Sex-specific adverse event profiles of PDE4 inhibitors: a comparative big-data pharmacovigilance study of apremilast, crisaborole, and roflumilast in FAERS (2004–2025)

Abstract

Introduction:

Phosphodiesterase-4 (PDE4) inhibitors, including apremilast, crisaborole and roflumilast, are widely prescribed for chronic inflammatory diseases. However, sex-specific safety profiles of these agents in routine clinical practice remain poorly characterized.

Methods:

We conducted a retrospective pharmacovigilance study using the US Food and Drug Administration Adverse Event Reporting System (FAERS) from Q1-2004 to Q1-2025. Data cleaning and deduplication were performed following the FDA Guidance for Industry: Pharmacovigilance Practices and Pharmacoepidemiologic Assessment (FDA, 2005). Specifically, deduplication was executed by retaining only the most recent FDA_DT for each unique CASEID, and demographic (DEMO), drug (DRUG), and reaction (REAC) files were merged. Reports listing apremilast, crisaborole, or roflumilast as primary suspect drugs formed three target cohorts. Disproportionality analyses were performed using reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN, information component), and multi-item gamma Poisson shrinker (MGPS, empirical Bayes geometric mean). Positive signals required meeting all four algorithmic thresholds. Signals were summarised at Preferred Term (PT) and System Organ Class (SOC) levels, stratified by sex, and visualised with volcano plots, forest plots, and SOC-level heatmaps. Time-to-onset (TTO) was calculated as the interval between treatment start and event onset.

Results:

A total of 127,516 Apremilast, 7,562 Crisaborole, and 2,142 Roflumilast reports were included. Clear sex-specific disproportionality patterns emerged across the three agents. For Apremilast, females exhibited higher reporting of dizziness, palpitations, and infection-related adverse events, whereas Crisaborole showed a modest male predominance for “product use issue,” and Roflumilast demonstrated male-skewed signals, particularly for malignancy- and metabolism-related events. System Organ Class (SOC) analyses further revealed distinct organ-system involvement for each agent. Time-to-onset profiles showed substantially delayed onset for Apremilast compared with the more immediate onset observed for Crisaborole and Roflumilast. The median TTO (IQR) was 24 (7–86) days for Apremilast, 4 (1–15) days for Crisaborole, and 42 (14–128) days for Roflumilast.

Discussion/Conclusion:

This large-scale real-world analysis reveals pronounced sex-specific and drug-specific heterogeneity in the safety profiles of PDE4 inhibitors. These findings highlight the need for sex-stratified risk communication, individualized monitoring strategies, and further mechanistic investigations.

1 Introduction

Phosphodiesterase-4 (PDE4) inhibitors constitute an important therapeutic class targeting cyclic adenosine monophosphate (cAMP) regulation and downstream inflammatory pathways. Currently approved agents—including Apremilast (oral), Crisaborole (topical), and Roflumilast (oral or topical formulations)—are widely used for psoriasis, psoriatic arthritis, atopic dermatitis, and chronic obstructive pulmonary disease (COPD), among other inflammatory and immune-mediated conditions (Kumar et al., 2013; Li et al., 2018). Although randomized clinical trials have established favorable efficacy and acceptable short-term tolerability profiles for these drugs, accumulating post-marketing evidence suggests a broader spectrum of adverse events (AEs) than previously recognized, with variations across organ systems and specific patient populations (Crocetti et al., 2022; Fan et al., 2024). Given the expanding real-world utilization of PDE4 inhibitors globally, a systematic characterization of their comparative safety profiles is urgently needed.

Drug safety assessment in clinical trials is inherently limited by restricted sample sizes, short follow-up periods, selected populations, and exclusion of patients with comorbidities or polypharmacy (Li et al., 2023; Linger and Martin, 2018). By contrast, the U.S. Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) offers a valuable resource for detecting rare, unexpected, or long-term AEs at a population scale (Li et al., 2025). FAERS-based disproportionality analyses have been widely applied to identify pharmacovigilance signals across dermatologic, immunologic, and respiratory therapeutics (Lei et al., 2025; Zhao et al., 2024; Que et al., 2025). However, despite the growing number of studies evaluating individual PDE4 inhibitors, no comprehensive class-level, multi-algorithm, real-world comparison of Apremilast, Crisaborole, and Roflumilast has been conducted. Likewise, existing literature has not examined organ-system patterns (SOC-level), onset timing, or cross-drug heterogeneity in a unified framework.

In addition to class-level safety, sex-specific differences in drug response and toxicity have increasingly attracted attention. Biological and immunological differences between males and females-including hormone-regulated signaling, body composition, metabolism, microbiome diversity, and immune responsiveness-are known to affect drug pharmacokinetics and pharmacodynamics (Zucker and Prendergast, 2020) (Dalla and Kokras, 2025). Regulatory agencies, including the FDA and NIH, have called for systematic integration of sex as a biological variable in safety research and risk communication (Evaluation of Sex Differences in, 2013) (NIH Policy on Sex as, 2015). Nevertheless, sex-stratified safety evidence for PDE4 inhibitors remains extremely limited. Only sporadic post-marketing observations have suggested that females may report more gastrointestinal or neuropsychiatric symptoms with Apremilast, whereas males may have distinct AE patterns with Roflumilast—yet these findings lack robust validation from large-scale real-world datasets (Shan et al., 2023). Moreover, no prior study has simultaneously evaluated sex-differential disproportionality, volcano-plot visualization, forest-plot quantification, and class-wide comparison across all three PDE4 inhibitors.

