Adverse events associated with candida infections in secukinumab treatment Authors: Chongli Yu, Tian Ma, Nan Wu, Chengxin Li, Rui Wang Published: 2026 ABSTRACT BackgroundSecukinumab, an interleukin-17 A inhibitor widely used in psoriasis and related disorders, has been linked to opportunistic infections. Candida infections are a key safety concern, yet their real-world clinical profiles, risk distribution, and onset patterns remain insufficiently characterized.ObjectivesTo evaluate the clinical characteristics, severity, onset patterns, and risk factors of Candida infections associated with secukinumab, thereby informing risk stratification and patient management.MethodsA retrospective pharmacovigilance analysis was performed using the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) from Q1 2004 to Q1 2025. Secukinumab reports were retrieved, and Candida-related adverse events were identified using a predefined Candida-specific MedDRA Preferred Term dictionary. Disproportionality analyses were conducted using reporting odds ratios (RORs) with 95% confidence intervals (CIs). Event severity, subgroup characteristics, and time-to-onset were assessed, and Weibull shape parameter modeling was used to characterize risk patterns.ResultsWe identified 1,075 Candida events (0.8% of all secukinumab reports). Oral, esophageal, and vulvovaginal candidiasis predominated. All eight focal PTs showed significant disproportionality, with genital candidiasis yielding the strongest signal (ROR 19.85; 95% CI 12.91–30.53). Over half of events met criteria for serious outcomes. Stronger reporting signals were observed in females and adults aged 18–64 years. Median onset was 2–3 months, and Weibull β < 1 indicated an early-failure pattern, suggesting heightened risk shortly after treatment initiation.ConclusionsCandida infections during secukinumab therapy are uncommon but often severe, particularly in defined high-risk groups. Early onset highlights the importance of vigilant monitoring during initial treatment months. Real-world pharmacovigilance remains essential for optimizing biologic safety.Clinical trial numberNot applicable.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12879-026-13782-w. Introduction Interleukin-17 A (IL-17 A) is essential for mucocutaneous antifungal defense, mediating neutrophil recruitment, epithelial integrity, and rapid clearance of Candida spp [1, 2]. Secukinumab, a monoclonal antibody that selectively neutralizes IL-17 A, provides rapid and durable efficacy in plaque psoriasis, psoriatic arthritis, and ankylosing spondylitis [3–5]. It exerts its therapeutic effect by suppressing the IL-23/IL-17 axis, modulating extensive downstream transcriptional pathways, and producing sustained anti-inflammatory responses [6]. However, IL-17 blockade compromises antifungal immunity, and accumulating evidence has linked secukinumab with mucocutaneous candidiasis, raising safety concerns in long-term biologic therapy [7–9]. To date, large-scale pharmacovigilance analyses evaluating the association between secukinumab and candidiasis in real-world settings are lacking. Existing evidence is often limited by selective trial populations, short follow-up durations, and underreporting of adverse events, potentially underestimating the true infection risk. Additionally, few studies have systematically investigated population-specific susceptibility, severity stratification, or the temporal dynamics of Candida-related adverse events. The integration of pharmacovigilance analysis with clinical prioritization frameworks also remains unexplored. These gaps limit clinical recognition of high-risk patients and hinder development of surveillance strategies. Our study advances previous real-world pharmacovigilance analyses by incorporating severity stratification, time-to-onset Weibull modeling, and clinical prioritization scoring, providing actionable insight into monitoring strategies and patient risk management. Based on FDA Adverse Event Reporting System (FAERS) Database, this study systematically carried out signal mining and clinical feature analysis of adverse reactions related to secukinumab candida infection. We further characterized the signal strength of secukinumab-associated Candida-related adverse events, analyzed their severity distribution and the characteristics of affected populations, prioritized risks using a clinical priority scoring framework, and evaluated time-to-onset patterns, thereby providing actionable evidence for real-world risk assessment and individualized monitoring and management. Materials and methods Data source and case identification This study was a real-world, observational pharmacovigilance disproportionality analysis based on spontaneously reported adverse event (AE) data from the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS), covering the period from the first quarter of 2004 to the first quarter of 2025. Quarterly FAERS data files were retrieved from the publicly available FAERS Quarterly Data Extract Files. The major datasets used for analysis included DEMO (demographic and administrative information), DRUG (drug information), REAC (adverse event terms), and OUTC (clinical outcomes). Adverse events in FAERS are coded using the Medical Dictionary for Regulatory Activities (MedDRA). To improve case specificity, only reports in which secukinumab was recorded as the primary suspect drug (ROLE_COD = “PS”) were retained. Data cleaning and deduplication were conducted before analysis in accordance with FDA-recommended case-level consolidation principles. Specifically, when multiple records shared the same CASEID, only the most recent version was retained according to FDA_DT; when both CASEID and FDA_DT were identical, the record with the higher PRIMARYID was retained. Thus, follow-up reports superseded earlier submissions for the same case, and duplicate records across quarterly data releases were consolidated so that each individual case safety report (ICSR) contributed only once to the final analytic dataset. After deduplication and quality control, 131,657 unique secukinumab primary-suspect ICSRs were included for downstream analyses. Definition of candida-related adverse events Candida-related adverse events were identified using a predefined Candida-specific MedDRA Preferred Term (PT) dictionary established