Across 5,298 FAERS echinocandin reports, caspofungin showed a unique DRESS signal (ROR 16.62) while caspofungin and micafungin showed exceptionally strong resistance-related signals

The study reports a unique positive signal for caspofungin and DRESS (drug reaction with eosinophilia and systemic symptoms): 46 cases, ROR = 16.62 (95% CI 12.43-22.22), IC025 = 3.21; the signal remained positive in a sensitivity analysis restricted to caspofungin as the sole suspected drug, with 31 cases, ROR = 14.89, IC025 = 3.05. The authors state that no previous pharmacovigilance study had quantified this risk at a population level, and that DRESS did not reach the top-20 PT threshold for micafungin, anidulafungin, or rezafungin. Based on four-method concordant positivity in FAERS spontaneous reports, plus a sensitivity analysis restricted to a single suspected drug; the authors also note that other medications (antibiotics, antiepileptic drugs, allopurinol) may contribute to DRESS and cannot be fully excluded in FAERS data.

The study reports exceptionally strong resistance-related signals for caspofungin and micafungin: caspofungin "pathogen resistance" ROR = 68.83 (IC025 = 4.64) and "drug resistance" ROR = 22.32 (IC025 = 3.62); micafungin "bronchopulmonary aspergillosis" ROR = 71.99 (IC025 = 4.55) and "Candida infection" ROR = 23.93 (IC025 = 3.59). The authors argue these ROR values far exceed typical thresholds for confounding by indication, suggesting clinical failure and suspected resistance are increasingly recognized in real-world practice, while stating explicitly that these signals do not confirm microbiologically confirmed resistance such as FKS mutations, which requires antifungal susceptibility testing and genetic analysis. Derived from simultaneous positivity across four disproportionality methods, but the authors note that terms such as "Aspergillus infection," "bronchopulmonary aspergillosis," and "fungal infection" partly reflect the underlying indications for echinocandin therapy.

The study confirms hepatobiliary and systemic signals shared across agents: caspofungin cholestasis (46 cases, ROR = 24.97, IC025 = 3.62) and blood alkaline phosphatase increased (47 cases, ROR = 19.10, IC025 = 3.37), micafungin transaminase elevation and hepatic function abnormal, anidulafungin hepatic failure (9 cases, ROR = 17.21); MODS, septic shock, sepsis, and respiratory failure were positive across all four drugs. The authors compare these results with product labels and prior meta-analyses, describing them as consistent with known safety information, and note that systemic events reflect the severe underlying conditions of treated patients rather than drug-specific toxicity. PT-level four-method concordant positive signals, corroborated at the SOC level: Infections and infestations had the highest proportion, General disorders and administration site conditions ranked second, and Hepatobiliary disorders and Investigations ranked third and fourth.

The study reports differences in report volume and signal detectability across agents: caspofungin 2,408 cases, micafungin 2,375, anidulafungin 416, rezafungin 99; 16 of caspofungin's top 20 PTs were positive signals, versus 14 for micafungin, 10 for anidulafungin, and only 5 for rezafungin. The authors use this to explain that the absence of resistance or DRESS signals for anidulafungin and rezafungin should not be interpreted as lower risk, because low report volume limits signal detection for rare events. Descriptive counts and per-drug signal counts; the authors explicitly caution that rezafungin results "should be interpreted with caution" and note it was approved only in 2023, with reports first appearing in 2024.

The work is aimed at clinicians using echinocandins, antifungal stewardship programs, and pharmacovigilance researchers, and applies to post-marketing safety surveillance of caspofungin, micafungin, anidulafungin, and rezafungin in real-world settings. The authors' proposed next steps include prospective registries to validate the caspofungin-DRESS association, pharmacogenomic studies to identify HLA or MRGPRX2 polymorphisms conferring susceptibility, and surveillance networks to monitor global echinocandin resistance trends. Their clinical recommendations are: monitor baseline and weekly eosinophil counts during caspofungin therapy, suspect DRESS and discontinue immediately with consideration of switching to micafungin or anidulafungin if fever, rash, eosinophilia, or visceral dysfunction develops; perform antifungal susceptibility testing for all echinocandins when response is poor after 5 to 7 days; and have local antifungal stewardship programs monitor resistance patterns.

The authors list scope limitations: FAERS is a spontaneous reporting system subject to under-reporting, selective reporting, and confounding by indication; the absence of denominator data precludes incidence calculation; US reports dominated, limiting generalizability; the time-to-onset analysis was compromised by data extraction artifacts; and rezafungin data are limited due to recent approval. The authors also caution that DRESS not appearing among the top 20 PTs for the other echinocandins should not be read as evidence of safety, since approval dates, global usage, reporting volumes, and reporter characteristics can substantially affect signal detection, and that resistance signals reflect reporting of suspected resistance or treatment failure rather than microbiologically confirmed resistance. The available text is an incomplete version: the contents of Figures 1, 2, and 3 and Supplementary Table S4 are not included, so temporal trends, time-to-onset distributions, and SOC proportions can only be taken from the narrative and the specific values in those figures cannot be checked.