Given these major evidence gaps, a systematic investigation integrating real-world pharmacovigilance, multi-dimensional signal analytics, and sex-specific risk assessment is critically needed. In this study, we leveraged the FAERS database (Yan et al., 2025), comprising over 19million deduplicated reports, to perform a comprehensive evaluation of Apremilast, Crisaborole, and Roflumilast. Using four complementary disproportionality algorithms—reporting odds ratio (ROR), proportional reporting ratio (PRR), Bayesian confidence propagation neural network (BCPNN), and multi-item Gamma–Poisson Shrinker (MGPS) (Bate, 2007)—we profiled drug-specific and sex-specific AE signals across preferred term (PT) and system organ class (SOC) levels (Kong et al., 2025). Additionally, we examined time-to-onset (TTO) differences to explore early vs. delayed AE patterns among the three agents, in line with growing interest in temporal pharmacovigilance analyses (Liu et al., 2024; Zheng et al., 2025).

This work provides the first class-wide, sex-stratified, multi-algorithm pharmacovigilance analysis of PDE4 inhibitors in real-world practice, uncovering novel risk patterns and heterogeneity that may inform patient counseling, individualized monitoring, and safer prescribing strategies.

2 Methods

2.1 Data source and study design

This retrospective pharmacovigilance study used publicly available data from the FAERS. All quarterly files from Q1 2004 to Q1 2025 were downloaded, including the DEMO, DRUG, REAC, and THER (Therapy) datasets to ensure complete information for time-to-onset calculation. The FDA Adverse Event Reporting System (FAERS) consists of large-scale, spontaneously reported Individual Case Safety Reports (ICSRs). While the database follows a structured format—including demographic, drug, and reaction fields—the clinical narratives and reporting behaviors are inherently heterogeneous. Reports are coded using the Medical Dictionary for Regulatory Activities (MedDRA). The study followed FDA recommendations for FAERS data processing and reporting, and adhered to established pharmacoepidemiology practices for disproportionality analysis. As FAERS contains anonymized public data, institutional review board approval was not required.

2.2 Deduplication of FAERS reports

Deduplication was performed according to FDA recommendations to ensure that each case was counted once. Reports were grouped by CASEID, and for each group, the entry with the most recent FDA_DT and highest PRIMARYID was retained. Demographic inconsistencies and biologically implausible values were excluded. After deduplication, 19,026,509 unique ICSRs were included for analysis.

2.3 Adverse event processing

AEs were mapped to MedDRA PTs and SOC categories. Reports lacking AE information were excluded; those missing demographic fields were retained for overall analyses but excluded from sex-stratified evaluations.

2.4 Disproportionality analysis

A case–non-case framework was applied using two-by-two contingency tables (Table 1). Four major signal-detection algorithms were used: ROR, PRR, BCPNN, and MGPS. The formulas, output metrics, and positive-signal criteria for each algorithm are summarized in Table 2. A signal was deemed positive when all thresholds were met: ROR lower 95% CI > 1; PRR ≥2 with χ² ≥ 4; IC025 > 0; and EBGM05 > 1.

2.5 Sex-specific disproportionality

Sex-stratified disproportionality was calculated by estimating ROR separately for males and females. Sex-differential risk was evaluated using log₂(ROR_male/ROR_female), where positive values reflect male-predominant risk and negative values reflect female-predominant risk. Volcano plots were generated using log₂ ratios and FDR-adjusted p-values. To quantify the magnitude of sex differences, the Ratio of Reporting Odds Ratios () was calculated: Statistical significance for the was determined using a Z-test on the log-transformed RORs, followed by False Discovery Rate (FDR) correction via the Benjamini–Hochberg method.

2.6 System-level AE patterns

To evaluate cross-drug organ-system heterogeneity, significant PT-level signals were aggregated to SOC categories, and SOC-level RORs were calculated by identifying unique caseIDs within each SOC category. This case-level aggregation ensures that each individual report is counted only once per SOC, even if multiple PTs were reported, thereby avoiding the bias of double-counting. Heatmaps were constructed to visualize SOC-level patterns for each PDE4 inhibitor.

2.7 Time-to-onset analysis

Time-to-onset (TTO) was calculated as the interval between treatment start (START_DT, extracted from the THER or DRUG files) and adverse event onset (EVENT_DT). Data cleaning was performed to exclude reports where: (i) EVENT_DT occurred before START_DT; (ii) dates were incomplete (missing month or day); or (iii) TTO exceeded 20 years, which was considered biologically implausible.

2.8 Statistical software

All analyses were performed using R (version 4.4.2) with openEBGM and related packages. False discovery rate (FDR) correction used the Benjamini–Hochberg method, with significance defined as p < 0.05.

3 Results

Across the deduplicated FAERS dataset (2004–2025Q1), a total of 137,220 adverse event (AE) reports were included for Apremilast (N = 127,516), Crisaborole (N = 7,562), and Roflumilast (N = 2,142). Baseline demographic, temporal, and indication distributions displayed substantial heterogeneity across the three agents, reflecting their distinct routes of administration, approved populations, and real-world utilization patterns (Table 3). The overall data extraction and analytic workflow is summarized in Figure 1.

