Real-world sex differences in treatment persistence and reasons for discontinuation in psoriatic arthritis patients: results from the German RABBIT-SpA register

Background In psoriatic arthritis (PsA), growing evidence indicates sex-specific differences regarding clinical manifestation and treatment outcomes. Research has highlighted that females may be less likely to achieve treatment targets and are more prone to discontinuing therapy, though data on sex-specific adverse events is limited. This analysis investigates sex differences in treatment outcomes, persistence, discontinuation reasons, and adverse events during first-line b/tsDMARD therapy.

In this analysis bionaïve patients with PsA from the RABBIT-SpA register were included at the start of their first b/tsDMARD. Therapy persistence was estimated using the Cox-regression adjusted for age. Descriptive analyses were used to examine and compare sex–specific differences on reasons for therapy discontinuation.

A total of 457 female patients and 343 male patients were included. Females exhibited more severe joint involvement and poorer patient-reported parameters, such as higher disease activity, more pain, and greater functional limitations. In contrast, males showed more pronounced skin involvement and a higher prevalence of nail psoriasis. Females had lower treatment persistence rates, both in the overall analysis of all first-line b/tsDMARDs and in subgroup analyses restricted to TNFi and IL17i therapies. At 12 months, 52% of females and 68% of males remained on their initial b/tsDMARD therapy. Notable sex differences were also observed in the reasons for therapy discontinuation: males more frequently discontinued due to lack of efficacy or remission, while females more often stopped treatment due to adverse events. Our safety analysis indicated that although female patients experienced a greater number of overall adverse events, males reported serious adverse events at twice the rate.

Our findings underscore the need for sex-specific treatment strategies and more comprehensive research into biological and sociocultural factors influencing therapy persistence and reasons for discontinuation in real-world settings. Tailored treatment strategies are needed with regard to biologic therapy to overcome worse therapeutic outcomes in female patients with PsA. Real-world sex differences in treatment persistence and reasons for discontinuation in psoriatic arthritis patients: results from the German RABBIT-SpA register Lisa Lindner1* , Anja Weiß1 , Andreas Reich1 , Christine Baumann2, Frank Behrens3 , Xenofon Baraliakos4 and Anne C. Regierer1 Page 2 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188

Psoriatic-Arthritis (PsA) is a chronic inflammatory dis - ease that not only affects the joints but can also affect the skin, and entheses, with overall prevalence considered relatively equal between females and males [ 1]. However, several factors including hormonal influences, environ - mental and sociocultural factors, and genetic predisposi - tions contribute to differences in disease expression and progression between the sexes [2]. Studies on PsA patients have shown sex-specific dif - ferences regarding clinical manifestation, disease impact and therapy response [ 3, 4]. Males typically have more axial and oligoarticular involvement, along with more severe skin manifestations and rapid radiographic dam - age. Females, on the other hand, are more likely to have polyarticular involvement, enthesitis and a higher disease burden including more functional limitations [5, 6]. Initial treatment of PsA includes conventional synthetic disease-modifying anti-rheumatic drugs (csDMARDs), while biologic and targeted synthetic DMARDs (b/tsD - MARDs) such as tumor necrosis factor inhibitors (TNFi) and Interleukin-17 inhibitors (IL17i) are considered for patients who do not respond adequately to first-line options. The primary treatment objective is to achieve remission or at least minimal disease activity (MDA), aiming to prevent further joint damage and enhance patients’ overall well-being. The current EULAR (Euro - pean Alliance of Associations for Rheumatology) recom - mendations stress the importance of a tailored treatment strategy, taking into account individual patients’ needs, comorbidities and preferences. To optimize individual - ized therapy, it is essential to recognize the influence of sex on these factors. Accordingly, research on the effect of sex on treatment choices, treatment efficacy, and treat- ment maintenance was prioritized in the recommenda - tions for future research in PsA [7]. Current studies suggest that male sex may have a favor- able impact on therapy persistence and outcome, with females less likely to achieve MDA [ 7– 9]. Reasons for this disparity are multifaceted and may involve a combi - nation of hormonal, biological, disease presentation and treatment-related factors. The specific mechanisms and reasons remain unclear. The most common reasons for early therapy discon - tinuation in longitudinal observational studies include remission, lack of efficacy, and serious and non-serious adverse events (SAEs and AE) [ 10, 11]. However, a nota - ble gap exists in sex-stratified data, particularly concern - ing adverse events during first-line b/tsDMARD therapy. Although some randomized trials have provided sex- stratified data on AEs [ 12], comprehensive observational studies with sex stratification remain limited. The objectives of our analyses were to examine sex- specific differences in clinical parameters and first-line therapy persistence, as well as reasons for discontinua - tion. Additionally, we provide sex-stratified data on fre - quencies of SAEs and AEs occurring during the first-line b/tsDMARD therapy and leading to discontinuation. Patients and methods Data source The RABBIT-SpA register is a German longitudinal observational cohort study focused on monitoring and evaluating the safety and effectiveness of b/tsDMARDs in patients with PsA in a real-world setting. Patients can be included by a rheumatologist with the start of a new b/ tsDMARD or with a conventional systemic treatment e.g. csDMARD or NSAID (nonsteroidal anti-inflammatory drugs) [ 13]. Physicians and patients enter data using a web-based documentation system. Clinical and patient- reported data is collected through an electronic case report form (eCRF) at baseline, at three and six months post-enrolment, and subsequently at six-month intervals for a follow-up period of up to ten years [14]. Patients All PsA patients enrolled between October 2017 and database closure in March 2024 were included. For these analyses, only b/tsDMARD-naïve patients with PsA who initiated their first-line b/tsDMARD and who have a fol - low-up time of at least one year were selected (Supple - mentary Figure S1). In our register, sex was reported as “female” , “male” , or “other” . Since no patient’s sex was reported as “other” , this study only refers to female and male patients. While the distinction between sex and gender is well established in the social sciences, it remains a complex and evolving issue in medical research [ 15]. In this study, we use the term sex acknowledging that both biological and sociocultural factors may influence treatment out - comes, although the documentation of these variables may vary across study sites and may not consistently reflect this distinction. Statistical analyses and variables Definition of variables Characteristics at the beginning of first-line therapy of female and male participants were compared using Clinical trial number Not applicable.

Observational study, Psoriatic arthritis, Gender, Sex, Women, Men, First-line therapy, bDMARD, tsDMARD, Biologics, Therapy outcome, Patient-reported outcome Page 3 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 appropriate descriptive statistics, including means, stan - dard deviations, and percentages. The physician eCRF contains data on treatment history and the current disease activity status and treatment reg - imen. Physicians record the onset of symptoms and the time of diagnosis. Disease duration is calculated in years since diagnosis, while diagnostic delay represents the time in years between onset of symptoms and diagnosis. Additional clinical parameters reported by the physician include dactylitis, axial involvement, and nail psoriasis, which are assessed using a binary yes or no response, enthesitis (SPARCC score including 16 entheses), swollen joint counts (66 joints, SJC-66), and tender joint counts (68 joints, TJC-68). The affected body surface area (BSA) is noted as a percentage and the C-Reactive Protein (CRP) value is provided as continuous value in mg/l. The DAS-28-CRP is calculated using 28 swollen and tender joint counts, CRP (mg/l), and patient global assessment. Physician global assessment is rated on a numeric rating scale (NRS 0–10). The DAPSA is a continuous measure focusing on joint involvement and providing granular information about disease activity [16]. Comorbidity data were obtained from the physician’s CRF, where comor - bidities are documented using a checklist of predefined conditions (e.g., cardiovascular disease, renal disease, malignancies, depression) with