before signal detection. Because adverse events in FAERS are coded using MedDRA terminology, final case retrieval was performed by exact PT-level matching in the REAC file. Candidate terms were generated from MedDRA LLT/PT terminology entries explicitly containing “Candida,” “candidiasis,” “monilia,” or “moniliasis,” and were mapped to their corresponding PTs. The candidate PTs were then reviewed for clinical relevance and retained if they explicitly denoted Candida or monilial infection involving mucosal, cutaneous, urinary, respiratory, peritoneal, or systemic/invasive sites. Selected PTs were cross-checked against their Standardised MedDRA Query (SMQ) mappings to document terminology consistency; however, no entire SMQ was used as the primary case definition because broader SMQs may capture non-Candida or nonspecific infectious events. Terms indicating non-Candida fungal infection, nonspecific colonization, prophylactic antifungal use, or nonspecific mucosal symptoms were excluded unless Candida or monilial etiology was explicitly specified at the PT level. The complete PT dictionary, including PT codes, SMQ mapping, and inclusion rationale, is provided in Supplementary Table S1. A report was classified as Candida-related if at least one predefined Candida-related PT was present. Using this report-level definition, 1,075 Candida-related secukinumab primary-suspect reports were identified. Candida attribution was therefore based on the coded MedDRA PT rather than inference from nonspecific symptoms. Because FAERS does not provide microbiological culture results, histopathology, or adjudicated diagnostic records, these cases should be interpreted as reported Candida-related adverse events rather than laboratory-confirmed Candida infections. For report-level descriptive analyses, each deduplicated individual case safety report (ICSR) was counted once. For PT-level profiling and the stratified analysis shown in Fig. 3, entries were analyzed at the PT-event level; therefore, a single ICSR could contribute more than one Candida-related event if multiple predefined Candida-related PTs were reported in the same case. Table 1Clinical characteristics of reports with secukinumab-associated Candida infectionCandida-report group(N = 1,075)Non-Candida report group(N = 130,582)Overall secukinumab reports (N = 131,657)Gender, n(%) Female696 (64.7%)73,848 (56.6%)74,544 (56.6%) Male298 (27.7%)48,772 (37.3%)49,070 (37.3%) unknown81 (7.5%)7962 (6.1%)8043 (6.1%)Age (years), n(%) <180 (0%)384 (0.3%)384 (0.3%) 18~65431 (40.1%)38,263 (29.3%)38,694 (29.4%) 65~8599 (9.2%)9322 (7.1%)9421 (7.2%) >851 (0.1%)254 (0.2%)255 (0.2%) missing544 (50.6%)82,359 (63.1%)82,903 (63.0%)Weight (kg), n(%) <5014 (1.3%)429 (0.3%)443 (0.3%) 50~100137 (12.7%)9275 (7.1%)9412 (7.1%) >10038 (3.5%)3272 (2.5%)3310 (2.5%) missing886 (82.4%)117,606 (90.1%)118,492 (90.0%)Outcome, n(%) Congenital Anomaly0 (0%)29 (0.0%)29 (0.0%) Death10 (0.9%)3164 (2.4%)3174 (2.4%) Disability5 (0.5%)605 (0.5%)610 (0.5%) Hospitalization140 (13.0%)10,852 (8.3%)10,992 (8.3%) Life-Threatening14 (1.3%)1211 (0.9%)1225 (0.9%) Other906 (84.3%)114,721 (87.9%)115,627 (87.8%)Reporter, n(%) Consumer500 (46.5%)82,906 (63.5%)83,406 (63.4%) Health Professional127 (11.8%)11,850 (9.1%)11,977 (9.1%) Other448 (41.7%)35,826 (27.4%)36,274 (27.5%)Reporter country, n(%) US449 (41.8%)64,211 (49.2%)64,660 (49.1%) Non-US626 (58.2%)66,371 (50.8%)66,997 (50.9%)Reporting year, n (%) 202511 (1.0%)2262 (1.7%)2273 (1.7%) 202470 (6.5%)9112 (7.0%)9182 (7.0%) 202391 (8.5%)15,799 (12.1%)15,890 (12.1%) 202299 (9.2%)15,132 (11.6%)15,231 (11.6%) 2021114 (10.6%)15,338 (11.7%)15,452 (11.7%) 2020163 (15.2%)18,739 (14.4%)18,902 (14.4%) 2019138 (12.8%)17,636 (13.5%)17,774 (13.5%) 2018137 (12.7%)18,539 (14.2%)18,676 (14.2%) 2017137 (12.7%)10,081 (7.7%)10,218 (7.8%) 201684 (7.8%)5464 (4.2%)5548 (4.2%) 201531 (2.9%)2480 (1.9%)2511 (1.9%)Percentages were calculated within each column. Missing or unknown values were retained as separate categories. FAERS outcome categories are not mutually exclusive Table 1Clinical characteristics of reports with secukinumab-associated Candida infectionCandida-report group(N = 1,075)Non-Candida report group(N = 130,582)Overall secukinumab reports (N = 131,657)Gender, n(%) Female696 (64.7%)73,848 (56.6%)74,544 (56.6%) Male298 (27.7%)48,772 (37.3%)49,070 (37.3%) unknown81 (7.5%)7962 (6.1%)8043 (6.1%)Age (years), n(%) <180 (0%)384 (0.3%)384 (0.3%) 18~65431 (40.1%)38,263 (29.3%)38,694 (29.4%) 65~8599 (9.2%)9322 (7.1%)9421 (7.2%) >851 (0.1%)254 (0.2%)255 (0.2%) missing544 (50.6%)82,359 (63.1%)82,903 (63.0%)Weight (kg), n(%) <5014 (1.3%)429 (0.3%)443 (0.3%) 50~100137 (12.7%)9275 (7.1%)9412 (7.1%) >10038 (3.5%)3272 (2.5%)3310 (2.5%) missing886 (82.4%)117,606 (90.1%)118,492 (90.0%)Outcome, n(%) Congenital Anomaly0 (0%)29 (0.0%)29 (0.0%) Death10 (0.9%)3164 (2.4%)3174 (2.4%) Disability5 (0.5%)605 (0.5%)610 (0.5%) Hospitalization140 (13.0%)10,852 (8.3%)10,992 (8.3%) Life-Threatening14 (1.3%)1211 (0.9%)1225 (0.9%) Other906 (84.3%)114,721 (87.9%)115,627 (87.8%)Reporter, n(%) Consumer500 (46.5%)82,906 (63.5%)83,406 (63.4%) Health Professional127 (11.8%)11,850 (9.1%)11,977 (9.1%) Other448 (41.7%)35,826 (27.4%)36,274 (27.5%)Reporter country, n(%) US449 (41.8%)64,211 (49.2%)64,660 (49.1%) Non-US626 (58.2%)66,371 (50.8%)66,997 (50.9%)Reporting year, n (%) 202511 (1.0%)2262 (1.7%)2273 (1.7%) 202470 (6.5%)9112 (7.0%)9182 (7.0%) 202391 (8.5%)15,799 (12.1%)15,890 (12.1%) 202299 (9.2%)15,132 (11.6%)15,231 (11.6%) 2021114 (10.6%)15,338 (11.7%)15,452 (11.7%) 2020163 (15.2%)18,739 (14.4%)18,902 (14.4%) 2019138 (12.8%)17,636 (13.5%)17,774 (13.5%) 2018137 (12.7%)18,539 (14.2%)18,676 (14.2%) 2017137 (12.7%)10,081 (7.7%)10,218 (7.8%) 201684 (7.8%)5464 (4.2%)5548 (4.2%) 201531 (2.9%)2480 (1.9%)2511 (1.9%)Percentages were calculated within each column. Missing or unknown values were retained as separate categories. FAERS outcome categories are not mutually exclusive Clinical characteristics of reports with secukinumab-associated Candida infection Percentages were calculated within each column. Missing or unknown values were retained as separate categories. FAERS outcome categories are not mutually exclusive Handling of missing data Demographic variables, including sex, age, and body weight, contained varying degrees of missingness within FAERS. To minimize