FIGURE 1

3.1 Overall AE distributions, sex composition, and baseline characteristics

Apremilast users demonstrated the broadest demographic spread, with a majority of female reporters (62.6%) and a long usage history dating back to 2014. Crisaborole reports were predominantly pediatric and adolescent, consistent with its topical dermatologic indications, while Roflumilast users were mainly older males, consistent with its COPD indication. These characteristic differences provide clinical context for downstream AE divergence.

3.2 Sex-differential AE patterns and drug-level contrasts

When sex-stratified disproportionality signals were examined across PT levels using and Z-tests, clear, non-random patterns emerged (Figures 2, 3). Rather than isolated outliers, the signals displayed structured gradients that aligned tightly with each agent’s pharmacologic properties and patient population. Across the class, three consistent themes were observed:

FIGURE 2

FIGURE 3

3.2.1 Magnitude gradient of sex influence across the class

Apremilast demonstrated a mixed but distinctly female-skewed tolerability profile, with higher female RORs for nausea, diarrhoea and headache. In contrast, Roflumilast showed the strongest male-biased AE elevation, with substantially higher male risks for death, psychiatric symptoms, respiratory exacerbations, metabolic changes, and gastrointestinal AEs. Crisaborole displayed minimal sex-differential amplification, consistent with negligible systemic absorption.

3.2.2 Bidirectional sex imbalance within apremilast

A “dual-direction” structure was evident: males showed elevated RORs for severe or systemic events (death, COPD-like symptoms, suicidal ideation), while females predominated in mucosal, autonomic, and neurologic tolerability events. This pattern reinforces that sex differences reflect mechanistic divergence rather than reporting noise.

3.2.3 Consistent male-dominant amplification with roflumilast

Male users exhibited markedly higher RORs across psychiatric, metabolic, respiratory, and systemic PT categories, forming the steepest sex gradient among the three agents. Even within the same PT category, male RORs frequently exceeded female estimates by large margins, reflecting heightened susceptibility.

Collectively, these observations illustrate that patient sex is associated with significant variations in drug-specific AE signatures in structured and clinically coherent patterns, rather than acting as a random confounder.

3.3 SOC-level clustering and organ-system integration

At the organ-system level, each PDE4 inhibitor demonstrated a unique SOC signature (Figure 4). To ensure methodological rigor, SOC-level RORs were calculated based on unique Case IDs rather than PT counts, effectively eliminating potential double-counting bias. Apremilast (N = 127,516) displayed strong enrichment in Gastrointestinal, Nervous System, and General Disorders. Roflumilast (N = 2,142) showed clustering in respiratory, psychiatric, and metabolic SOCs; and Crisaborole was almost exclusively represented within skin and subcutaneous tissue disorders. Importantly, several SOC-level differences were themselves sex-dependent. Females showed stronger neurological and mucosal SOC signals with Apremilast, whereas males showed elevated psychiatric, metabolic, and respiratory SOC signals for Roflumilast. These findings suggest that sex interacts not only with specific symptom types but with broader physiologic and organ-system domains.

FIGURE 4

3.4 Time-to-onset (TTO) divergence and sex-associated temporal patterns

Marked differences were observed in TTO distributions across the three agents (Figure 5).The median TTO (IQR) was 24 (7–86) days for Apremilast, 4 (1–15) days for Crisaborole, and 42 (14–128) days for Roflumilast. Crisaborole produced rapid-onset events, typically within days, consistent with its topical mechanism. Apremilast demonstrated a broader TTO profile, with gastrointestinal events peaking during the first month of titration. Roflumilast showed a biphasic profile: rapid onset for psychiatric AEs, followed by later-emerging metabolic events. Sex differences were also reflected in onset timing; for instance, females exhibited earlier autonomic and gastrointestinal onset with Apremilast, aligning with the female-biased signal intensity observed in the sex-stratified analysis.

FIGURE 5

3.5 Validation of signals against clinical benchmarks

To validate the reliability of our findings, we compared FAERS-derived signals with FDA product labels and pivotal Phase III trial data (Supplementary Table S6). (Papp et al., 2015; Paller et al., 2016; Gyldenløve et al., 2024) (Administration USFaD, 2011; Administration USFaD, 2014; Administration USFaD, 2016). Common adverse events demonstrated both high RORs and high clinical incidences (e.g., diarrhea for Apremilast), confirming the sensitivity of our model. Crucially, our study also identified signals with high RORs but very low incidence rates in clinical trials (e.g., psychiatric events for Roflumilast and severe mucosal reactions for Apremilast). In pharmacovigilance, these low-incidence/high-ROR signals are of paramount significance as they may represent rare but serious risks that were potentially underpowered in randomized controlled trials, thereby justifying the need for continuous post-marketing surveillance.

3.6 Integrated cross-drug, cross-sex AE architecture

Taken together, these multidimensional analyses reveal a coherent class-level architecture in which: (1) pharmacology and route of administration shape baseline risk; (2) patient sex is associated with variations in AE likelihood, reporting severity, and temporal onset; (3) drug identity determines organ-system vulnerability. These findings underscore that PDE4 inhibitors do not share a uniform safety profile but exhibit structured, sex-influenced, drug-specific AE patterns detectable only through large-scale pharmacovigilance interrogation.

