and have more regulatory T cells. 10   the potential for risk identification at the level of statistical screening within large databases. Further exploration of spontaneous databases with more extensive analyses could usher in a new era of "precision pharmacovigilance."

ETHICS STATEMENT
The authors state that no ethical approval was needed.