Huixiang Li1†
Haijian Wei
Jianxin Cui
Jiaojiao Chen- 1Department of Pharmacy, Yantai Yuhuangding Hospital Affiliated to Qingdao University, Yantai, Shandong, China
- 2Department of Organ Transplantation, Yantai Yuhuangding Hospital Affiliated to Qingdao University, Yantai, Shandong, China
- 3Department of pharmacy, Dongyang Red Cross Hospital, Dongyang, Zhejiang, China
Background: Alopecia is a significant adverse effect that profoundly impacts quality of life. Although numerous medications are implicated, the real-world risk profiles across drug classes and patient demographics remain poorly quantified.
Objective: To identify and characterize drugs associated with alopecia using real-world data from the FDA Adverse Event Reporting System (FAERS).
Methods: FAERS reports from Q1 2004 to Q4 2024 were analyzed using four disproportionality methods (ROR, PRR, BCPNN, MGPS) to detect signals of drug-alopecia associations. Subgroup analyses were conducted by age, gender, and drug category. Time-to-onset (TTO) was analyzed using the Weibull distribution model.
Results: A total of 181,838 reports with drug-associated alopecia were identified. The mean age was 53.84 ± 16.28 years, and 76.82% of reports were from females. Oncology medications showed strongest association (37.5%), especially docetaxel (ROR = 70.38). Endocrine (18.8%) and immune system medications (10.9%) were also prominent. The TTO analysis revealed a bimodal distribution, with 40.2% of cases occurring within 30 days and 13.1% manifesting at 240–360 days. Males experienced a significantly shorter onset latency compared to females (108 days vs. 236 days, P < 0.001). Oncology drugs also showed shorter latency than non-oncology agents (198 vs. 308 days, P < 0.001). Notably, comparison with United States prescribing information revealed that 23.4% of high-signal drugs lacked documentation of alopecia in their official labels.
Conclusion: This large-scale pharmacovigilance study identified 64 drugs with significant alopecia signals, highlighting distinct demographic patterns and latency periods. The findings underscore the need for heightened clinical vigilance, gender-specific monitoring, and updates to labels to better reflect real-world risks.
1 Introduction
Alopecia, characterized by abnormal hair loss, significantly impairs quality of life (Vary, 2015; Kearney et al., 2025). It encompasses various clinical types, among which androgenetic alopecia (AGA) and alopecia areata (AA) are the most prevalent (Alkhalifah et al., 2010; Mubki et al., 2014). AGA exhibits an extremely high prevalence, affecting approximately 80% of men and 50% of women by age 70 (Devjani et al., 2023; Gupta et al., 2025). As the second most common type, the incidence rate of AA has increased significantly in recent years, particularly among children and adolescents (McKenzie et al., 2022). Despite not being life-threatening, alopecia profoundly impacts psychological wellbeing and social functioning (Hunt and McHale, 2005; Muntyanu et al., 2023).
The etiology of alopecia is complex and multifactorial, involving genetic, hormonal, autoimmune, nutritional, and environmental factors (Pratt et al., 2017; Kinoshita-Ise et al., 2023). Critically, medications are recognized as important modifiable triggers. (Alhanshali et al., 2023; Ezemma et al., 2024; Ravipati et al., 2024). Drug-induced alopecia (DIA) typically presents as non-scarring, diffuse hair loss. It primarily occurs through two mechanisms: anagen effluvium, involving rapid hair loss due to direct cytotoxicity (e.g., chemotherapy), and the more common telogen effluvium, characterized by delayed, gradual shedding. (Piraccini et al., 2006; Trüeb, 2010). A key clinical challenge is identifying the causative drug, often complicated by variable latency periods and concurrent use of multiple medications.
The FDA Adverse Event Reporting System (FAERS), which collates adverse events (AE) data from healthcare professionals, patients, and manufacturers, provides a valuable resource for detecting drug safety signals in real-world populations (Sakaeda et al., 2013). While Hill et al. previously described drug-induced hair loss using FAERS data, their study did not include quantitative analyses stratified by key variables like age, gender, or drug category, nor did it identify specific risk signals for DIA (Hill et al., 2025). Therefore, our study aimed to systematically identify and characterize DIA signals through comprehensive disproportionality analyses to delineate risk profiles and provide actionable clinical insights.
2 Methods
2.1 Data source and processing
The data for this study were extracted from FAERS database, covering reports from the first quarter of 2004 (Q1 2004) to the fourth quarter of 2024 (Q4 2024). Raw quarterly ASCII files were downloaded from the FAERS public repository (https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html). The downloaded data were imported into SAS software (version 9.4) for cleaning and statistical analysis. To ensure data integrity, a structured de-duplication process was applied. Duplicate reports were identified and removed as recommended by the FDA (Tregunno et al., 2014). Specifically, for reports sharing the same CASEID, the record with the most recent FDA_DT was retained. If the CASEID and FDA_DT were the same, the report with the highest PRIMARYID was retained.
2.2 Identification of target AE reports
In the FAERS database, AE reports are coded using the preferred terms (PT) from the Medical Dictionary for Regulatory Activities (MedDRA). To identify reports of alopecia, we employed Standardized MedDRA Queries (SMQs) (MedDRA v26.0) in this study. SMQs are predefined sets of PTs that represent related clinical conditions or syndromes (Mozzicato, 2007), with two search modalities: broad-scope search and narrow-scope search. In this study, we adopted the narrow-scope search, restricting inclusion to PTs with a well-established relationship to alopecia (Supplementary Table S1). AE reports were considered as target cases if they contained any PTs specified in Supplementary Table S1. Crucially, to address potential confounding from polypharmacy (multiple drugs reported per case) and to focus on the associations with the highest suspicion, our analysis was strategically restricted to drugs designated as the “Primary Suspect” in the reported events. Reports listing the drug of interest as “Secondary Suspect”, “Concomitant”, or “Interacting” were excluded. Furthermore, each case report contributed only once to the disproportionality analysis for a specific drug, and no statistical weighting was applied for the presence of other drugs within the same report. This conservative approach prioritizes signal specificity over sensitivity, aiming to highlight the most robust drug-event associations.