4 Discussion

This large-scale real-world pharmacovigilance study demonstrates that PDE4 inhibitors share a common anti-inflammatory mechanism shaped by both patient sex and drug-specific pharmacological properties (e.g., systemic absorption and route of administration). Using more than 19 million deduplicated FAERS reports, we identified structured, non-random patterns in sex-differential disproportionality, leading adverse events, organ-system involvement, and time-to-onset patterns. Collectively, these findings highlight that PDE4 inhibitors cannot be viewed as a homogeneous therapeutic class from a safety perspective, and that patient sex acts as a significant biological determinant of adverse event susceptibility within this class (Ibrahim et al., 2021).

A central observation was the gradient of sex influence across the PDE4 inhibitor class. The magnitude of sex-differential reporting followed a consistent pattern, with Roflumilast showing the strongest male-predominant risks, Apremilast demonstrating a female-skewed tolerability profile, and Crisaborole exhibiting minimal sex-specific divergence. This gradient aligns with their differing pharmacokinetic and pharmacodynamic characteristics (Aljohmani and Yildiz, 2025). Roflumilast’s systemic exposure, hepatic metabolism, and catabolic effects may create vulnerability in males, who have distinct inflammatory responses, metabolic profiles, and COPD-related comorbid burden (Klein and Flanagan, 2016) (Kawamatawong, 2021) (Buhl et al., 2012). In contrast, Apremilast’s gastrointestinal and neurologic tolerability profile aligns with known sex differences in autonomic sensitivity, nociception, and immune-mediated reactions, which are heightened among females (Waldman et al., 2007) (Kavanaugh et al., 2019). Crisaborole’s lack of systemic absorption likely explains its flat sex-specific landscape, underscoring how route of administration and systemic exposure govern sex-related risk differences (Dalla and Kokras, 2025) (Silverberg et al., 2024).

The identification of sex-differential clusters of adverse events further suggests that the patterns observed in this study are unlikely to be attributable solely to random noise (Sportiello and Capuano, 2024) (de Vries et al., 2019). Female-predominant neurologic and cardiovascular AEs with Apremilast may be partly influenced by known sex-related differences in autonomic reactivity, β-adrenergic responsiveness, serotonin pathways, and vascular regulation, although these mechanisms have not been specifically validated in the context of PDE4 inhibition (Liccardo et al., 2022) (Ruschil et al., 2021) (Castle and Flanigan, 2024). Conversely, the male-skewed psychiatric and metabolic signals associated with Roflumilast may reflect interactions between the drug’s systemic pharmacology and sex-related differences in dopaminergic modulation, inflammatory signaling, and the greater systemic inflammatory and metabolic burden often observed among male patients with COPD (Cahill, 2006) (Milne et al., 2024). These interpretations remain hypothesis-generating rather than confirmatory, but they are directionally consistent with broader evidence regarding sex-specific susceptibilities to neurologic, cardiovascular, psychiatric, and metabolic adverse drug reactions in real-world populations (Shan et al., 2023). The consistency of these sex-differential signals across multiple disproportionality algorithms further supports their robustness at the signal-detection level, while acknowledging that disproportionality metrics cannot establish causal relationships (Gravel et al., 2024).

At the organ-system level, the three agents demonstrated highly distinctive SOC-level clustering, suggesting that each PDE4 inhibitor engages different physiologic pathways or interacts with distinct comorbidity profiles. Apremilast’s gastrointestinal and neurologic clusters reflect known PDE4-related cAMP elevations affecting emesis pathways and central neurotransmission (Zhao et al., 2013). Roflumilast’s enrichment in respiratory, psychiatric, and metabolic SOCs aligns with its indication in COPD, its known propensity for weight loss and systemic inflammation, and reports of hepatic and metabolic adverse reactions in real-world use (Fabbri and Rabe, 2007) (Gyldenløve et al., 2023). Crisaborole’s concentration in cutaneous SOCs mirrors its topical route of administration and demonstrated low systemic absorption in healthy-volunteer pharmacokinetic studies (Cao et al., 2024). Importantly, several SOC-level differences appear to be sex-dependent, consistent with broader pharmacovigilance evidence showing that sex influences adverse drug reaction profiles across organ systems (Lacroix et al., 2022) (de Vries et al., 2019).

The time-to-onset differences observed among the three PDE4 inhibitors reflect distinct “failure-type” patterns. The sharp, early-onset peak for Crisaborole aligns with an early-failure pattern ( < 1), characteristic of its topical application. Conversely, the delayed distribution for Apremilast and the biphasic profile of Roflumilast suggest systemic accumulation or adaptation-related mechanisms, potentially linked to their oral route and broader organ-system engagement. For Apremilast, gastrointestinal and neurologic adverse events typically emerge over the first several weeks of treatment in clinical trials, consistent with our finding of a delayed time-to-onset distribution (Paul et al., 2015) (Mease et al., 2023). In contrast, Crisaborole—administered topically with minimal systemic exposure—demonstrates application-site burning or stinging that often occurs shortly after treatment initiation, aligning with the early-onset peak observed in our dataset (Eichenfield et al., 2017) (Jarnagin et al., 2016). For Roflumilast, our data revealed a biphasic pattern in which gastrointestinal and neuropsychiatric events tend to occur early in therapy, whereas metabolic events such as progressive weight loss accumulate over longer durations. This temporal dissociation corresponds with known safety profiles from COPD trials, where early diarrhea and insomnia contrast with longer-term systemic metabolic effects (Au Duong et al., 2005). Together, these time-to-onset signatures offer clinically relevant clues, as early versus delayed patterns are frequently used to help differentiate drug-related from disease-related events in pharmacovigilance assessments.