the option to add further conditions in free-text fields. The Rheumatic Disease Comorbidity Index (RDCI) is used to quantify the bur - den of comorbid conditions in patients with rheumatic diseases [ 17] including PsA [ 18]. The index incorporates ten comorbidities (lung disease, myocardial infarction, stroke or other cardiovascular diseases, hypertension, fracture, depression, diabetes, cancer, and ulcers or stom- ach symptoms). These comorbidities, as documented by the physician, are assigned specific weights to generate a composite score ranging from 0 to 9. A higher value reflects a higher burden of comorbid conditions. Patients provide data on sociodemographic charac - teristics and a variety of patient-reported outcome mea - sures. The patient global assessment, pain, and sleep disturbance are rated on NRS 0–10. The WHO-5 Well- Being Index (WHO-5) is a widely used five-item instru - ment for screening depressive symptoms, yielding a total score from 0 to 100, with scores below 29 indicating moderate to severe depressive symptoms. Higher scores reflect greater well-being [ 19, 20]. The Dermatology Life Quality Index (DLQI) is reported annually by the patients to assess the impact of dermatological conditions on their quality of life. The Health Assessment Questionnaire (HAQ) is administered at each time point to evaluate the patients’ functional ability and the extent of disability related to their condition. Therapy persistence rate The Kaplan-Meier method was utilized to estimate ther - apy persistence to visualize the time-to-treatment dis - continuation of the first-line b/tsDMARD therapy for the two groups of interest. One therapy episode was defined as an episode of the same active substance that has not been paused for more than 90 days. Patients who contin - ued therapy after database close or were lost-to-follow up were censored. To analyze the direct effect of sex on therapy persis - tence, we used a minimally adjusted model. Age was included as covariate, as it may influence therapy per - sistence independently, through factors such as comor - bidities, treatment tolerability, or healthcare-seeking behavior, but is not considered a mediator in the rela - tionship between sex and therapy persistence. Additional covariates that could potentially act as mediators, such as disease severity or other sociodemographic factors, were deliberately excluded to avoid attenuating or obscuring the direct effect of sex [ 21, 22]. By adjusting for age only, we aimed to isolate the specific contribution of sex to therapy persistence, without introducing other variables that might blur this relationship. The age-adjusted analy - sis was performed using a Cox Regression model. Reasons for discontinuation Physician-reported reasons for discontinuation were cat - egorized into adverse events, lack of efficacy, remission and other reasons. For all patients who stopped their first-line therapy during observation proportion of rea - sons for discontinuation were analyzed. Adverse events analysis To enhance clarity, SAEs and AEs are collectively referred to as events unless otherwise specified and needed. In this event analysis we only included patients for whom the reason to discontinue was reported as “(seri - ous) adverse event” . The specific event(s) that led to ther- apy discontinuation cannot definitively be identified, due to the way events are reported in the register, because physicians are not explicitly asked to specify which spe - cific event or a combination of events was the reason for discontinuation. Therefore, to identify an event as a treatment-related reason for discontinuation, we used the following assumptions: The primary criterion to classify an event, as either treatment-related or non-treatment-related, was the causal relationship to the therapy indicated by the physician as ‘definite, ’ ‘probable, ’ or ‘possible’ . Secondly, if no information on the causal relationship was provided or it was indicating an ‘unlikely’ , ‘no causal’ or ‘unknown’ relationship, the event was selected time-based. This means if the event occurred no more than 90 days before and 14 days after the date of therapy discontinuation, it Page 4 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 was eligible for selection and therefore matching to the therapy episode. The risk windows for neoplasms and malignancies were extended to the end of observation time. If an event could not be classified as either having a causal or time-based relationship to the first-line therapy, it was considered out of range and excluded from further event analysis.