2.3 Detection of ADR signals
Disproportionality analysis is a special data mining algorithm developed for the quantitative detection of ADR signals in large pharmacovigilance databases (Yan et al., 2022). Based on the standard 2 × 2 contingency table (Supplementary Table S2), this method quantifies the disparity between the observed frequency and the expected background frequency for a specific drug-AE pair, thereby establishing potential statistical associations between drugs and AEs (Li et al., 2024). In this study, we employed four disproportionality analysis methods to detect positive signals of drug-associated alopecia: reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multi-item gamma Poisson shrinker (MGPS) (Candore et al., 2015). Positive signal criteria are provided in Supplementary Table S3. To strengthen evidence for robust association between the identified drugs and alopecia, a drug was selected as a positive signal only if it met the criteria of all four methods simultaneously. Subgroup analyses stratified by gender and age were conducted to explore the potential variation in drug-associated alopecia signals across these demographic dimensions. P-values from these subgroup analyses were adjusted for multiple comparisons using the Bonferroni correction, with statistical significance defined by the corrected threshold. Furthermore, in order to conduct a more in-depth investigation into the relationship between the positive drugs and AEs, stratification analyses were performed based on number of reports, drug category, and signal intensity. Additionally, to characterize the latency period of drug-associated alopecia, the Weibull distribution model was used to analyze the time-to-onset (TTO) of AEs for each drug (Kinoshita et al., 2020). We further compared differences in TTO distributions across drug categories and between genders. Finally, to assess the alignment between our pharmacovigilance signals and established safety labels, we further verified the signals by checking for adverse reactions in the official United States prescribing information (FDA’s Drugs@FDA search database). For each drug, we systematically examined the “Adverse Reactions” section of the current, approved label for any mention of “alopecia”, “hair loss”, or related terms.
2.4 Statistical analysis
Descriptive analysis was performed to summarize and present the clinical characteristics of the patients in drug-associated alopecia reports. Categorical variables were presented as frequencies and percentages, while continuous variables were expressed as median with interquartile range (IQR) or mean ± standard deviation (SD) based on distribution normality. Between-group comparisons of the TTO distributions were performed across drug categories and gender strata using non-parametric tests (Wilcoxon rank-sum test). Statistical analysis was conducted using a combination of software tools, including Microsoft Excel 2019, Origin (version 2021), SPSS (version 23.0; IBM, United States), GraphPad Prism (version 8.0.2) and R (version 4.2.3), where p < 0.05 was considered statistically significant.
3 Results
3.1 Descriptive analysis of drug-associated alopecia reports
From Q1 2004 to Q4 2024, the FAERS database recorded a total of 22,375,298 patients reporting AEs, of which 181,049 patients were reported to have experienced drug-associated alopecia, involving 181,838 reports. Figure 1 illustrated the detailed data processing. And Table 1 presented the clinical characteristics of AEs associated with alopecia. The mean age of the patients was 53.84 ± 16.28 years, with females accounting for the majority at 76.82%. Specifically, in females, the reported age for drug-associated alopecia was primarily concentrated in the 50–65 years range, whereas in males, it was concentrated in the 55–70 years range (Figure 2A). Furthermore, we observed a gradual increase in the number of reported cases of drug-related alopecia over the years, peaking in 2018 and subsequently showing a downward trend in the past 6 years. Notably, the incidence was significantly higher in females than in males annually (Figure 2B). The patients had a median weight of 72.00 kg (IQR 60.33–86.18 kg). Consumers were the main reporters, accounting for 55.88%. In terms of the severity of the AE reports, the proportions of patients with non-serious and serious reports were roughly similar, at 53.31% and 46.69% respectively. As for the outcomes of the aforementioned AEs, the most frequent outcome was “Other Serious (important medical event)” (66.21%: 56,729 females, 7,020 males), followed by “Hospitalization-Initial or Prolonged” (16.84%: 14,561 females, 2,509 males) (Figure 2C). Among patients with known administration routes, oral (28.92%), subcutaneous (10.85%), and intravenous (10.02%) administration were the predominant routes (Figure 2D). Regarding reporting region, the United States submitted the majority of reports, followed by Canada and United Kingdom, with 134,199, 13,238 and 5,380 reports, respectively (Figure 2E). For more details, please refer to Figure 2 and Table 1.
Figure 1. Flowchart for cleaning process for drug-associated alopecia data in the FAERS database. Abbreviations: FAERS, Food and Drug Administration Adverse Event Reporting; ROR, reporting odds ratio; PRR, proportional reporting ratio; BCPNN, Bayesian confidence propagation neural network; MGPS, Multi-item Gamma Poisson Shrinker.
Table 1. Baseline data of drug-related alopecia patients reported in FAERS database.
Figure 2. Baseline characteristics of adverse reactions associated with alopecia. (A) Population pyramid of patients with drug-associated alopecia categorized by gender and age. (B) Bar chart showing the annual reporting counts of drug-associated alopecia by gender. (C) Distribution of adverse reaction outcomes for patients with drug-associated alopecia categorized by gender. (D) Bar chart of drug administration routes for patients with drug-associated alopecia. (E) Heatmap of reporting countries for patients with drug-associated alopecia.
3.2 Distribution of drug categories associated with alopecia signals
Disproportionality analysis was used to identify drugs with positive signals for alopecia. Firstly, the volcano plot was generated to visualize the relationship between alopecia reports and the suspected drugs (Figure 3). In the plot, the x-axis represents the logarithm of the ROR. A positive value on x-axis suggests that AEs associated with drug-associated alopecia were reported more frequently than other AEs. The y-axis shows the negative logarithm of the p-adjust value, which is derived from the p-value following Fisher’s exact test and Bonferroni correction. A positive value on the y-axis indicates a highly significant difference. Therefore, drugs located in the upper-right quadrant of the graph exhibited both a strong association signal (high ROR) and statistical significance.
Figure 3. Volcano plot of drug-associated alopecia. Abbreviations: ROR, reporting odds ratio; P-adjust, p-value after Bonferroni correction.
Subsequently, we analyzed a dataset of 1,545 drugs that reported AEs related to alopecia and identified 71 drugs exhibiting positive signals using all four predefined screening criteria simultaneously (Figure 4A). To focus on drugs where alopecia is likely an unintended adverse effect, we excluded those which are approved or recommended by clinical guidelines for the treatment of alopecia (e.g., AA, AGA) (Supplementary Table S4). Their positive signals were considered potentially confounded by reports of treatment failure or inadequate efficacy. This resulted in 64 drugs for final analysis. Among these, oncology medication constituted the largest proportion (24 drugs, 37.5%), followed by endocrine system medication (12 drugs, 18.8%), immune system medication (7 drugs, 10.9%), skin system medication (6 drugs, 9.4%), nervous system medication (4 drugs, 6.3%), and other therapeutic categories (11 drugs, 17.2%). Figure 4B visualized this distribution with circle sizes proportional to AE report numbers. The top three drugs by AE reports per category were: Oncology - docetaxel, palbociclib, vismodegib; Endocrine system - levothyroxine, letrozole, anastrozole; Immune system - teriflunomide, peginterferon alfa-2a, leflunomide; Skin system - fumaric acid, ketoconazole, acitretin; Nervous system - erenumab, galcanezumab, fremanezumab; Other medications - pentosan polysulfate, pegvaliase, permethrin. Additional details were provided in Figure 4B.