Collectively, these integrated findings support a conceptual model wherein the safety profile of each PDE4 inhibitor is the combined product of its pharmacologic architecture, route of administration, target population, and sex-dependent biological responses (Younes, 2024). The results emphasize that sex is not an incidental variable but a determinant of real-world drug safety, consistent with pharmacovigilance evidence that females and males exhibit systematically different ADR patterns (Shan et al., 2023). This is in line with growing evidence that females and males differ in immune function, hormonal regulation, metabolism, microbiome ecology, and nociceptive pathways—factors that can directly influence adverse drug reactions (Sharma et al., 2025) (Massey et al., 2020) (Shobeiri et al., 2022). Our study extends this knowledge into the PDE4 inhibitor class and provides class-level clarity using population-scale pharmacovigilance data.

The study also offers practical clinical implications. First, clinicians should anticipate gastrointestinal and neurologic tolerability issues with Apremilast—particularly in subgroups already known to be vulnerable to such events—and may consider slower titration or more proactive counseling (Mease et al., 2020). Second, given Roflumilast’s well-recognized gastrointestinal, psychiatric, and metabolic adverse event profile, together with our observation of male-biased disproportionality signals, closer monitoring of male patients—including early symptom screening for mood changes and weight loss—may help reduce AE burden (Yu et al., 2013) (Cilli et al., 2019). Third, Crisaborole’s predominantly cutaneous and generally mild adverse events, combined with low systemic absorption and reassuring long-term safety across age groups, suggest relatively limited concern regarding sex-specific risk and support its use across diverse patient populations (Paller et al., 2016) (Zane et al., 2016).

Finally, several limitations should be acknowledged. First, FAERS is a spontaneous reporting system subject to underreporting, reporting bias, and a lack of denominator data. Second, the distinct indications and demographic structures—such as older male COPD patients for Roflumilast versus female psoriasis patients for Apremilast—may act as important confounders for the observed sex-biased patterns. Third, while median and interquartile ranges provided a macro-level overview of temporal AE onset, future studies employing Weibull shape parameter () analysis are warranted to determine the specific failure types (e.g., early-type vs. wear-out type) for individual drug-AE combinations. Despite these constraints, the large sample size and multi-algorithm approach provide a robust signal-detection framework.

5 Conclusion

This large-scale real-world evaluation demonstrates that PDE4 inhibitors, despite sharing a common pharmacologic mechanism, exhibit distinct and sex-dependent safety profiles. Using comprehensive FAERS data and multiple complementary disproportionality algorithms, we identified consistent sex-differential AE patterns, drug-specific organ-system clustering, and divergent temporal onset characteristics across Apremilast, Crisaborole, and Roflumilast. These findings reveal that sex is a major modifier of PDE4 inhibitor tolerability and should be incorporated into risk communication and individualized monitoring strategies. Clinically, the results support heightened awareness of female-predominant neurologic and mucosal events with Apremilast, increased vigilance for male-dominant psychiatric and metabolic events with Roflumilast, and confidence in the relatively low sex-related divergence for Crisaborole. Overall, this work provides a class-level framework for understanding PDE4 inhibitor safety and underscores the value of large real-world pharmacovigilance datasets in informing precision medicine.

References

• 1

  Administration USFaD (2014). Otezla (apremilast) tablets prescribing information. Silver Spring, MD: U.S. Food and Drug Administration.

  • Google Scholar
  • View reference in article

• 2

  Administration USFaD (2011). Daliresp (roflumilast) tablets prescribing information. Silver Spring, MD: U.S. Food and Drug Administration.

  • Google Scholar
  • View reference in article

• 3

  Administration USFaD (2016). Eucrisa (crisaborole) ointment prescribing information. Silver Spring, MD: U.S. Food and Drug Administration.

  • Google Scholar
  • View reference in article

• 4

  AljohmaniA.YildizD. (2025). Biological sex differences in pharmacokinetics and adverse drug reactions. Naunyn Schmiedeb. Arch. Pharmacol. 399 (3), 3285–3301. 10.1007/s00210-025-04721-8

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  Au DuongM.-V.AudoinB.BoulanouarK.IbarrolaD.MalikovaI.Confort-GounyS.et al (2005). Altered functional connectivity related to white matter changes inside the working memory network at the very early stage of MS. J. Cereb. Blood Flow. Metab.25 (10), 1245–1253. 10.1038/sj.jcbfm.9600122

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  BateA. (2007). Bayesian confidence propagation neural network. Drug Saf.30 (7), 623–625. 10.2165/00002018-200730070-00011