selection bias and retain maximal information, missing entries were recoded as “unknown” rather than removed during preprocessing. Reports containing missing demographic fields were included in the overall signal detection analysis to preserve dataset completeness, but were excluded from subgroup stratification analyses that required explicit demographic classification. The extent of missingness for key demographic and reporting variables is summarized in Supplementary Table S2. Reports with missing values were retained in the overall disproportionality analysis after recoding as “unknown”, but were excluded from subgroup-specific stratified analyses requiring explicit category assignment. Disproportionality analysis Disproportionality analysis was performed using the reporting odds ratio (ROR), a widely used case–non-case approach in pharmacovigilance signal detection [10]. A standard 2 × 2 contingency table was constructed to compare the reporting frequency of target events for secukinumab with that for all other drugs in the FAERS database. RORs and corresponding two-sided 95% confidence intervals (CIs) were calculated. A statistically significant disproportionality signal was defined as a Preferred Term (PT) with at least three reported cases and a lower bound of the 95% CI of the ROR greater than 1 [11]. Because disproportionality analysis reflects reporting imbalance rather than incidence or causality, the detected signals were interpreted as hypothesis-generating. Stratification analysis To further explore variations in disproportionality signals across subpopulations, stratified analyses were conducted according to sex, age, body weight, and reporter type. These analyses were performed at the REAC PT-event level. In Fig. 3, values under the “secukinumab” and “non-secukinumab” columns are presented as Candida-related PT-event entries/non-Candida PT-event entries within each stratum. The ROR for each stratum was calculated by comparing the event-level reporting odds for secukinumab primary-suspect reports with the corresponding event-level reporting odds for all other drugs in FAERS. For signal stratification, age was categorized as < 18 years, 18–65 years, and > 65 years, and body weight was categorized as < 80 kg, 80–100 kg, and > 100 kg, consistent with the grouping used in Fig. 3 and the underlying analytic code. RORs and 95% confidence intervals (CIs) were estimated separately for each stratum. Bonferroni correction was applied to adjust for multiple comparisons and control the family-wise error rate. Pearson’s chi-square test was used to compare the distribution of event frequencies across subgroups. For descriptive analyses shown in Tables 1 and 2, age and body weight were presented using separate clinically oriented categories (< 18, 18–65, 65–85, and > 85 years for age; <50, 50–100, and > 100 kg for body weight) to improve clinical interpretability. Table 2Comparison of clinical characteristics between serious and non-serious secukinumab-associated Candida reportsSerious cases(N = 574)Non-serious cases(N = 501)Statisticp-valueSEX, n(%) Female386 (67.2%)310 (61.9%)4.1930.041 Male153 (26.7%)145 (28.9%)7.5950.006WT (kg), n(%) <5010 (1.7%)4 (0.8%)-0.421 50~100113 (19.7%)24 (4.8%)33.452< 0.01 >10024 (4.2%)14 (2.8%)0.3070.580AGE (years), n(%) 18~64.9302 (52.6%)129 (25.7%)26.909< 0.01 65~8561 (10.6%)38 (7.6%)0.3440.558 > 851 (0.2%)0 (0%)-1Outcome, n(%) Death10 (1.7%)0 (0%)-- Disability5 (0.9%)0 (0%)-- Hospitalization140 (24.4%)0 (0%)-- Life-Threatening14 (2.4%)0 (0%)-- Required Intervention to Prevent Permanent Impairment / Damage1 (0.2%)0 (0%)-- Other404 (70.4%)0 (0%)--Types of AEs, n(%) Candida infection21527151.328< 0.01 Gastrointestinal candidiasis15--- Genital candidiasis14110.0800.778 Oesophageal candidiasis1133-- Oral candidiasis2181864.5080.034 Peritoneal candidiasis1--- Respiratory moniliasis3--- Skin candida1981.5150.218Serious and non-serious reports were classified according to the seriousness criteria recorded in FAERS. A report was considered serious if it included at least one seriousness outcome, such as death, life-threatening condition, hospitalization, disability, congenital anomaly, or other medically important condition; otherwise, it was classified as non-serious. Percentages were calculated using the total number of reports in each seriousness group as the denominator. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. The “Statistic” column reports the corresponding test statistic. A dash (–) indicates that the comparison was not applicable or could not be reliably estimated because of zero or very small cell counts Table 2Comparison of clinical characteristics between serious and non-serious secukinumab-associated Candida reportsSerious cases(N = 574)Non-serious cases(N = 501)Statisticp-valueSEX, n(%) Female386 (67.2%)310 (61.9%)4.1930.041 Male153 (26.7%)145 (28.9%)7.5950.006WT (kg), n(%) <5010 (1.7%)4 (0.8%)-0.421 50~100113 (19.7%)24 (4.8%)33.452< 0.01 >10024 (4.2%)14 (2.8%)0.3070.580AGE (years), n(%) 18~64.9302 (52.6%)129 (25.7%)26.909< 0.01 65~8561 (10.6%)38 (7.6%)0.3440.558 > 851 (0.2%)0 (0%)-1Outcome, n(%) Death10 (1.7%)0 (0%)-- Disability5 (0.9%)0 (0%)-- Hospitalization140 (24.4%)0 (0%)-- Life-Threatening14 (2.4%)0 (0%)-- Required Intervention to Prevent Permanent Impairment / Damage1 (0.2%)0 (0%)-- Other404 (70.4%)0 (0%)--Types of AEs, n(%) Candida infection21527151.328< 0.01 Gastrointestinal candidiasis15--- Genital candidiasis14110.0800.778 Oesophageal candidiasis1133-- Oral candidiasis2181864.5080.034 Peritoneal candidiasis1--- Respiratory moniliasis3--- Skin candida1981.5150.218Serious and non-serious reports were classified according to the seriousness criteria recorded in FAERS. A report was considered serious if it included at least one seriousness outcome, such as death, life-threatening condition, hospitalization, disability, congenital anomaly, or other medically important condition; otherwise, it was classified as non-serious. Percentages were calculated using the total number of reports in each seriousness group as the denominator. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. The “Statistic” column reports the corresponding test statistic. A dash (–) indicates that the comparison was not applicable or could not be reliably estimated because of zero or very small cell counts Comparison of clinical characteristics between serious and non-serious secukinumab-associated Candida reports Serious and non-serious reports were classified according to the seriousness criteria recorded in FAERS. A report was considered serious if it included at least one seriousness outcome, such as death, life-threatening condition, hospitalization, disability, congenital anomaly, or other medically important condition; otherwise, it was classified as non-serious. Percentages were calculated using the total number of reports in each seriousness group as the denominator. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. The “Statistic” column reports the corresponding test statistic. A dash (–) indicates that the comparison was not applicable or could not be reliably estimated because of zero or very small cell counts Clinical prioritization and time-to-onset analysis To distinguish statistical disproportionality from clinical relevance, PTs showing a statistically significant association were further evaluated using a semiquantitative clinical prioritization framework adapted from published pharmacovigilance methodology. Four criteria were assessed: clinical relevance, reporting rate, reported case fatality rate, and signal stability. Clinical relevance was determined according to whether a PT was classified as an Important Medical Event (IME) or Designated Medical Event (DME). Reporting rate was defined as the proportion of a given PT among all secukinumab-related adverse event reports and categorized as very common (≥ 10%), common (1–10%), or uncommon (≤ 1%). Reported case fatality rate was categorized as > 50%, 25–50%, or < 25%. Signal stability was defined according to the consistency of signal detection across four pharmacovigilance algorithms (ROR, PRR, MGPS, and BCPNN). Each criterion was assigned 0, 1, or 2 points, yielding a total score ranging from 0 to 8. Total scores of 0–2, 3–5, and 6–8 indicated low, moderate, and high clinical priority, respectively [12]. Time-to-onset (TTO) was defined as the interval, in days, between the start date of secukinumab therapy (START_DT) and the date of adverse event onset (EVENT_DT). Only reports with complete and logically consistent date fields were included in the TTO analysis. TTO distributions were summarized according to clinical priority category. For each priority group, Weibull shape parameter analysis was performed to characterize onset patterns. The scale parameter (α) and shape parameter (β) were estimated. A β < 1 indicates an early-failure type pattern, suggesting that the hazard decreases over time, whereas a β > 1 indicates a late-failure type pattern. Median TTO and interquartile range (IQR) were also calculated to improve clinical interpretability. Statistical analysis and visualization All statistical analyses and data visualizations were conducted using R software (version 4.3.1). Group comparisons for categorical variables were performed using Pearson’s chi-square test, with statistical significance set at a two-tailed p-value < 0.05. For analyses involving multiple comparisons, Bonferroni correction was applied to adjust significance thresholds, thereby ensuring the stability and reliability of the findings. Results Descriptive analysis A total of 131,657 deduplicated FAERS reports with secukinumab recorded as the primary suspect drug were identified, including 1,075 reports (0.8%) with predefined Candida-related adverse events and 130,582 reports without Candida-related adverse events (Fig. 1; Table 1). In terms of sex distribution, 64.7% of reports in the candida-report group involved female patients, significantly higher than male patients (27.7%). Regarding age, most secukinumab-related AEs occurred in individuals aged 18–64 years (40.1%), followed by those aged 65–85 years (9.2%), while reports from patients aged > 85 years and < 18 years were rare (0.1% and 0%, respectively). Fig. 1Flowchart of Data Extraction and Analysis Framework. FAERS data (2004Q1–2025Q1) were processed to identify 131,657 AE reports associated with secukinumab as the primary suspect drug, from which 1,075 Candida-related reports were extracted after deduplication. Subsequent analyses included clinical characteristics, disproportionality analysis, stratification analysis, clinical prioritization, and time-to-onset assessment Flowchart of Data Extraction and Analysis Framework. FAERS data (2004Q1–2025Q1) were processed to identify 131,657 AE reports associated with secukinumab as the primary suspect drug, from which 1,075 Candida-related reports were extracted after deduplication. Subsequent analyses included clinical characteristics, disproportionality analysis, stratification analysis, clinical prioritization, and time-to-onset assessment In terms of body weight, the majority of reports came from patients weighing 50–100 kg (12.7%), followed by those > 100 kg (3.5%) and < 50 kg (1.3%). Regarding clinical outcomes, the mortality rate in the Candida-report group was 0.9%, slightly lower than that in the non-Candida secukinumab-report group (2.4%). The hospitalization rate was significantly higher in the Candida-report group (13.0%) compared to the non-Candida secukinumab-report group (8.3%). Life-threatening events were also more frequent in the Candida-report group (1.3% vs. 0.9%). As for reporter type, 46.5% of reports in the Candida-report group were submitted by consumers, which was significantly lower than in the non-Candida secukinumab-report group. Reports submitted by healthcare professionals accounted for 11.8%, higher than the 9.1% observed in the non-Candida secukinumab-report group. Geographically, 58.2% of reports in the Candida-report group originated from non-U.S. regions, while 41.8% came from the United States. In terms of reporting year, most secukinumab-associated AE reports were concentrated in 2020 (15.2%), 2019 (12.8%), and 2021 (10.6%). Notably, the number of reports has declined in recent years, with only 6.5% reported in 2024 and 1.0% in 2025. Disproportionality analysis Among the 35 predefined Candida-related PTs used for case retrieval, eight PTs met the prespecified minimum reporting threshold and were therefore included in the PT-level disproportionality analysis. In this part of the analysis, PTs were evaluated at the event level rather than the deduplicated report level. The