Figure 4. Distribution of drugs with positive signals for drug-associated alopecia. (A) Venn diagram of four disproportionality analysis methods. (B) Doughnut chart of drugs distribution. The size of each circle represents the number of adverse event reports. Abbreviations: ROR, reporting odds ratio; PRR, proportional reporting ratio; BCPNN, Bayesian confidence propagation neural network; MGPS, Multi-item Gamma Poisson Shrinker.
3.3 Risk assessment of drugs with positive signals for alopecia
The top 50 drugs with positive signals from four disproportionality analyses were presented in Figure 5. Docetaxel showed the highest association strength (ROR = 70.38, 95%CI: 69.43–71.33), followed by Spinosad (ROR = 41.01, 95%CI: 16.12–104.33) and Selenium (ROR = 25.29, 95%CI: 10.15–62.97). Detailed results are available in Table 2 and Figure 5. Subgroup analyses by age and gender further delineated the association profiles. In the age-stratified analysis, 10, 31, 28, and 34 drugs showed significant signals in the <18, 18–44, 45–64, and ≥65-year-old subgroups, respectively. After applying the Bonferroni correction, five additional drugs, anastrozole, tisotumab vedotin, enfortumab vedotin, tazemetostat, and colchicine and probenecid, lost statistical significance in the age-stratified analysis (Supplementary Figure S1). Similarly, in the gender subgroup analysis, 45 and 32 drugs showed positive signals in the subgroups of female and male, respectively. And seven drugs, futibatinib, colchicine and probenecid, infigratinib, acetohydroxamic acid, nirogacestat, liothyronine, and levothyroxine and liothyronine, were no longer significant after Bonferroni correction (Supplementary Figure S2). All Bonferroni-adjusted p-values were presented in the forest plots.
Figure 5. Forest plot and signal value heatmap of top 50 drugs with positive signals from four disproportionality analysis methods. Abbreviations: ROR, reporting odds ratio; PRR, proportional reporting ratio; BCPNN, Bayesian confidence propagation neural network; MGPS, Multi-item Gamma Poisson Shrinker; CI, confidence interval.
Table 2. Disproportionality analysis results of drugs with positive signals for drug-associated alopecia.
3.4 Stratified evaluation based on reporting frequency, drug category, risk level, and drug-induced time
The drugs sorted by the number of reports were presented in Figure 6A. During the study period, the top four drugs reporting the highest number of alopecia-associated AEs were docetaxel, teriflunomide, levothyroxine and palbociclib, each exceeding 5,000 reports. For each drug category, the top three drugs by ROR values for endocrine medication were estradiol, norethisterone and relugolix (ROR = 16.99, 95%CI: 13.33–21.65), levothyroxine (ROR = 8.11, 95%CI: 7.89–8.33), elagolix, estradiol and norethisterone (ROR = 7.01, 95%CI: 5.39–9.12); for immune system medication, the top three drugs in ROR values were teriflunomide (ROR = 13.06, 95%CI: 12.72–13.42), leniolisib (ROR = 8.44, 95%CI: 5.21–13.67), and teprotumumab (ROR = 5.86, 95%CI: 5.28–6.52); for nervous system medication, the top three drugs in ROR values were phentermine and topiramate (ROR = 5.12, 95%CI: 4.37–6.00), galcanezumab (ROR = 4.01, 95%CI: 3.71–4.33), and erenumab (ROR = 3.56, 95%CI: 3.37–3.77); for oncology medication, the top three drugs in ROR values were docetaxel (ROR = 70.38, 95%CI: 69.43–71.33), vismodegib (ROR = 20.88, 95%CI: 19.75–22.08), and ripretinib (ROR = 10.35, 95%CI: 9.40–11.39); for skin system medication, the top three drugs in ROR values were selenium sulfide (ROR = 11.94, 95%CI: 8.80–16.19), ketoconazole (ROR = 8.60, 95%CI: 7.71–9.60), and benzyl alcohol (ROR = 8.06, 95%CI: 2.56–25.35); for other medication, the top three drugs in ROR values were Spinosad (ROR = 41.01, 95%CI: 16.12–104.33), selenium (ROR = 25.29, 95%CI: 10.15–62.97), and permethrin (ROR = 11.38, 95%CI: 9.30–13.93). More details were provided in Figure 6B. Based on the IC025 values, the 64 drugs were classified into high-risk (IC025 > 3), medium-risk (1.5 < IC025 ≤ 3), and low-risk (IC025 ≤ 1.5) categories (Wu et al., 2025). This resulted in 6 high-risk drugs (9.38%), 26 medium-risk drugs (40.63%), and 32 low-risk drugs (50.00%). Docetaxel (IC025 = 5.61), vismodegib (IC025 = 4.18), and teriflunomide (IC025 = 3.57) represented the top three highest-risk drugs (Figure 7A). Additionally, drugs were categorized by drug-induced onset time duration, segmented by quartiles. The top three drugs with the longest median drug-induced onset time were pegvaliase (median time = 187 days), thyroid (median time = 146 days), and mogamulizumab (median time = 143 days) (Figure 7B).
Figure 6. Reporting frequency and therapeutic categorization of drugs with positive alopecia signals. (A) Drugs ranked by the number of adverse event reports. (B) Drugs categorized by therapeutic class and ranked by ROR within each category.Abbreviations ROR, reporting odds ratio.
Figure 7. Risk level and onset time distribution of drugs with positive alopecia signals. (A) Drugs categorized by risk level based on IC025 values. (B) Drugs ranked by median drug-induced onset time. Abbreviations: IC025, lower limit of the 95% CI of the information component.
