Pharmacovigilance insights into drug-induced hypercalcemia: evidence from the U.S. FDA Adverse Event Reporting System...

Drug-induced hypercalcemia (DIH) is an often-overlooked adverse event with serious consequences, yet evidence remains limited. This study aimed to identify drugs with disproportionate reporting signals for hypercalcemia and characterize their clinical features using the U.S. FDA Adverse Event Reporting System (FAERS) database, thereby providing pharmacovigilance evidence to support clinical risk awareness and monitoring prioritization. A worldwide observational case-non-case study was conducted using FAERS data from January 2004 to December 2024. Duplicate reports were excluded, and drug names were standardized via the DrugBank database. Four algorithms-reporting odds ratio [ROR], proportional reporting ratio [PRR], information component [IC], and empirical Bayes geometric mean [EBGM]) - were applied to screen for potential signals by detecting statistical disproportionality. Cumulative risk curves were employed to characterize the time-to-onset patterns of DIH. A total of 18,806 valid DIH reports were identified for 967 drug pairs, of which 107 drugs were identified as potential signals based on disproportionality analysis (meeting all four algorithm criteria). The top three drug classes associated with DIH were antineoplastic and immunomodulating agents (35.5%), systemic hormonal preparations (21.9%), and musculo-skeletal system drugs (10.6%). Teriparatide (2539 cases), denosumab (894 cases), and lenalidomide (800 cases) were the most frequently reported drugs associated with hypercalcemia. Additionally, statistically significant disproportionality signals were detected for several drugs not currently labeled for DIH risk, such as bortezomib, octreotide, and atezolizumab, warranting further evaluation. The median onset time of DIH was 52 days (IQR, 9-223 days): the anti-infective drug group had the earliest DIH onset time (median 12 days; IQR, 3-51 days), while the musculo-skeletal system drug group had a relatively longer median latency period (151 days; IQR, 17-510 days). This FAERS-based disproportionality analysis identified 107 drugs across 14 classes with reporting signals for hypercalcemia and delineated their time-to-onset patterns. Although these signals do not establish causality and are susceptible to reporting bias and confounding, they identify therapies for which targeted calcium monitoring may be warranted, and may help prioritize future regulatory review, including potential labeling revisions. Further confirmation in studies with laboratory-verified calcium values and robust confounding control is needed.