distribution and disproportionality analysis results are shown in Fig. 2. The most frequently reported PT was “Candida Infection” (n = 486), followed by “Oral Candidiasis” (n = 404), “Oesophageal Candidiasis” (n = 116), and “Vulvovaginal Candidiasis” (n = 77), most of which represent mucosal-associated infections. Fig. 2Disproportionality analysis of secukinumab-associated Candida-related adverse events (AEs). The red dashed line indicates the null value (ROR = 1). ROR, reporting odds ratio; CI, confidence interval Disproportionality analysis of secukinumab-associated Candida-related adverse events (AEs). The red dashed line indicates the null value (ROR = 1). ROR, reporting odds ratio; CI, confidence interval Compared with non-secukinumab-related reports, all included PTs showed statistically significant disproportionality signals, as the lower bounds of the 95% confidence intervals of the RORs exceeded 1. Genital candidiasis showed the highest disproportionality magnitude, with an ROR of 19.85 (95% CI: 12.91–30.53). According to the prespecified descriptive stratification based on RORL, skin candida, vulvovaginal candidiasis, gastrointestinal candidiasis, and genital candidiasis were categorized as strong statistical signals, whereas oral candidiasis, oesophageal candidiasis, and oropharyngeal candidiasis were categorized as moderate statistical signals. Serious vs. non-serious cases A total of 1,075 reports of Candida infections associated with secukinumab were identified, among which 574 cases (53.4%) were classified as serious adverse events (AEs), and 501 cases (46.6%) were non-serious (Table 2). Regarding sex distribution, both female (χ² = 4.193, p = 0.041) and male patients (χ² = 7.595, p = 0.006) showed statistically significant differences between serious and non-serious groups, suggesting that sex may influence the severity of adverse reactions. Age was also a significant factor: in the 18–64-year age group, the proportion of serious events was significantly higher than that in the non-serious group (52.6% vs. 25.7%, χ² = 26.909, p < 0.01), whereas no significant differences were observed in the 65–85-year (p = 0.558) and > 85-year (p = 1.000) groups. For body weight, patients weighing 50–100 kg had a significantly higher proportion of serious events compared to the non-serious group (19.7% vs. 4.8%, p < 0.01). In terms of clinical outcomes, hospitalization was the most common serious outcome, accounting for 24.4% of cases. Other severe outcomes such as death, disability, and life-threatening events were exclusively reported in the serious group, further highlighting their clinical importance. Among different infection types, “Candida infection” (unspecified) (χ² = 51.328, p < 0.01) and “Oral candidiasis” (χ² = 4.508, p = 0.034) were significantly more frequent in serious AEs compared to non-serious events. In contrast, “Genital candidiasis” and “Skin Candida” showed no statistically significant differences between the two groups. Stratification analysis To further explore the influence of population characteristics on the disproportionality signal of secukinumab-associated Candida-related adverse events, stratified analyses were conducted according to sex, age, body weight, and reporter type (Fig. 3). Unlike the descriptive analyses in Table 1, which were performed at the deduplicated report level, Fig. 3 summarizes Candida-related entries at the PT-event level. Therefore, the total number of Candida-related events shown in Fig. 3 is higher than the 1,075 Candida-related reports reported in Table 1, because a single ICSR may include more than one predefined Candida-related PT. Fig. 3Stratified event-level disproportionality analysis of secukinumab-associated Candida-related adverse events. Stratified analyses were performed at the REAC PT-event level. Under the “secukinumab” column, each value is shown as Candida-related PT-event entries/non-Candida PT-event entries among secukinumab primary-suspect reports within the corresponding stratum. Under the “non-secukinumab” column, each value represents the corresponding Candida-related PT-event entries/non-Candida PT-event entries for all other drugs in FAERS. Therefore, the event-level values shown in Fig. 3 are not expected to equal the 1,075 deduplicated Candida-related reports shown in Table 1. RORs were calculated by comparing these two event-level odds. CI, confidence interval; FAERS, FDA Adverse Event Reporting System; PT, Preferred Term; ROR, reporting odds ratio Stratified event-level disproportionality analysis of secukinumab-associated Candida-related adverse events. Stratified analyses were performed at the REAC PT-event level. Under the “secukinumab” column, each value is shown as Candida-related PT-event entries/non-Candida PT-event entries among secukinumab primary-suspect reports within the corresponding stratum. Under the “non-secukinumab” column, each value represents the corresponding Candida-related PT-event entries/non-Candida PT-event entries for all other drugs in FAERS. Therefore, the event-level values shown in Fig. 3 are not expected to equal the 1,075 deduplicated Candida-related reports shown in Table 1. RORs were calculated by comparing these two event-level odds. CI, confidence interval; FAERS, FDA Adverse Event Reporting System; PT, Preferred Term; ROR, reporting odds ratio The overall disproportionality signal was statistically significant, with a reporting odds ratio (ROR) of 4.21 (95% CI: 3.98–4.46, p < 0.05). All subgroups showed significantly elevated RORs, with the lower bounds of the 95% confidence intervals exceeding 1, indicating a consistent disproportionality signal across strata. In the sex-stratified analysis, the signal was slightly stronger in females (ROR = 4.43, 95% CI: 4.13–4.76) than in males (ROR = 3.45, 95% CI: 3.09–3.85). Age-stratified analysis showed elevated signals in both the 18–65 years group (ROR = 3.88, 95% CI: 3.54–4.25) and the > 65 years group (ROR = 3.74, 95% CI: 3.11–4.51). Stratification by body weight showed heterogeneity across categories, with the highest ROR observed in patients weighing > 100 kg (ROR = 5.06, 95% CI: 3.82–6.71), whereas the 80–100 kg subgroup showed the lowest signal (ROR = 2.98, 95% CI: 2.28–3.90). When stratified by reporter type, reports submitted by health professionals showed a stronger signal (ROR = 