3.5 Proportional distribution of drugs at different PT levels
To further summarize the overall drug characteristics of DIA, we integrated the positive/negative distribution of ADR signal for drugs at the PT level and the corresponding drug class distributions (anatomical therapeutic chemical (ATC) classification system) (Supplementary Figure S3). Overall, the number of drugs with positive ADR signals was lower than those with negative ADR signals in all groups. Notably, antineoplastic agents (L01) and immunosuppressants (L04) exhibited the highest proportion of positive signals at most PT levels, except for non-scarring alopecia, injection site alopecia, and application site alopecia. No positive signals were observed for loose anagen syndrome, and no drugs were reported to induce seborrhoeic alopecia.
Figure 8 showed the top 10 drugs with the highest reporting proportions at the PT level. Docetaxel was the most frequently reported drug for alopecia, AA, alopecia totalis, and AGA, with reporting proportions of 46.98%, 65.10%, 67.71%, and 15.49%, respectively. Levothyroxine showed the highest reporting proportion for diffuse alopecia (75.79%). Exenatide (9.52%), ribavirin (18.39%), and erlotinib (30.00%) accounted for the highest proportions of hypotrichosis, alopecia universalis, and alopecia scarring, respectively. Only botulinum toxin type a, deoxycholic acid, and peginterferon beta-1a were associated with injection site alopecia. Deoxycholic acid, triamcinolone, and ketoconazole were the only drugs reported to cause application site alopecia, while only adalimumab and olmesartan reported non-scarring alopecia.
Figure 8. The distribution of top 10 drugs at different performed terms.
3.6 Comparison of drug-induced onset time among gender and drug category
Figure 9A illustrated the time distribution of drug-induced onset time for drugs with positive signals. Among them, 40.16% (5,054 reports) of cases induced alopecia AEs within 30 days. And there were 2,380 reports documented onset times of alopecia between 240 and 360 days Figure 9B presented the TTO bubble plot for individual drugs, where circle size corresponded to report frequency. Among them, docetaxel had the highest number of reports (median time = 119 days). Pegvaliase showed the longest median onset time, at 187 days, although only 3 reports have reporting time data. The induction time of most drugs was within 15–50 days. See Table 3 for further details. We further analyzed gender differences in TTO and found that only docetaxel use resulted in a significantly shorter latency in males compared to females (P < 0.001, Supplementary Table S5).
Figure 9. (A) The time distribution of drug-induced alopecia onset time for drugs with positive signals. (B) The TTO bubbles distribution diagram for each drug. Abbreviations: TTO, time to onset.
Table 3. Time to onset for drug-induced alopecia by different drugs.
The cumulative curves and violin plots were used to evaluate the differences in drug-induced time between genders and different category of drugs, as shown in Figure 10. The results illustrated that males experienced a significantly shorter drug-induced onset time compared to females (mean days, 108 days vs. 236 days, respectively; P < 0.001). Furthermore, a comparison between oncology and non-oncology medications demonstrated a significant difference, with oncology medications inducing a notably shorter onset time than non-oncology medications (mean time, 198 days vs. 308 days, respectively; P < 0.001).
Figure 10. Comparison of time-to-onset for drug-induced alopecia by gender and drug categories. (A) Cumulative risk curve for drug-induced alopecia onset times between female and male groups. (B) Violin plot for drug-induced alopecia onset times between female and male groups. The results illustrated that males have significantly shorter drug induction times compared to females (P < 0.01). (C) Cumulative risk curve for drug-induced alopecia onset times between oncology and non-oncology medications. (D) Violin plot for drug-induced alopecia onset times between oncology and non-oncology medications. The results illustrated that oncology medications have significantly shorter drug induction times compared to non-oncology medications (P < 0.01).
3.7 The relationship between medications and alopecia
Based on reports from PubMed and Web of Science, medications with well-established associations with DIA are primarily concentrated in five major therapeutic classes: antineoplastic agents (e.g., cyclophosphamide, paclitaxel), immunomodulators (e.g., tacrolimus, interferon alfa-2b), antiseizure medications (e.g., valproate, carbamazepine), anti-TNF biologics (e.g., infliximab, adalimumab), and hormonal agents (e.g., levonorgestrel-containing contraceptives). The mechanisms underlying DIA vary among these categories. Antineoplastic medications primarily inhibit the mitosis of hair follicle cells, leading to anagen effluvium. Immunomodulatory medications may disrupt immune cell function, resulting in telogen effluvium. Antiseizure medications, anti-TNFs, and oral contraceptives may cause telogen effluvium through metabolic interference in hair follicles, follicle-targeted inflammatory responses, and hormonal disruption of the follicle cycle, respectively (Figure 11A). By comparing with the United States prescribing information, we found that alopecia was documented in the labels of 49 drugs (76.6%) among the 64 drugs with positive signals. In contrast, 15 drugs (23.4%), such as erlotinib, ketoconazole, fremanezumab, galcanezumab, lacked this adverse reaction documentation in their official labels, despite showing significant signals (Figure 11B).
Figure 11. (A) The five common drug categories and representative drugs that are established or strongly suspected to be associated with alopecia based on existing literature. (B) Drugs with positive signals stratified by presence of alopecia in the adverse reactions section of drug inserts. Values in parentheses represent the reporting odds ratio (ROR).
4 Discussion
Our large-scale pharmacovigilance study utilized real-world FAERS data to systematically identify and characterize signals of drug-associated alopecia using disproportionality analysis. We identified 64 drugs with significant alopecia signals, revealing distinct risk profiles across therapeutic categories, demographic groups, and latency periods. By quantifying these signals at a population level, our findings expand the current understanding of suspected DIA and provide clinically actionable insights. Furthermore, they highlight specific drugs that may warrant updates to their package inserts based on accumulated real-world evidence.
In terms of patient demographics, reports were predominantly from females (76.82%), peaking in the 50–65 age range. This likely reflects both a higher prevalence of polypharmacy for chronic conditions in this group and a greater perceived impact of hair loss. These findings are consistent with established evidence that women, particularly during hormonal transition periods like perimenopause, experience higher rates of DIA (Mirmirani, 2013). Interestingly, despite higher reporting rates by women, most signaled drugs showed higher relative alopecia risk in men, suggesting potential under-reporting by males or greater biological susceptibility (Supplementary Figure S4). This paradox may be explained by heightened female vigilance regarding hair changes and a greater likelihood to seek medical advice and report events, alongside physiological factors such as higher androgen sensitivity in males potentially increasing their susceptibility (Cash et al., 1993; Trüeb, 2002). Addressing this may require raising awareness among male patients. Furthermore, significant alopecia incidence occurred in older adults (≥65 years), consistent with known elevated ADR risks from age-related pharmacokinetic changes and polypharmacy (Hadia et al., 2022; Ngcobo, 2025). In elderly patients, reduced drug distribution volumes can lead to higher drug concentrations, increasing the risk of adverse events, particularly for drugs with a narrow therapeutic index. This necessitates vigilant monitoring and potential dose adjustments in this population.