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  BuhlR.KunaP.PetersM. J.AnderssonT. L. G.NayaI. P.PetersonS.et al (2012). The effect of budesonide/formoterol maintenance and reliever therapy on the risk of severe asthma exacerbations following episodes of high reliever use: an exploratory analysis of two randomised, controlled studies with comparisons to standard therapy. Respir. Res.13 (1), 59. 10.1186/1465-9921-13-59

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  CahillL. (2006). Why sex matters for neuroscience. Nat. Rev. Neurosci.7 (6), 477–484. 10.1038/nrn1909

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  CaoX.-C.LuJ.-W.FengY.-F.SongL.-B.LuY. (2024). Efficacy and safety of Ruxolitinib, Crisaborole, and Tapinarof for mild-to-moderate atopic dermatitis: a Bayesian network analysis of RCTs. Naunyn Schmiedeb. Arch. Pharmacol.397 (7), 4657–4662. 10.1007/s00210-024-02971-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  CastleM. E.FlaniganM. E. (2024). The role of brain serotonin signaling in excessive alcohol consumption and withdrawal: a call for more research in females. Neurobiol. Stress30, 100618. 10.1016/j.ynstr.2024.100618

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  CilliA.BalH.GunenH. (2019). Efficacy and safety profile of roflumilast in a real-world experience. J. Thorac. Dis.11 (4), 1100–1105. 10.21037/jtd.2019.04.49

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  CrocettiL.FlorestaG.CilibrizziA.GiovannoniM. P. (2022). An overview of PDE4 inhibitors in clinical trials: 2010 to early 2022. Molecules27 (15), 4964. 10.3390/molecules27154964

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  DallaC.KokrasN. (2025). Pharmacokinetics and pharmacodynamics: making clinical sense of sex differences in psychopharmacology. Expert Opin. Drug Metab. Toxicol. 10.1080/17425255.2025.2597341

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  de VriesS. T.DenigP.EkhartC.BurgersJ. S.KleefstraN.MolP. G. M.et al (2019). Sex differences in adverse drug reactions reported to the National pharmacovigilance centre in the Netherlands: an explorative observational study. Br. J. Clin. Pharmacol.85 (7), 1507–1515. 10.1111/bcp.13923

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  EichenfieldL. F.CallR. S.ForshaD. W.FowlerJ.HebertA. A.SpellmanM.et al (2017). Long-term safety of crisaborole ointment 2% in children and adults with mild to moderate atopic dermatitis. J. Am. Acad. Dermatol77 (4), 641–649. 10.1016/j.jaad.2017.06.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  Evaluation of sex differences in clinical investigations silver spring, MD: U.S. Food and Drug Administration (2013). Silver Spring, MD: U.S. Food and Drug Administration. Available online at: https://www.fda.gov/regulatory-information/search-fda-guidance-documents/evaluation-sex-differences-clinical-investigations (Accessed November 15, 2025).

  • Google Scholar
  • View reference in article

• 17

  FabbriL. M.RabeK. F. (2007). From COPD to chronic systemic inflammatory syndrome?Lancet.370 (9589), 797–799. 10.1016/S0140-6736(07)61383-X

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  FanT.WangW.WangY.ZengM.LiuY.ZhuS.et al (2024). PDE4 inhibitors: potential protective effects in inflammation and vascular diseases. Front. Pharmacol., 15–2024. 10.3389/fphar.2024.1407871

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  GravelC. A.BaiW.DourosA. (2024). Comparators in pharmacovigilance: a quasi-quantification bias analysis. Drug Saf.47 (8), 809–819. 10.1007/s40264-024-01433-5

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  GyldenløveM.MeteranH.SørensenJ. A.FageS.YaoY.LindhardsenJ.et al (2023). Efficacy and safety of oral roflumilast for moderate-to-severe psoriasis-a randomized controlled trial (PSORRO). Lancet Reg. Health Eur.30, 100639. 10.1016/j.lanepe.2023.100639

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  GyldenløveM.SørensenJ. A.FageS.MeteranH.SkovL.ZachariaeC.et al (2024). Effects of oral roflumilast therapy on body weight and cardiometabolic parameters in patients with psoriasis - results from a randomized controlled trial (PSORRO). J. Am. Acad. Dermatol91 (1), 64–71. 10.1016/j.jaad.2024.02.036

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  IbrahimH.AbdoA.El KerdawyA. M.EldinA. S. (2021). Signal detection in pharmacovigilance: a review of informatics-driven approaches for the discovery of drug-drug interaction signals in different data sources. Artif. Intell. Life Sci.1, 100005. 10.1016/j.ailsci.2021.100005

  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  JarnaginK.ChandaS.CoronadoD.CiaravinoV.ZaneL. T.Guttman-YasskyE.et al (2016). Crisaborole topical ointment, 2%: a nonsteroidal, topical, anti-inflammatory phosphodiesterase 4 inhibitor in clinical development for the treatment of atopic dermatitis. J. Drugs Dermatol15 (4), 390–396.