5.68, 95% CI: 5.10–6.34) than consumer-submitted reports (ROR = 4.15, 95% CI: 3.81–4.52). This pattern may reflect differences in recognition and reporting practices between reporter groups. Clinical prioritization of the disproportionality signals Among the eight PTs included in the analysis, oesophageal candidiasis and gastrointestinal candidiasis were classified as IMEs/DMEs, whereas the remaining six PTs were not categorized as IMEs/DMEs (Table 3). Detailed sub-scores and total scores for each PT are provided in Supplementary Table S3, According to the clinical prioritization scoring system, six adverse events (AEs) were rated as low priority (score = 2), and two (25.00%) as moderate priority (score = 3). No high-priority signals (score ≥ 6) were identified. Notably, “Genital candidiasis,” despite showing the strongest disproportionality signal (ROR = 19.85, 95% CI: 12.91–30.53), was still categorized as low-priority due to the absence of serious clinical outcomes. Table 3Clinical prioritization of secukinumab-associated candida-related Preferred TermsPTReports(n)ROR (95% CI)Deaths(n)IME/DMEsTotal scoreClinical priorityCandida infection4863.73 (3.40–4.08)602LowOral candidiasis4045.14 (4.65–5.68)502LowOesophageal candidiasis1164.91 (4.08–5.92)013ModerateVulvovaginal candidiasis779.40 (7.44–11.87)002LowOropharyngeal candidiasis325.46 (3.82–7.79)002LowSkin candida278.28 (5.59–12.26)002LowGenital candidiasis2519.85 (12.91–30.53)002LowGastrointestinal candidiasis156.40 (3.80–10.80)013ModerateClinical priority was assessed using a semiquantitative four-item scoring framework, including clinical relevance, reporting rate, reported case fatality rate, and signal stability. Each item was assigned 0–2 points, and total scores of 0–2, 3–5, and 6–8 indicated low, moderate, and high clinical priority, respectively Table 3Clinical prioritization of secukinumab-associated candida-related Preferred TermsPTReports(n)ROR (95% CI)Deaths(n)IME/DMEsTotal scoreClinical priorityCandida infection4863.73 (3.40–4.08)602LowOral candidiasis4045.14 (4.65–5.68)502LowOesophageal candidiasis1164.91 (4.08–5.92)013ModerateVulvovaginal candidiasis779.40 (7.44–11.87)002LowOropharyngeal candidiasis325.46 (3.82–7.79)002LowSkin candida278.28 (5.59–12.26)002LowGenital candidiasis2519.85 (12.91–30.53)002LowGastrointestinal candidiasis156.40 (3.80–10.80)013ModerateClinical priority was assessed using a semiquantitative four-item scoring framework, including clinical relevance, reporting rate, reported case fatality rate, and signal stability. Each item was assigned 0–2 points, and total scores of 0–2, 3–5, and 6–8 indicated low, moderate, and high clinical priority, respectively Clinical prioritization of secukinumab-associated candida-related Preferred Terms Clinical priority was assessed using a semiquantitative four-item scoring framework, including clinical relevance, reporting rate, reported case fatality rate, and signal stability. Each item was assigned 0–2 points, and total scores of 0–2, 3–5, and 6–8 indicated low, moderate, and high clinical priority, respectively Time-to-onset analysis The results of time-to-onset (TTO) analysis and Weibull shape parameter (WSP) modeling are presented in Table 4. When stratified by clinical prioritization level, both moderate- and low-priority signals exhibited relatively delayed onset patterns. Specifically, the median TTO was 57.5 days (interquartile range [IQR]: 27.5–125.75) for moderate-priority events and 74 days (IQR: 28–185) for low-priority events. Table 4Time-to-onset and weibull parameter estimates for adverse events associated with secukinumab, stratified by clinical priority levelPrioritizationCasesTTO (days)Scale parameterShape parameter n Median (IQR)Min - maxα95% CIβ95% CIModerate3657.5(27.5-125.75)1-838109.2556.83-161.680.720.55–0.9Low24574(28–185)1-1630135.79110.4-161.190.710.64–0.77n, number of cases with available time-to-onset data. TTO, time-to-onset, defined as the interval (in days) between the initiation of secukinumab treatment and the occurrence of the adverse event. IQR, interquartile range. α, Weibull scale parameter; higher values suggest greater variability in onset time. β, Weibull shape parameter; β < 1 indicates a “early failure type” distribution, where risk decreases over time. When TTO = 0, the adverse event occurred on the same day as drug administration Table 4Time-to-onset and weibull parameter estimates for adverse events associated with secukinumab, stratified by clinical priority levelPrioritizationCasesTTO (days)Scale parameterShape parameter n Median (IQR)Min - maxα95% CIβ95% CIModerate3657.5(27.5-125.75)1-838109.2556.83-161.680.720.55–0.9Low24574(28–185)1-1630135.79110.4-161.190.710.64–0.77n, number of cases with available time-to-onset data. TTO, time-to-onset, defined as the interval (in days) between the initiation of secukinumab treatment and the occurrence of the adverse event. IQR, interquartile range. α, Weibull scale parameter; higher values suggest greater variability in onset time. β, Weibull shape parameter; β < 1 indicates a “early failure type” distribution, where risk decreases over time. When TTO = 0, the adverse event occurred on the same day as drug administration Time-to-onset and weibull parameter estimates for adverse events associated with secukinumab, stratified by clinical priority level n, number of cases with available time-to-onset data. TTO, time-to-onset, defined as the interval (in days) between the initiation of secukinumab treatment and the occurrence of the adverse event. IQR, interquartile range. α, Weibull scale parameter; higher values suggest greater variability in onset time. β, Weibull shape parameter; β < 1 indicates a “early failure type” distribution, where risk decreases over time. When TTO = 0, the adverse event occurred on the same day as drug administration In the Weibull distribution fitting, the scale parameter (α) for low-priority signals was 135.79 (95% CI: 110.4–161.19), slightly higher than that of moderate-priority signals (α = 109.25, 95% CI: 56.83–161.68), suggesting a longer and more variable onset distribution for the former. Notably, the shape parameters (β) for both priority groups were less than 1—0.71 (95% CI: 0.64–0.77) for low-priority and 0.72 (95% CI: 0.55–0.90) for moderate-priority events—indicating a decreasing hazard over time, consistent