Most alopecia reports were submitted by consumers (55.88%), suggesting patients perceive this as a greater concern than healthcare providers. Regarding administration routes, oral medications accounted for the most alopecia reports, followed by subcutaneous and intravenous drugs. This pattern likely reflects the widespread use of oral drugs for chronic conditions, resulting in a large at-risk population. The notable proportion of subcutaneous reports likely stems from the high mechanistic potency of biologics and targeted therapies to disrupt hair follicles. In contrast, intravenous drugs, though potent, are used for shorter durations, leading to fewer reports. Understanding these differences is essential for clinicians to monitor the safety of treatments effectively.
Oncology drugs constituted the largest proportion of high-risk medications for alopecia (37.5%), with docetaxel exhibiting the strongest association (ROR = 70.38). This is aligned with the well-established risk of chemotherapy-induced alopecia (CIA), primarily through anagen effluvium (Trüeb, 2010). Notably, targeted agents like vismodegib and palbociclib also showed prominent signals, suggesting that non-cytotoxic mechanisms, specifically through inhibition of hedgehog signaling or CDK4/6, may similarly disrupt follicular cycling, leading to anagen effluvium. While CIA is usually reversible, (Kinoshita et al., 2019), persistent, or permanent CIA (pCIA) can occur, with taxanes like docetaxel carrying a particularly high risk (Perez et al., 2024). Although no definitive cure for pCIA exists, treatment options such as minoxidil, photobiomodulation, and platelet-rich plasma injections are being explored (Wikramanayake et al., 2023). Scalp cooling remains the primary FDA-approved preventive measure, reducing the relative risk of significant alopecia by over 40% (Rugo and Voigt, 2018; Yin et al., 2023).
Endocrine medications, such as levothyroxine and anastrozole, also featured prominently in the list of drugs associated with alopecia. Levothyroxine may cause hair thinning via effects on thyroid hormone levels, which are crucial for follicle function (van Beek et al., 2008). Notably, thyroid dysfunction itself (encompassing hypothyroidism, hyperthyroidism, and parathyroid disorders) is a well-established contributor to alopecia (Vincent and Yogiraj, 2013). Anastrozole, an aromatase inhibitor for breast cancer treatment, likely induces alopecia through estrogen depletion, potentially by shortening the anagen phase, increasing shedding (telogen effluvium), or contributing to female pattern hair loss (Karatas et al., 2016). Interestingly, apart from interferon alpha, teriflunomide, leniolisib, and teprotumumab, other immunosuppressive drugs were also identified as notable contributors in this study. This highlights the growing recognition of the potential for immunomodulatory therapies to cause alopecia. Migraine medication-calcitonin gene-related peptide (CGRP) inhibitors, such as erenumab (n = 1,287, ROR = 3.56), galcanezumab (n = 651, ROR = 4.01), and fremanezumab (n = 237, ROR = 3.00), also exhibited higher risk signals with more frequent reports. This is consistent with previous reports that CGRP inhibitors can cause alopecia ADR (Woods, 2022; Ruiz et al., 2023), suggesting that drug regulatory agencies should issue warnings and revise the drug label for this issue.
The observed bimodal distribution of TTO for DIA, with a substantial proportion of cases (40.2%) occurring within 30 days and a later peak (13.1%) manifesting at 240–360 days, strongly reflects distinct underlying pathophysiological mechanisms. The early-onset cases (≤30 days) are highly consistent with anagen effluvium. This is characterized by the acute, direct cytotoxicity of medications, most notably antineoplastic agents (e.g., docetaxel) and kinase inhibitors (e.g., sorafenib, vemurafenib). These agents rapidly disrupt the intense mitotic activity of hair matrix keratinocytes during the anagen (growth) phase, leading to abrupt hair loss within days to weeks of drug exposure (Paus et al., 2013). This pathophysiology directly supports our finding of a significantly shorter mean TTO for oncology drugs compared to non-oncology agents (P < 0.001). Conversely, the delayed-onset cases (typically ≥90 days and extending beyond 240 days) align with the pattern of telogen effluvium. This form of hair loss is associated with drugs that subtly disrupt the normal hair cycle over a prolonged period, such as endocrine medications (e.g., thyroid agents) and immunomodulatory drugs (e.g., teriflunomide). These drugs may gradually alter hormonal equilibrium or follicular cycling, precipitating a premature and synchronized transition of numerous hair follicles from anagen to the telogen (resting/shedding) phase. The resulting hair loss becomes clinically apparent only after the 2–3 months latency period of the telogen phase, explaining the observed longer TTO (Chien Yin et al., 2021). Consequently, clinical management can reference the agent-specific TTO data: Patients initiating rapid-onset drugs (median ≤15 days) require education on vigilant monitoring within the first 2 weeks for signs like increased shedding on brushes/pillows, while those on delayed-onset agents (median >100 days) need extended surveillance (≥6 months) and counseling to prevent misattribution of late hair loss to aging or stress.
A critical finding was the significant sex-based disparity in DIA latency, with a markedly shorter onset in males (108 days) versus females (236 days). This aligned with established biological differences (Trüeb, 2002; Thornton, 2013). Males exhibit higher androgen sensitivity, where elevated dihydrotestosterone levels accelerate follicular miniaturization and amplify drug-triggered anagen-to-telogen transition. Conversely, estrogen’s protective role in prolonging anagen phase may delay DIA manifestation in females. Notably, while this robust population-level trend was statistically significant only for docetaxel in our drug-specific analyses (P < 0.001), the observed discrepancy can be attributed to the combined effects of statistical power and the characteristics of data within spontaneous reporting systems. The population-level analysis benefits from a large aggregated sample, whereas drug-specific comparisons are often underpowered due to limited TTO data for individual agents, failing to reach statistical significance despite a consistent directional trend. Consequently, the finding for docetaxel confirms the general trend is detectable at the individual drug level when data are sufficient. Beyond biological mechanisms, clinical and reporting factors contribute to this disparity (Thornton, 2013). Females more frequently use chronic therapies associated with delayed alopecia (e.g., hormonal agents, antidepressants), and often seek treatment in the hair loss process, typically leading to delayed reporting. Conversely, males report earlier due to visible patterning (e.g., frontal recession) and reduced attribution bias. This underscores the need for sex-specific monitoring protocols in clinical practice. Males should maintain vigilance within 30–60 days for high-risk agents like taxanes and retinoids, while females require extended surveillance of at least 3 months for endocrine or immunomodulatory drugs.
