Safety evaluation of irinotecan: a real-world disproportionality analysis using FAERS and JADER databases during the time period 2004-2024

Abstract

Introduction: Irinotecan is a widely used chemotherapeutic agent for treating colorectal, pancreatic, and ovarian cancers. Despite its therapeutic efficacy, the safety profile of irinotecan necessitates continuous pharmacovigilance due to its association with severe adverse drug events (ADEs). Given its global use, cross-national signal detection may reveal region-specific risks or unrecognized adverse effects.

Methods: We conducted a retrospective pharmacovigilance analysis of irinotecan-associated ADEs using two large spontaneous reporting systems: the U.S. FDA Adverse Event Reporting System (FAERS) and the Japan Adverse Drug Event Report (JADER) database. ADE reports between 2004 and 2024 were extracted. Disproportionality analyses were performed using four methods: Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and Multi-item gamma Poisson shrinker (MGPS).

Results: A total of 11,344 ADE reports from FAERS and 7,822 from JADER were identified. These reports involved 27 system organ classes (SOCs). In FAERS, the most frequently affected SOC was gastrointestinal disorders (n = 6,888), while in JADER it was blood and lymphatic system disorders (n = 3,389). Disproportionality analysis revealed 388 and 67 preferred terms (PTs) significantly associated with irinotecan in FAERS and JADER, respectively, with 38 overlapping signals. These included both expected ADEs (e.g., neutropenia, diarrhea, thrombocytopenia, stomatitis) and unexpected signals such as second primary malignancies, hyperammonaemia, and hiccups. Notable FAERS-specific signals included skin toxicity (n=100, ROR 33.89 (27.79-41.34), PRR 33.80, EBGM05 28.03, IC025 4.76), aphasia [n=65, ROR 3.57 (2.8-4.55), PRR 3.56, EBGM05 2.90, IC025 1.47], and hepatic failure [n=56, ROR 3.09 (2.38-4.02), PRR 3.09, EBGM05 2.48, IC025 1.24], while JADER-specific signals included fatigue [n=73, ROR 4.69 (3.71-5.93), PRR 4.67, EBGM05 3.57, IC025 0.51], hyperammonaemia [n=67, ROR 7.24 (5.56-9.27), PRR 7.21, EBGM05 5.32, IC025 1.10], and cholinergic syndrome [n=27, ROR 5.54 (3.76-8.16), PRR 5.53, EBGM05 3.61, IC025 0.74]. Over half of all reported ADEs occurred within one month of irinotecan administration (53.1% in FAERS, 61.7% in JADER). The median time to onset was 28 days [IQR 9-76] in FAERS and 17 days [IQR 9-57] in JADER.

Discussion: This comparative analysis revealed multiple consistent and unexpected signals related to irinotecan use. The findings emphasize the importance of region-specific pharmacovigilance and the need for heightened awareness of both labeled and unlabeled toxicities. Our results support continued monitoring and further investigation into temporal patterns and regional differences in irinotecan-related adverse events to enhance clinical safety.

Keywords: FAERS; JADER; adverse drug event; irinotecan; real-world pharmacovigilance analysis.

FIGURE 1

Flowchart of the entire study, including data cleaning, analysis methods, and main results. FAERS: The U.S. FDA Adverse Event Reporting System; JADER: Japanese Adverse Drug Event Report; Q1: first quarter; Q2: second quarter; Q3: third quarter; Q4: fourth quarter; PT: preferred term; PS: primary suspect.

FIGURE 2

Signal detection at the SOC level. (A) Annual distribution of ADE reports in the FAERS and JADER databases, displayed as a bar chart. Submitter outcome data from FAERS (B) and JADER (C) were illustrated using donut plots to visually represent the distribution of reported clinical outcomes. (D) Number of ADE reports in FAERS and JADER at the SOC level. (E, F) Signal detection at the SOC level in FAERS and JADER, with ROR values and their 95% confidence intervals (95% CI) visualized. SOCs meeting the signal threshold are highlighted in red. SOC: system organ class; ADE: adverse drug event; FAERS: The U.S. FDA Adverse Event Reporting System; JADER: Japanese Adverse Drug Event Report; ROR: reporting odds ratio; Q3: third quarter; Q4: fourth quarter.

FIGURE 3

Bar plots displaying the top 50 PTs in terms of reported cases in FAERS (A) and JADER (B) databases. The color indicates the SOC of the corresponding PT. The percentage values labeled in the figure represent the proportion of cases with such ADEs out of the total reported ADEs. SOC: system organ class; PT: preferred term.

FIGURE 4

Forest plot showing the top 50 PT entries (ranked by reported cases) that simultaneously satisfy the four disproportionality methods with positive signal strength in the FAERS database. The organ arrows indicate that the lower limit of the 95% confidence interval for the ROR exceeds 20. The heatmap on the right displays the signal values under different algorithms. The darker the color, the stronger the signal value. Asterisks (*) indicate unexpected signals not listed in the drug label. PT: preferred term; FAERS: The U.S. FDA Adverse Event Reporting System.

FIGURE 5

Forest plot showing PT entries that simultaneously satisfy the four disproportionality methods with positive signal strength and at least 10 reported cases in the JADER database. The blue arrow indicates that the lower limit of the 95% confidence interval for the ROR exceeds 20. The heatmap on the right displays the signal values under different algorithms. The darker the color, the stronger the signal value. Asterisks (*) indicate unexpected signals not listed in the drug label. PT: preferred term; JADER: Japanese Adverse Drug Event Report.

FIGURE 6

Time to onset (TTO) analysis (counted in days) of irinotecan-related ADEs. Bar graphs depict the number and proportion of ADE reports at different time intervals in FAERS (A) and JADER (B). Overall description and Weibull distribution test analysis of valid TTO reports in FAERS (C) and JADER (D). ADE: adverse drug event; IQR: interquartile range.