with an early-failure event pattern. Discussion This study systematically evaluated the association between secukinumab and Candida infections using real-world pharmacovigilance data from the FAERS database. By integrating severity stratification, subgroup analysis, clinical prioritization scoring, and time-to-onset assessment, our findings support a clinically meaningful reporting association between IL-17 A blockade and Candida infections, especially oral, oesophageal, and genital presentations. These observations are consistent with current mechanistic understanding of IL-17-mediated antifungal defense and provide clinically relevant information for risk recognition and early monitoring during secukinumab therapy. IL-17 A plays a pivotal role in mucosal host defense against Candida by promoting epithelial antimicrobial peptide production, neutrophil recruitment, and maintenance of barrier integrity at oral, gastrointestinal, and genitourinary surfaces [13–15]. The IL-23/IL-17 pathway is regarded as a core immune axis in protection against fungal invasion such as Candida, coordinating barrier integrity and immune responses at mucosal sites [16, 17]. Inhibition of IL-17 A will weaken the above defense mechanisms, resulting in a decrease in the local antifungal ability of the mucosa and making patients more prone to Candida colonization and infection [18]. This biological mechanism is well aligned with our findings, in which Candida infection, oral candidiasis, oesophageal candidiasis, and vulvovaginal candidiasis were the most frequently reported event types. The predominance of mucosal phenotypes in our dataset further supports the mechanistic plausibility that secukinumab-associated Candida events are linked to impaired IL-17-dependent antifungal immunity rather than representing a random reporting pattern. Importantly, variation in Candida-related infection patterns across biologic classes further supports the biological relevance of our findings. Prior pooled clinical and real-world studies have shown that Candida infections are more frequently reported with IL-17-pathway inhibitors than with biologics targeting other inflammatory axes. Prior pooled analyses have reported exposure-adjusted Candida infection rates of 2.2, 1.5, and 0.7 per 100 person-years among patients with psoriasis, psoriatic arthritis, and ankylosing spondylitis receiving IL-17 inhibitors, respectively—substantially higher than those reported with TNF-α inhibitors [19].Recent multicenter real-world evidence further supports this pattern. In a 16-week multicenter real-world study of bimekizumab in patients with plaque psoriasis, candidiasis was reported in 10.5% of treated patients, with oral candidiasis being the most frequent presentation, further suggesting that Candida-related events represent a clinically relevant class-associated safety issue during IL-17-pathway inhibition [20]. By contrast, long-term safety data for the IL-23 inhibitor risankizumab have shown a very low burden of opportunistic infections, with no reported systemic candidiasis in the published extension analysis, supporting the view that Candida-related adverse events may be less prominent under IL-23 inhibition [21]. This elevated risk has also been observed with other IL-17 pathway inhibitors. For instance, brodalumab—a monoclonal antibody targeting the IL-17 receptor A—was associated with a 4.8% incidence of mucosal candidiasis in psoriasis trials, compared to only 1.3% in patients treated with the IL-12/23 inhibitor ustekinumab [22]. Additionally, case reports have documented recurrent post-injection candidiasis during secukinumab therapy in otherwise immunocompetent individuals, which resolved following a switch to IL-23 inhibition with guselkumab [23]. Collectively, these data reinforce the indispensable role of IL-17 in maintaining mucosal immunity and highlight the need for careful infection monitoring during IL-17 A inhibition. The present study was intentionally designed as a secukinumab-focused pharmacovigilance analysis rather than a formal comparative class analysis. The objective was to provide an in-depth characterization of Candida-related reporting patterns, severity distribution, clinical priority, and temporal features for a single widely used IL-17 A inhibitor. However, these comparisons should be interpreted as contextual rather than inferential, because they are derived from independent studies with different designs, populations, and outcome definitions. Among 1,075 Candida cases, mucosal infections dominated the clinical spectrum. This distribution aligns well with the known role of the IL-17 pathway in maintaining mucosal barrier integrity and regulating antifungal immunity. Compared with earlier randomized controlled trials, post-marketing pharmacovigilance data may capture a greater burden of reported Candida-related events. According to existing post-marketing surveillance data, the exposure-adjusted incidence rate of secukinumab-associated candidiasis is estimated at 2.16 per 100 person-years, with approximately 15% classified as serious adverse events [24]. Moreover, prior studies have predominantly focused on oral and esophageal candidiasis in the context of secukinumab use [25]. A large U.S.-based real-world cohort study involving 33,536 patients reported that the incidence of Candida infection with secukinumab (5.1 per 1,000 person-years) was higher than that observed with several other biologics, with esophageal candidiasis being the most frequently reported subtype [26]. Notably, genital candidiasis showed the strongest disproportionality signal in our study, despite being less emphasized in earlier literature than oral or oesophageal candidiasis. This finding suggests that genital Candida-related events may be underrecognized in routine clinical settings and deserve greater attention during biologic safety monitoring. Although previous reports suggest that most Candida infections associated with secukinumab are mild to moderate in severity [27, 28], our analysis indicates that a substantial proportion of reported cases were serious, including hospitalization and other clinically important outcomes. Hospitalization was the most common severe outcome, accounting for 24.4% of serious cases. Fatal, disabling, or life-threatening events occurred exclusively in