Our analysis further revealed an apparent gap between real-world evidence and official documentation, as 23.4% of drugs with significant alopecia signals did not list alopecia as an adverse reaction in their U.S. prescribing information. For instance, CGRP inhibitors (fremanezumab, galcanezumab), increasingly prescribed for migraine prevention, demonstrated consistent signals across multiple disproportionality metrics, suggesting an under-recognized risk that warrants regulatory attention. These omissions may lead to under-diagnosis or misattribution of alopecia to other causes, delaying appropriate intervention. From a regulatory perspective, this disconnect highlights the imperative for proactive post-marketing surveillance. Agencies like the FDA and EMA should leverage real-world data to issue safety communications and update prescribing information, particularly for high-risk, widely used drugs. Moreover, encouraging healthcare providers to report suspected DIA cases is crucial to improve signal detection and risk characterization.
In clinical practice, a thorough medication history is essential for any patient presenting with unexplained alopecia. Diagnosis of DIA requires establishing a temporal relationship between drug initiation and alopecia onset, along with the exclusion of other etiologies such as nutritional deficiencies, autoimmune diseases, or genetic predisposition. For confirmed cases of DIA, the first-line approach is discontinuing the causative drug, if clinically feasible, which often leads to hair regrowth within 3–6 months (Paus and Cotsarelis, 1999; Kinoshita et al., 2019). When drug cessation is insufficient or impractical, several therapies for AGA or AA can be considered for DIA, based on clinical experience and shared pathomechanisms. For instance, topical minoxidil, FDA-approved for AGA, promotes hair growth through vasodilation, antiandrogen effects, and modulation of the hair cycle phases (Gupta et al., 2022). Oral and sublingual minoxidil are also effective off-label options (Awad et al., 2023). Similarly, 5α-reductase inhibitors, a mainstay of AGA treatment, can be used, with combination therapy of finasteride and topical minoxidil demonstrating superior efficacy to monotherapy (Chen et al., 2020). Additionally, Janus kinase (JAK) inhibitors, including baricitinib, ritlecitinib, and deuruxolitinib, have proven effective for AA by prolonging the anagen phase and promote regrowth (Kim et al., 2025). Beyond these established options, emerging modalities like small-molecule inhibitors, biologics, and stem cell-based therapies also hold promise for more precise interventions (Kim et al., 2025).
This study represents the largest and most comprehensive pharmacovigilance analysis of DIA to date. Leveraging 2 decades of FAERS data, it identifies high-risk drugs, characterizes latency periods, and reveals demographic disparities. Our findings provide clinically actionable insights for risk mitigation and patient management. However, these results must be interpreted in the context of the inherent limitations of spontaneous reporting systems. First and foremost, disproportionality analysis can identify statistical associations indicative of potential safety signals, but not definitive cause-effect relationships. Second, FAERS data are subject to substantial reporting biases, including underreporting and demographic biases (such as the overrepresentation of reports from the United States and from females, as observed in our study) (Maciá-Martínez et al., 2016). Third, the lack of a definitive denominator makes it impossible to calculate true incidence rates or directly compare risks across drugs. Fourth, inconsistent data quality and frequent absence of key clinical details limited potential confounders adjustment. Finally, while our label review was confined to United States prescribing information, cross-regional comparisons may provide valuable insights into global disparities in safety communication. Despite these limitations, our study robustly identifies potential DIA risks. It provides a clear direction for more targeted and rigorous future research.
5 Conclusion
This comprehensive pharmacovigilance study systematically identified and characterized the risk of DIA across a wide range of medications. Our study identified 64 drugs with significant alopecia signals, predominantly from oncology, endocrine, and immunomodulatory classes. Notably, we observed distinct demographic patterns, with females and older adults being disproportionately affected, highlighting the importance of considering patient-specific factors in clinical practice. The study also revealed significant disparities in the latency period of DIA between males and females, underscoring the need for sex-specific monitoring protocols. Additionally, a substantial proportion (23.4%) of these high-risk drugs lacked documentation of alopecia in their official labels, indicating a concerning gap in current drug safety information. These findings enhanced the understanding of DIA epidemiology and provide practical guidance for improved patient risk assessment and management. Furthermore, they emphasized the necessity for regulatory agencies to incorporate real-world evidence into post-marketing surveillance and label updates. Overall, this study highlighted the importance of pharmacovigilance in identifying and mitigating the risk of DIA and supports the continuous improvement of drug safety profiles to enhance patient outcomes.
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found here: https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html.
Author contributions
HL: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Writing – original draft. HW: Conceptualization, Methodology, Software, Visualization, Writing – original draft. QS: Data curation, Methodology, Software, Writing – original draft. JnC: Funding acquisition, Investigation, Methodology, Validation, Writing – review and editing. JoC: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review and editing.
Funding
The authors declare that financial support was received for the research and/or publication of this article. This research was funded by Shandong Pharmaceutical Association Research Project on Pharmacovigilance in Medical Institutions (Grant No. 2024ywjj02); Yantai Yuhuangding Hospital Scientific Research and Development Foundation (Grant No. 2023-11).