Patients characteristics: demographics, disease variables and patient-reported outcomes Table 1 presents the characteristics at start of first-line therapy of 800 patients with PsA, stratified by female sex ( n = 457) and male sex ( n = 343). The mean age was 52.3 years in females, while males were slightly younger. A higher proportion of females reported being current smokers and they were more frequently obese. Disease duration for joints and skin involvement was similar for both sexes. Females exhibited a longer diagnostic delay for joint and skin involvement compared to males. And they had less BSA affected. Nail psoriasis and dactyli - tis were more common among males. The proportion of patients with ≥ 5  mg/l CRP was similar between the sexes. The average number of SJC-66 and TJC-68 was higher in females, as well as the proportion of patients with ≥ 5 affected joints. Females also had a higher propor - tion of enthesitis and more enthesitis sites. Females had a slightly higher mean physician global disease activity score, DAPSA and DAS-28-CRP compared to males. Female patients reported higher values in most patient- related outcome measures including patient global, pain, sleep disturbance, functional status, and well-being. The mean DLQI score was comparable for both sexes. Treatments Table  2 presents the proportion of first-line b/tsD - MARDs, concomitant treatments, and therapy continu - ation after one year, stratified by sex. Among first-line b/ tsDMARD therapies, TNFi was the most commonly used treatment in both females and males, prescribed to 54% of females and 57% of males. The second most frequently prescribed therapy was IL17i, administered to 26% of females and 31% of males. NSAID use was reported for 43% of females compared to 35% of males. Glucocorticoids were prescribed in 32% of the overall population, with a higher use among females. Additionally, both non-opioid and opioid anal - gesics were more commonly used in females. Persistence rates Figure 1 presents the age-adjusted therapy persistence rates across all therapies (1a) (including TNFi, IL17i and other modes of action) as well as specifically for TNFi (1b) and IL17i (1c). Overall, 59% of patients remained on their first-line b/tsDMARD therapy after one year of observation, with 52% of females and 68% of males continuing their initial treatment (HR: 0.62 [0.50–0.76], p = 0.000). Males demonstrated higher persistence rates across all therapies after adjustment, as well as for TNFi (HR: 0.56 [0.42–0.76], p = 0.000) and IL17i (HR: 0.63 [0.42–0.96], p = 0.030) in particular. Reason for therapy discontinuation Of 800 patients, 328 patients discontinued their first- line treatment during follow-up within 12 months. The reason for discontinuation was reported in 213 patients (65%) (Table 3). The most common reasons were lack of efficacy and AEs. A higher proportion of males discon - tinued therapy due to lack of efficacy. In contrast, more females discontinued due to AEs (22%) compared to males (13%). Remission was reported as a reason for dis - continuation in 2% of patients. Both sexes discontinued therapy for other unspecified reasons at an equal rate, including for example non-compliance, patient decision, and other medical interventions unrelated to the rheu - matic disease. Serious and non-serious adverse events In total, 171 events occurred in 63 patients for whom the reason to discontinue was reported as “(serious) adverse event” (Table 4). The overall number of events was 127 in females and 44 in male patients. Out of 127 events, 10 were classified as SAE and 117 as AE in females. In male patients, 11 were SAEs and 33 were AEs. For females 91 and for males 30 events could either be related to a causal or time-based relationship. A total of fifty events