  • Pubmed Abstract
  • Google Scholar
  • View reference in article

• 24

  KavanaughA.GladmanD. D.EdwardsC. J.SchettG.GueretteB.DelevN.et al (2019). Long-term experience with apremilast in patients with psoriatic arthritis: 5-year results from a PALACE 1-3 pooled analysis. Arthritis Res. Ther.21 (1), 118. 10.1186/s13075-019-1901-3

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  KawamatawongT. (2021). Phosphodiesterase-4 inhibitors for Non-COPD respiratory diseases. Front. Pharmacol.12, 518345. 10.3389/fphar.2021.518345

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  KleinS. L.FlanaganK. L. (2016). Sex differences in immune responses. Nat. Rev. Immunol.16 (10), 626–638. 10.1038/nri.2016.90

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  KongL.YangX.XuJ. (2025). Comparison of safety of lecanemab and aducanumab: a real-world disproportionality analysis using the FDA adverse event reporting system. Front. Pharmacol.16, 1593989. 10.3389/fphar.2025.1593989

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  KumarN.GoldminzA. M.KimN.GottliebA. B. (2013). Phosphodiesterase 4-targeted treatments for autoimmune diseases. BMC Med.11, 96. 10.1186/1741-7015-11-96

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  LacroixC.MaurierA.LargeauB.DestereA.ThillardE.-M.DriciM.et al (2022). Sex differences in adverse drug reactions: are women more impacted?Therapie78 (2), 175–188. 10.1016/j.therap.2022.10.002

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  LeiL.WangL.ShenS.ZhaoG.ZhaoX.DongH. (2025). Pharmacovigilance study of the frequency of gastrointestinal ulceration reports associated with immune checkpoint inhibitors: insights from the FDA adverse event reporting system. Front. Pharmacol.16, 1682259. 10.3389/fphar.2025.1682259

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  LiH.ZuoJ.TangW. (2018). Phosphodiesterase-4 inhibitors for the treatment of inflammatory diseases. Front. Pharmacol.9, 1048. 10.3389/fphar.2018.01048

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  LiY.WuY.JiangT.XingH.XuJ.LiC.et al (2023). Opportunities and challenges of pharmacovigilance in special populations: a narrative review of the literature. Ther. Adv. Drug Saf.14, 20420986231200746. 10.1177/20420986231200746

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  LiJ.HanZ. W.GaoS. S.GuoQ.CangH. Q. (2025). Risk of pulmonary fungal infections associated with biologics: a FAERS database disproportionality analysis. Front. Immunol.16, 1672343. 10.3389/fimmu.2025.1672343

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 34

  LiccardoD.ArosioB.CorbiG.CannavoA. (2022). Sex/Gender- and age-related differences in β-Adrenergic receptor signaling in cardiovascular diseases. J. Clin. Med.11 (15), 4280. 10.3390/jcm11154280

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 35

  LingerM.MartinJ. (2018). Pharmacovigilance and expedited drug approvals. Aust. Prescr.41 (2), 50–53. 10.18773/austprescr.2018.010

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 36

  LiuQ.CuiZ.DengC.YangC.ShiT. (2024). A real-world pharmacovigilance analysis of adverse events associated with irbesartan using the FAERS and JADER databases. Front. Pharmacol.15, 1485190. 10.3389/fphar.2024.1485190

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 37

  MasseyS. C.WhitmireP.DoyleT. E.IppolitoJ. E.MrugalaM. M.HuL. S.et al (2020). Sex differences in health and disease: a review of biological sex differences relevant to cancer with a spotlight on glioma. Cancer Lett.498, 178–187. 10.1016/j.canlet.2020.07.030

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 38

  MeaseP. J.GladmanD. D.Gomez-ReinoJ. J.HallS.KavanaughA.LespessaillesE.et al (2020). Long-Term safety and tolerability of apremilast Versus placebo in psoriatic arthritis: a pooled safety analysis of three phase III, randomized, controlled trials. ACR Open Rheumatol.2 (8), 459–470. 10.1002/acr2.11156

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 39

  MeaseP. J.HatemiG.ParisM.ChengS.MaesP.ZhangW.et al (2023). Apremilast long-term safety up to 5 years from 15 pooled randomized, placebo-controlled studies of psoriasis, psoriatic arthritis, and behçet's syndrome. Am. J. Clin. Dermatol24 (5), 809–820. 10.1007/s40257-023-00783-7

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 40

  MilneK. M.MitchellR. A.FergusonO. N.HindA. S.GuenetteJ. A. (2024). Sex-differences in COPD: from biological mechanisms to therapeutic considerations. Front. Med. (Lausanne)11, 1289259. 10.3389/fmed.2024.1289259

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41

  NIH policy on sex as a biological variable (SABV)Bethesda, MD: National Institutes of Health (2025). Available online at: https://orwh.od.nih.gov/sex-as-biological-variable (Accessed November 20, 2025).