Acknowledgements
We sincerely express our gratitude to the FAERS for high-quality data for our study.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The authors declare that no Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2025.1703423/full#supplementary-material
References
Alhanshali, L., Buontempo, M., Shapiro, J., and Lo Sicco, K. (2023). Medication-induced hair loss: an update. J. Am. Acad. Dermatol 89 (2s), S20–s28. doi:10.1016/j.jaad.2023.04.022
Alkhalifah, A., Alsantali, A., Wang, E., McElwee, K. J., and Shapiro, J. (2010). Alopecia areata update: part I. Clinical picture, histopathology, and pathogenesis. J. Am. Acad. Dermatol 62 (2), 177–190. doi:10.1016/j.jaad.2009.10.032
Awad, A., Chim, I., Sharma, P., and Bhoyrul, B. (2023). Low-dose oral minoxidil improves hair density in traction alopecia. J. Am. Acad. Dermatol 89 (1), 157–159. doi:10.1016/j.jaad.2023.02.024
Candore, G., Juhlin, K., Manlik, K., Thakrar, B., Quarcoo, N., Seabroke, S., et al. (2015). Comparison of statistical signal detection methods within and across spontaneous reporting databases. Drug Saf. 38 (6), 577–587. doi:10.1007/s40264-015-0289-5
Cash, T. F., Price, V. H., and Savin, R. C. (1993). Psychological effects of androgenetic alopecia on women: comparisons with balding men and with female control subjects. J. Am. Acad. Dermatol 29 (4), 568–575. doi:10.1016/0190-9622(93)70223-g
Chen, L., Zhang, J., Wang, L., Wang, H., and Chen, B. (2020). The efficacy and safety of finasteride combined with topical minoxidil for androgenetic alopecia: a systematic review and meta-analysis. Aesthetic Plast. Surg. 44 (3), 962–970. doi:10.1007/s00266-020-01621-5
Chien Yin, G. O., Siong-See, J. L., and Wang, E. C. E. (2021). Telogen effluvium - a review of the science and current obstacles. J. Dermatol Sci. 101 (3), 156–163. doi:10.1016/j.jdermsci.2021.01.007
Devjani, S., Ezemma, O., Kelley, K. J., Stratton, E., and Senna, M. (2023). Androgenetic alopecia: therapy update. Drugs 83 (8), 701–715. doi:10.1007/s40265-023-01880-x
Ezemma, O., Devjani, S., Jothishankar, B., Kelley, K. J., and Senna, M. (2024). Drug-induced alopecia areata: a systematic review. J. Am. Acad. Dermatol 90 (1), 133–134. doi:10.1016/j.jaad.2023.05.022
Gupta, A. K., Talukder, M., Venkataraman, M., and Bamimore, M. A. (2022). Minoxidil: a comprehensive review. J. Dermatol. Treat. 33 (4), 1896–1906. doi:10.1080/09546634.2021.1945527
Gupta, A. K., Wang, T., and Economopoulos, V. (2025). Epidemiological landscape of androgenetic alopecia in the US: an all of us cross-sectional study. PLoS One 20 (2), e0319040. doi:10.1371/journal.pone.0319040
Hadia, R., Joshi, D., Bhil, D., and Maheshwari, R. (2022). Incidence of adverse drug reactions among elderly patients: a systematic review and meta-analysis. J. Sci. Soc. 49 (2), 91–102. doi:10.4103/jss.jss_50_22
Hill, R. C., Zeldin, S. D., and Lipner, S. R. (2025). Drug-induced hair loss: analysis of the food and drug administration's adverse events reporting system database. Skin. Appendage Disord. 11 (1), 63–69. doi:10.1159/000540104
Hunt, N., and McHale, S. (2005). The psychological impact of alopecia. Bmj 331 (7522), 951–953. doi:10.1136/bmj.331.7522.951
Karatas, F., Sahin, S., Sever, A. R., and Altundag, K. (2016). Management of hair loss associated with endocrine therapy in patients with breast cancer: an overview. Springerplus 5, 585. doi:10.1186/s40064-016-2216-3
Kearney, C. A., Maguire, C. A., Oza, V. S., Oh, C. S., Occidental, M. A., Shapiro, J., et al. (2025). Alopecia in children with cancer: a review from pathophysiology to management. Am. J. Clin. Dermatol 26, 747–759. doi:10.1007/s40257-025-00960-w
Kim, J., Song, S. Y., and Sung, J. H. (2025). Recent advances in drug development for hair loss. Int. J. Mol. Sci. 26 (8), 3461. doi:10.3390/ijms26083461
Kinoshita, T., Nakayama, T., Fukuma, E., Inokuchi, M., Ishiguro, H., Ogo, E., et al. (2019). Efficacy of scalp cooling in preventing and recovering from chemotherapy-induced alopecia in breast cancer patients: the HOPE study. Front. Oncol. 9, 733. doi:10.3389/fonc.2019.00733
Kinoshita, S., Hosomi, K., Yokoyama, S., and Takada, M. (2020). Time-to-onset analysis of amiodarone-associated thyroid dysfunction. J. Clin. Pharm. Ther. 45 (1), 65–71. doi:10.1111/jcpt.13024
Kinoshita-Ise, M., Fukuyama, M., and Ohyama, M. (2023). Recent advances in understanding of the etiopathogenesis, diagnosis, and management of hair loss diseases. J. Clin. Med. 12 (9), 3259. doi:10.3390/jcm12093259
Li, D., Wang, H., Qin, C., Du, D., Wang, Y., Du, Q., et al. (2024). Drug-induced acute pancreatitis: a real-world pharmacovigilance study using the FDA adverse event reporting system database. Clin. Pharmacol. Ther. 115 (3), 535–544. doi:10.1002/cpt.3139
Maciá-Martínez, M. A., de Abajo, F. J., Roberts, G., Slattery, J., Thakrar, B., and Wisniewski, A. F. (2016). An empirical approach to explore the relationship between measures of disproportionate reporting and relative risks from analytical studies. Drug Saf. 39 (1), 29–43. doi:10.1007/s40264-015-0351-3
McKenzie, P. L., Maltenfort, M., Bruckner, A. L., Gupta, D., Harfmann, K. L., Hyde, P., et al. (2022). Evaluation of the prevalence and incidence of pediatric alopecia areata using electronic health record data. JAMA Dermatol 158 (5), 547–551. doi:10.1001/jamadermatol.2022.0351
Mirmirani, P. (2013). Managing hair loss in midlife women. Maturitas 74 (2), 119–122. doi:10.1016/j.maturitas.2012.10.020
Mozzicato, P. (2007). Standardised MedDRA queries: their role in signal detection. Drug Saf. 30 (7), 617–619. doi:10.2165/00002018-200730070-00009
