A disproportionality analysis of adverse events caused by GnRHas from the FAERS and JADER databases

Abstract

Background: Gonadotrophin-releasing hormone analogs (GnRHas) play a significant role in addressing gynecological diseases, central precocious puberty, and cancer. However, ensuring the safety of GnRHas in real-world applications requires continuous vigilance. In light of this, we undertook a disproportionality analysis focused on adverse events (AEs) associated with GnRHas using data from both the FDA Adverse Event Reporting System (FAERS) and the Japanese Adverse Drug Event Report (JADER). We evaluated GnRHas-associated AEs and characterized the clinical priority of unlisted AEs caused by each GnRHa from the different databases.

Methods: In the disproportionality analysis, we applied two adjusted algorithms to identify signals related to GnRHas in the FAERS and JADER databases from 2004 to 2023. Additionally, we utilized the Statistical Analysis System (SAS, 9.4) to examine potential and high-aROR (adjusted reporting odds ratio) signals associated with GnRHas. We performed clinical priority assessment for suspicious PTs and an analysis of serious/non-serious outcomes. We also gathered information on the onset times of AEs linked with GnRHas from both databases.

Results: From January 2004 to September 2023, FAERS and JADER recorded a total of 50,360,413 and 1,440,200 AEs, respectively. Employing two algorithms, the suspicious preferred terms (PTs) related to leuprolide (Leu) were 562 potential PTs (44 unlisted in specifications), followed by goserelin (Gos) with 189 PTs (28 unlisted), triptorelin (Tri) with 172 PTs (28 unlisted), and Leu-JADER with 85 PTs (10 unlisted). At the same PT level, the differences in GnRHas between the two databases were observed, such as cardiac failure, diabetes mellitus, liver disorder, dementia, suicidal ideation, interstitial lung disease, urinary disorders, and hypertensive crisis. In an analysis of serious vs. non-serious outcomes, a total of 43 AEs of Leu were more likely to be reported as serious AEs with p < 0.05 (such as asthenia, urinary retention, diabetes mellitus, interstitial lung disease, gait disturbance, and so on), following by Tri (6 AEs), and Gos (4 AEs). Based on the clinical priority score, 41 PTs of Leu, 26 PTs of Tri, 24 PTs of Gos, and 8 PTs of Leu-JADER were graded as weak. There were 3 PTs of Leu, 2 PTs of Tri, 4 PTs of Gos, and 2 PTs of Leu-JADER that were graded as moderate. Notably, in the assessment of the relevant evidence, 2 PTs (loss of libido and urinary tract toxicity caused by Leu), 1 PT (electrolyte imbalance caused by Tri), and 2 PTs (anorexia and suicidal ideation caused by Gos) showed a strong level of evidence with "++." The differences in the signal strength of the same PTs from two databases were also worth noting. Moreover, the median onset time for GnRHas (Leu, Tri, and Gos) was 23 days (0, 298), 22 days (0, 181), and 217 days (29, 706), respectively, as median (Q1, Q3).

Conclusion: An examination of two databases revealed suspicious AEs associated with GnRHas. Our study found potential new AE signals of GnRHas and supported continuous clinical monitoring, pharmacovigilance, regional differences, and further studies of GnRHas.

Keywords: FAERS; GnRHas; JADER; adverse events; data mining.

FIGURE 1

Signal strength of adverse events of GnRHas at the SOC level from FAERS database. -, showed a negative signal (aROR₀₂₅ ≤ 1 and IC₀₂₅ ≤ 0); GnRHas, gonadotropin-releasing hormone analogs; SOC, system organ class; FAERS, FDA Adverse Event Reporting System; N, number of target adverse events; Leu, leuprolide; Tir, triptorelin; Gos, goserelin; aROR, adjusted reporting odds ratio; IC₀₂₅, adjusted lower limit of 95% confidence interval of the information component of BCPNN; BCPNN, Bayesian confidence propagation neural network.

FIGURE 2

Signal strength of the same adverse events of GnRHas at the PT level from FAERS database. -, showed a negative signal (aROR₀₂₅ ≤ 1 or IC₀₂₅ ≤ 0); p < 0.05, showed the significant difference by Pearson χ² test. GnRHas, gonadotropin-releasing hormone analogs; SOC, system organ class; PT, preferred term; FAERS, FDA Adverse Event Reporting System; N, cases of target adverse events; Leu, leuprolide; Tir, triptorelin; Gos, goserelin; aROR, adjusted reporting odds ratio; IC₀₂₅, adjusted lower limit of 95% confidence interval of the information component of BCPNN; BCPNN, Bayesian confidence propagation neural network.

FIGURE 3

Signal strength of the same adverse events of leuprolide and goserelin at the PT level from JADER databases. -, showed a negative signal (aROR₀₂₅ ≤ 1, and IC₀₂₅ ≤ 0); p < 0.05, showed significant difference by Pearson χ² test; IC₀₂₅ > 0 and aROR₀₂₅ > 1, showed a suspicious signal. PT, preferred term; JADER, Japanese Adverse Drug Event Report; SOC, system organ class; JADER, Japanese Adverse Drug Event Report; Leu, leuprolide; Gos, goserelin; N, number of target adverse events; aROR, adjusted reporting odds ratio; IC₀₂₅, adjusted lower limit of 95% confidence interval of the information component of BCPNN; BCPNN, Bayesian confidence propagation neural network.

FIGURE 4

Signal profiles of different GnRHas from FAERS and JADER databases (top 30 based on IC₀₂₅ of each GnRHas). GnRHas, gonadotropin-releasing hormone analogs; FAERS, FDA Adverse Event Reporting System; JADER, Japanese Adverse Drug Event Report; SOC, system organ class; PT, preferred term; Leu, leuprolide; Tir, triptorelin; Gos, goserelin; IC₀₂₅, adjusted the lower limit of 95% confidence interval of the information component of BCPNN; BCPNN, Bayesian confidence propagation neural network. IC₀₂₅ > 3.0 indicates a strong signal; 1.5 < IC₀₂₅ ≤ 3.0 indicates a medium-intensity signal; 0 < IC₀₂₅ ≤ 1.5 indicates a weak-intensity signal.