  • Google Scholar

• 42

  PallerA. S.TomW. L.LebwohlM. G.BlumenthalR. L.BoguniewiczM.CallR. S.et al (2016). Efficacy and safety of crisaborole ointment, a novel, nonsteroidal phosphodiesterase 4 (PDE4) inhibitor for the topical treatment of atopic dermatitis (AD) in children and adults. J. Am. Acad. Dermatol75 (3), 494–503. 10.1016/j.jaad.2016.05.046

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  PappK.ReichK.LeonardiC. L.KircikL.ChimentiS.LangleyR. G. B.et al (2015). Apremilast, an oral phosphodiesterase 4 (PDE4) inhibitor, in patients with moderate to severe plaque psoriasis: results of a phase III, randomized, controlled trial (Efficacy and safety trial evaluating the effects of apremilast in psoriasis [ESTEEM] 1). J. Am. Acad. Dermatol73 (1), 37–49. 10.1016/j.jaad.2015.03.049

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44

  PaulC.CatherJ.GooderhamM.PoulinY.MrowietzU.FerrandizC.et al (2015). Efficacy and safety of apremilast, an oral phosphodiesterase 4 inhibitor, in patients with moderate-to-severe plaque psoriasis over 52 weeks: a phase III, randomized controlled trial (ESTEEM 2). Br. J. Dermatol173 (6), 1387–1399. 10.1111/bjd.14164

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  QueH.ZhengX.LiX.WangW. (2025). Risk factors for drug-induced acute generalized exanthematous Pustulosis(AGEP) from 2004 to 2024: a real-world study based on the FAERS. Clin. Cosmet. Investig. Dermatol18, 2835–2845. 10.2147/CCID.S544845

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46

  RuschilV.MazurakN.HofmannM.LoskutovaE.EnckP.FreilingerT.et al (2021). Decreased autonomic reactivity and psychiatric comorbidities in neurological patients with medically unexplained sensory symptoms: a case-control study. Front. Neurol.12, 713391. 10.3389/fneur.2021.713391

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 47

  ShanY.CheungL.ZhouY.HuangY.HuangR. S. (2023). A systematic review on sex differences in adverse drug reactions related to psychotropic, cardiovascular, and analgesic medications. Front. Pharmacol.14, 1096366. 10.3389/fphar.2023.1096366

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48

  SharmaS.GibbonsA.SaphireE. O. (2025). Sex differences in tissue-specific immunity and immunology. Science.389 (6760), 599–603. 10.1126/science.adx4381

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49

  ShobeiriP.KalantariA.TeixeiraA. L.RezaeiN. (2022). Shedding light on biological sex differences and microbiota–gut–brain axis: a comprehensive review of its roles in neuropsychiatric disorders. Biol. Sex. Differ.13 (1), 12. 10.1186/s13293-022-00422-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50

  SilverbergJ. I.KirsnerR. S.MargolisD. J.TharpM.MyersD. E.AnnisK.et al (2024). Efficacy and safety of crisaborole ointment, 2%, in participants aged ≥45 years with stasis dermatitis: results from a fully decentralized, randomized, proof-of-concept phase 2a study. J. Am. Acad. Dermatol90 (5), 945–952. 10.1016/j.jaad.2023.12.048

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51

  SportielloL.CapuanoA. (2024). Sex and gender differences and pharmacovigilance: a knot still to be untied. Front. Pharmacol.15, 1397291. 10.3389/fphar.2024.1397291

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52

  WaldmanS. A.ChristensenN. B.MooreJ. E.TerzicA. (2007). Clinical pharmacology: the science of therapeutics. Clin. Pharmacol. & Ther.81 (1), 3–6. 10.1038/sj.clpt.6100032

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53

  YanL.ZhangL.XuZ.LuoZ. (2025). A real-world disproportionality analysis of FDA adverse event reporting system (FAERS) events for lecanemab. Front. Pharmacol.16, 1559447. 10.3389/fphar.2025.1559447

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 54

  YounesS. (2024). The relationship between gender and pharmacology. Curr. Res. Pharmacol. Drug Discov.7, 100192. 10.1016/j.crphar.2024.100192

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 55

  YuT.FainK.BoydC. M.SinghS.WeissC. O.LiT.et al (2013). Benefits and harms of roflumilast in moderate to severe COPD. Thorax69 (7), 616–622. 10.1136/thoraxjnl-2013-204155

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 56

  ZaneL. T.ChandaS.JarnaginK.NelsonD. B.SpelmanL.GoldL. S. (2016). Crisaborole and its potential role in treating atopic dermatitis: overview of early clinical studies. Immunotherapy8 (8), 853–866. 10.2217/imt-2016-0023

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57

  ZhaoY. E.MaJ. X.HuL.XiaoS. X.ZhaoY. L. (2013). Meta‐analysis of the association between psoriasis and human leucocyte antigen‐B. Br. J. Dermatology169 (2), 417–427. 10.1111/bjd.12387

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58

  ZhaoB.FuY.CuiS.ChenX.LiuS.LuoL. (2024). A real-world disproportionality analysis of everolimus: data mining of the public version of FDA adverse event reporting system. Front. Pharmacol.15, 1333662. 10.3389/fphar.2024.1333662

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 59

  ZhengZ.WangZ.GuoQ.ChenX.ShenS.ZhangL.et al (2025). Disproportionality analysis comparing safety profiles of sonidegib and vismodegib based on the FAERS database. Sci. Rep.15 (1), 38170. 10.1038/s41598-025-22083-2

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 60

  ZuckerI.PrendergastB. J. (2020). Sex differences in pharmacokinetics predict adverse drug reactions in women. Biol. Sex. Differ.11 (1), 32. 10.1186/s13293-020-00308-5

  • Pubmed Abstract
  • CrossRef
  • Google Scholar