Mubki, T., Rudnicka, L., Olszewska, M., and Shapiro, J. (2014). Evaluation and diagnosis of the hair loss patient: part I. History and clinical examination. J. Am. Acad. Dermatol 71 (3), 415.e1–415.e15. doi:10.1016/j.jaad.2014.04.070
Muntyanu, A., Gabrielli, S., Donovan, J., Gooderham, M., Guenther, L., Hanna, S., et al. (2023). The burden of alopecia areata: a scoping review focusing on quality of life, mental health and work productivity. J. Eur. Acad. Dermatol Venereol. 37, 1490–1520. doi:10.1111/jdv.18926
Ngcobo, N. N. (2025). Influence of ageing on the pharmacodynamics and pharmacokinetics of chronically administered medicines in geriatric patients: a review. Clin. Pharmacokinet. 64 (3), 335–367. doi:10.1007/s40262-024-01466-0
Paus, R., and Cotsarelis, G. (1999). The biology of hair follicles. N. Engl. J. Med. 341 (7), 491–497. doi:10.1056/nejm199908123410706
Paus, R., Haslam, I. S., Sharov, A. A., and Botchkarev, V. A. (2013). Pathobiology of chemotherapy-induced hair loss. Lancet Oncol. 14 (2), e50–e59. doi:10.1016/s1470-2045(12)70553-3
Perez, A. M., Haberland, N. I., Miteva, M., and Wikramanayake, T. C. (2024). Chemotherapy-induced alopecia by docetaxel: prevalence, treatment and prevention. Curr. Oncol. 31 (9), 5709–5721. doi:10.3390/curroncol31090423
Piraccini, B. M., Iorizzo, M., Rech, G., and Tosti, A. (2006). Drug-induced hair disorders. Curr. Drug Saf. 1 (3), 301–305. doi:10.2174/157488606777934477
Pratt, C. H., King, L. E., Messenger, A. G., Christiano, A. M., and Sundberg, J. P. (2017). Alopecia areata. Nat. Rev. Dis. Prim. 3, 17011. doi:10.1038/nrdp.2017.11
Ravipati, A., Pradeep, T., and Tosti, A. (2024). A cross-sectional analysis of medications used by patients reporting alopecia areata on the FDA adverse events reporting system. Int. J. Dermatol 63 (4), 497–502. doi:10.1111/ijd.17014
Rugo, H. S., and Voigt, J. (2018). Scalp hypothermia for preventing alopecia during chemotherapy. A systematic review and meta-analysis of randomized controlled trials. Clin. Breast Cancer 18 (1), 19–28. doi:10.1016/j.clbc.2017.07.012
Ruiz, M., Cocores, A., Tosti, A., Goadsby, P. J., and Monteith, T. S. (2023). Alopecia as an emerging adverse event to CGRP monoclonal antibodies: cases series, evaluation of FAERS, and literature review. Cephalalgia 43 (2), 3331024221143538. doi:10.1177/03331024221143538
Sakaeda, T., Tamon, A., Kadoyama, K., and Okuno, Y. (2013). Data mining of the public version of the FDA adverse Event reporting system. Int. J. Med. Sci. 10 (7), 796–803. doi:10.7150/ijms.6048
Thornton, M. J. (2013). Estrogens and aging skin. Dermatoendocrinol 5 (2), 264–270. doi:10.4161/derm.23872
Tregunno, P. M., Fink, D. B., Fernandez-Fernandez, C., Lázaro-Bengoa, E., and Norén, G. N. (2014). Performance of probabilistic method to detect duplicate individual case safety reports. Drug Saf. 37 (4), 249–258. doi:10.1007/s40264-014-0146-y
Trüeb, R. M. (2002). Molecular mechanisms of androgenetic alopecia. Exp. Gerontol. 37 (8-9), 981–990. doi:10.1016/s0531-5565(02)00093-1
Trüeb, R. M. (2010). Chemotherapy-induced alopecia. Curr. Opin. Support Palliat. Care 4 (4), 281–284. doi:10.1097/SPC.0b013e3283409280
van Beek, N., Bodó, E., Kromminga, A., Gáspár, E., Meyer, K., Zmijewski, M. A., et al. (2008). Thyroid hormones directly alter human hair follicle functions: anagen prolongation and stimulation of both hair matrix keratinocyte proliferation and hair pigmentation. J. Clin. Endocrinol. Metab. 93 (11), 4381–4388. doi:10.1210/jc.2008-0283
Vary, J. C. (2015). Selected disorders of skin Appendages--Acne, alopecia, hyperhidrosis. Med. Clin. North Am. 99 (6), 1195–1211. doi:10.1016/j.mcna.2015.07.003
Vincent, M., and Yogiraj, K. (2013). A descriptive study of alopecia patterns and their relation to thyroid dysfunction. Int. J. Trichology 5 (1), 57–60. doi:10.4103/0974-7753.114701
Wikramanayake, T. C., Haberland, N. I., Akhundlu, A., Laboy Nieves, A., and Miteva, M. (2023). Prevention and treatment of chemotherapy-induced alopecia: what is available and what is coming? Curr. Oncol. 30 (4), 3609–3626. doi:10.3390/curroncol30040275
Woods, R. H. (2022). Alopecia signals associated with calcitonin gene-related peptide inhibitors in the treatment or prophylaxis of migraine: a pharmacovigilance study. Pharmacotherapy 42 (10), 758–767. doi:10.1002/phar.2725
Wu, S. N., Chen, X. D., Yan, D., Wang, Y. Q., Wang, S. P., Guan, W. Y., et al. (2025). Drug-associated glaucoma: a real-world study based on the food and drug administration adverse event reporting system database. Clin. Exp. Ophthalmol. 53 (2), 140–160. doi:10.1111/ceo.14454
Yan, Y. D., Zhao, Y., Zhang, C., Fu, J., Su, Y. J., Cui, X. L., et al. (2022). Toxicity spectrum of immunotherapy in advanced lung cancer: a safety analysis from clinical trials and a pharmacovigilance system. EClinicalMedicine 50, 101535. doi:10.1016/j.eclinm.2022.101535
Keywords: drug-induced alopecia, pharmacovigilance, FAERS, disproportionality analysis, time-to-onset, risk assessment
Citation: Li H, Wei H, Shentu Q, Cui J and Chen J (2025) Real-world pharmacovigilance insights into drug-induced risk of alopecia. Front. Pharmacol. 16:1703423. doi: 10.3389/fphar.2025.1703423
Received: 11 September 2025; Accepted: 17 November 2025;
Published: 28 November 2025.
Edited by:
Zhao-you Meng, Xinqiao Hospital, ChinaReviewed by:
Mokshal Porwal, Allegheny Health Network, United StatesSamah Alfahl, Taibah University, Saudi Arabia
Copyright © 2025 Li, Wei, Shentu, Cui and Chen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Jianxin Cui, Y3VpamlhbnhpbnFoQHFkdS5lZHUuY29t; Jiaojiao Chen, Y2pqOTMwNjA2QDE2My5jb20=
†These authors have contributed equally to this work and share first authorship