the serious AE group, underscoring the importance of monitoring high-risk individuals. In addition, Candida infection not otherwise specified and oral candidiasis were more frequently observed in serious than in non-serious reports, whereas genital candidiasis, despite showing the highest disproportionality magnitude, was not significantly enriched among serious cases. These findings suggest that signal strength and clinical severity do not necessarily overlap, underscoring the value of combining disproportionality measures with clinical-priority assessment. Subgroup analyses showed variation in the magnitude of disproportionality signals across sex, age, body weight, and reporter strata. Female reports showed a slightly stronger signal than male reports, and the > 100 kg subgroup showed the highest ROR among the weight categories, whereas the 80–100 kg subgroup showed the lowest. Age-stratified analysis demonstrated significant signals in both the 18–65 years and > 65 years groups. However, these subgroup patterns should be interpreted cautiously. Because FAERS is a spontaneous reporting system that captures reported events rather than exposed populations, a greater number of reports or a higher disproportionality estimate in a given subgroup does not necessarily indicate a higher true incidence of Candida infection in that subgroup. For example, the larger number of reports observed in adults aged 18–65 years may reflect the greater number of exposed individuals in this treatment population, differences in healthcare use, or reporting behavior, rather than a true age-specific drug effect. Similarly, reports submitted by healthcare professionals showed stronger signals than those from non-professional reporters, which may reflect differences in recognition and reporting practices rather than true biological differences. Therefore, the subgroup patterns identified in FAERS should be interpreted as reporting signals rather than evidence of subgroup-specific excess incidence or causal vulnerability. The clinical-priority analysis further refined interpretation of the PT-level signals. Oesophageal candidiasis and gastrointestinal candidiasis were classified as important medical events, whereas several other PTs with high disproportionality values remained low priority after integration of clinical relevance, reporting frequency, case fatality, and signal stability. This distinction is important because disproportionality magnitude alone may overemphasize events that are frequently reported but not necessarily clinically severe. Conversely, less frequently reported PTs may deserve closer attention when they involve more serious clinical consequences. The integration of statistical and clinical dimensions therefore provides a more balanced framework for identifying signals of practical relevance in pharmacovigilance research. Time-to-onset analysis showed that both moderate- and low-priority signals followed an early-failure pattern, with Weibull shape parameters below 1 and median onset times concentrated within the first several months of treatment. These results suggest that the reporting risk of Candida-related events may be greatest during the initial phase of secukinumab exposure, highlighting the first 12 weeks as a clinically relevant monitoring window. From a practical perspective, early vigilance for oral, oesophageal, and vulvovaginal symptoms may facilitate timely recognition and treatment, reduce progression to more severe infection, and minimize unnecessary interruption of effective biologic therapy. Despite the strengths of a large real-world pharmacovigilance dataset, several limitations should be acknowledged. FAERS is a spontaneous reporting system and is therefore subject to under-reporting, selective reporting, stimulated reporting, and incomplete clinical information. More severe or clinically unusual events may be preferentially reported, which could distort the apparent severity distribution of Candida-related reports. Because reliable exposure denominators are unavailable, disproportionality signals cannot be interpreted as estimates of incidence, absolute risk, or causality. Candida-related case identification was based on coded MedDRA Preferred Terms rather than microbiologically adjudicated diagnoses. Although the predefined PT dictionary was restricted to terms explicitly indicating Candida or monilial etiology, FAERS does not provide culture results, histopathology, detailed clinical narratives, or standardized diagnostic adjudication; therefore, residual diagnostic miscoding and false-positive classification cannot be excluded. The retrieved cases should therefore be interpreted as reported Candida-related adverse events rather than laboratory-confirmed Candida infections. In addition, missing demographic or clinical variables may affect subgroup comparability, and event-level analyses should not be directly equated with report-level descriptive counts. Finally, although other IL-17- and IL-23-targeted biologics were discussed for clinical context, this study was limited to secukinumab and was not designed for formal cross-drug comparison. Based on large-scale real-world pharmacovigilance data from FAERS, this study characterized the signal strength, severity profile, clinical priority level, and time-to-onset pattern of secukinumab-associated Candida infections. Although most reported events were mild to moderate and clinically manageable, the predominance of early-onset cases and the presence of vulnerable subpopulations highlight the need for strengthened monitoring during the initial treatment period. To mitigate infection-related risks while preserving therapeutic benefit, clinicians are advised to conduct individualized risk assessment before therapy initiation, maintain vigilance within the first 12 weeks of treatment, and closely inspect high-risk mucosal sites—particularly the oral and vulvovaginal regions. Early recognition and timely intervention may prevent localized infections from progressing to systemic disease and reduce treatment interruption. In summary, while secukinumab remains an effective therapy for inflammatory disorders, awareness of Candida infection risk and proactive management strategies are essential to ensure safe and optimized clinical use. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1