could not be identified as causal or time-based relationship and were excluded from further analysis. Out of all treatment related events, more events were reported as serious in males (20%) than in females (8%). In Table  5, the treatment related events are further classified. The most common events were infections, neurologic events, and other events, which were not specifically categorized. A list of preferred terms (within the MEdDRA hierarchy) can be found in the supplement (Supplementary Table S2). In fifteen females (14 had AEs and 1 had a SAE) twenty- two infections (21 AEs and 1 SAE) and in 3 males (2 had AEs and 1 had a SAE) four infections (3 AEs and 1 SAE) occurred. The most common infections were influenza, oral herpes, nasopharyngitis, and cystitis. One serious flare event occurred in one female. One serious cardio - logic event was reported for one male patient, which was an intracardiac thrombus. Neurologic/psychiatric events were reported in nine females (12 AEs) and three males (3 AEs) but no serious adverse events occurred in this category. Neurologic/psychiatric events include, among Page 5 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 Table 1 Sex-Stratified Patient Characteristics in RABBIT-SpA Patient Characteristics Females (n = 457) Males (n = 343) Total (n = 800) Missings In % Sociodemographic Factors Age, years, Mean (SD) 52.3 (12.7) 49.9 (13) 51.3 (12.9) 0 BMI, kg/m2, Mean (SD) 28.8 (6.5) 28.4 (4.7) 28.7 (5.8) 1 BMI > 30, kg/m2, Yes, n (%) 166 (37) 109 (32) 275 (35) Current Smoker, Yes, n (%) 144 (36) 71 (24) 215 (31) 14 Physician-Reported Measure Disease Duration, years Joints, Mean (SD) 4.9 (7.0) 4.5 (6.3) 4.7 (6.7) 10 Skin, Mean (SD) 12.9 (14.5) 12.8 (13.8) 12.9 (14.2) 33 Diagnostic Delay, years Joints, Mean (SD) 3.1 (6.0) 2.1 (4.4) 2.6 (5.4) 10 Skin, Mean (SD) 3.5 (8.4) 2.2 (6.3) 2.9 (7.6) 36 Dactylitis, Yes, n (%) 76 (17) 70 (21) 146 (18) 1 Axial Involvement, Yes, n (%) 89 (20) 63 (19) 152 (19) 1 Nail Psoriasis, Yes, n (%) 152 (34) 160 (47) 312 (39) 1 BSA, Mean (SD) 6.1 (11.6) 10.7 (16) 8 (13.8) 3 CRP , mg/L 6 Mean (SD) 6.3 (10.2) 7 (12.5) 6.6 (11.2) ≥ 5 mg/l, n (%) 162 (38) 128 (39) 290 (39) HLA-B27, Positive, n (%) 45 (16) 46 (19) 91 (17) 35 Enthesitis, Yes, n (%) 122 (27) 69 (20) 191 (24) 1 Mean (SD) 0.8 (2.0) 0.5 (1.3) 0.7 (1.7) Tender Joints (TJC-68), Yes, n (%) 387 (85) 270 (79) 657 (82) 0 Mean (SD) 6.8 (7.2) 5.3 (7.0) 6.1 (7.1) No Affected Joints, n (%) 69 (15) 72 (21) 141 (18) ≥ 5 affected joints, n (%) 226 (50) 122 (36) 348 (44) Swollen Joints (SJC-66), Yes, n (%) 315 (69) 216 (63) 531 (67) 0 Mean (SD) 3.3 (4.3) 2.8 (3.9) 3.1 (4.2) No affected joints, n (%) 141 (31) 126 (37) 267 (33) ≥ 5 affected joints, n (%) 107 (23) 67 (20) 174 (22) Physician Global Disease Activity*, Mean (SD) 5.1 (1.9) 4.7 (2) 4.9 (2) 2 DAPSA, Mean (SD) 22.9 (12.7) 19 (12.3) 21.2 (12.6) 16 REM (0–4), n (%) 11 (3) 18 (6) 29 (4) LDA (5–14), n (%) 82 (21) 93 (32) 175 (26) MoDA (15–28), n (%) 182 (48) 135 (46) 317 (47) HDA (> 28), n (%) 108 (28) 46 (16) 154 (23) DAS-28-CRP , Mean (SD) 3.6 (1.1) 3.2 (1.1) 3.4 (1.1) 16 No. of comorbidities (0–47), Mean (SD) 1.9 (2.0) 1.7 (2.2) 1.8 (2.1) ≥ 3 comorbidities, n (%) 138 (30) 78 (23) 216 (27) RDCI (0–9), Mean (SD) 0.9 (1.1) 0.7 (1.0) 0.8 (1.1) 0 Depression as comorbidity, n (%) 61 (13) 19 (6) 80 (10) Fibromyalgia as comorbidity, n (%) 23 (5) 2 (1) 25 (3) Patient-Reported Outcomes Patient Global*, Mean (SD) 5.8 (2.3) 5.0 (2.5) 5.5 (2.5) 10 Patient Pain*, Mean (SD) 5.8 (2.3) 4.7 (2.5) 5.3 (2.4) 10 Patient Sleep Disturbance*, Mean (SD) 5.5 (3) 3.9 (2.9) 4.8 (3) 12 WHO5 (0-100), Mean (SD) 42.0 (22.1) 52.1 (22.3) 46.3 (22.8) 13 Moderate/Severe (< 29), n (%) 133 (33) 56 (19) 189 (27) DLQI, Mean (SD) 5.2 (6.2) 5.1 (5.7) 5.1 (6) 13 HAQ, Mean (SD) 1.0 (0.6) 0.6 (0.6) 0.9 (0.6) 11 BSA: Body surface area; DAPSA: Disease Activity in Psoriatic Arthritis; REM: Remission; LDA: Low Disease Activity; MoDA: Moderate Disease Activity, HDA: High Disease Activity; RDCI: Rheumatic Disease Comorbidity Index; WHO-5: World Health Organization-Five Well-Being Index; DLQI: Dermatology Life Quality Index; HAQ: Health Assessment Questionnaire. *NRS-Scale (0–10) Page 6 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 others, depression, headache, dizziness, and mood swings. One female had a haematologic AE, which was leucopenia. A benign neoplasia event was documented in three females. These were pulmonary mass, papilloma and uterine leiomyoma. The most frequently documented events were labeled as “other events” , and included for example allergic condi- tions, diarrhea, skin disorders, alopecia, other musculo - skeletal disorders, and pruritus.

This study investigated the differences between female and male patients with PsA, regarding variations in dis - ease characteristics, therapy persistence, and reasons for discontinuation. Furthermore, we placed a special focus on sex differences in treatment-related adverse events in a real-world setting. We analyzed 800 patients starting their first b/tsD - MARD therapy. We had a higher number of female patients in the cohort, and they were, on average, two years older than males. Females had a one year longer diagnostic delay, exhibited worse health-related out - comes, and showed higher musculoskeletal disease activ - ity at the start of treatment while males presented more severe skin involvement. These findings align with previ - ously presented data from systematic reviews on obser - vational studies and randomized clinical trials [ 3, 23]. While males had a higher mean BSA affected by psoria - sis, the disease burden, as measured by the DLQI, was similar for both sexes, which is consistent with the find - ings of a scoping review by dermatologists [24]. TNFi was the most frequently prescribed b/tsDMARD among both sexes, followed by IL17i, which were more commonly prescribed to males, while females more often received IL23i. As IL17i are particularly effective in treating skin manifestations [ 25], the more frequent prescription of IL17i in males may be due to the greater skin involvement at start of the first line therapy in our cohort. Other modes of action were comparable in fre - quency. The frequent use of glucocorticoids, even more common in females than in males, does not correspond to current treatment recommendations [26]. Regarding treatment, females had lower persistence rates for both TNFi and IL17i in our study. This corre - sponds to results from the Danish DANBIO register, where males also remained on TNFi first-line therapy for a longer duration [ 27]. Another recent study analyzing pooled data from 13 European registries within the Euro- SpA collaboration found that females were more likely to discontinue TNFi therapy within a two-year period with a higher likelihood of discontinuation due to insuf - ficient efficacy or side effects [ 10]. Further supporting this, an Italian cohort study observed that females were more likely to discontinue anti-TNFi treatments earlier than males, although smaller sex differences were noted for IL17i and IL12/23i therapies. However, the smaller sample sizes in these groups limit the generalizability of these findings [ 28]. Similarly, a multinational prospec - tive cohort study by Van Kuijl et al. demonstrated that females who started ustekinumab, regardless of the ther - apy line, showed lower persistence [29]. A trend observed across multiple studies suggests that females experience higher rates of discontinuation due to side effects, particularly with TNFi and IL17i therapies. This observation is further supported by a meta-analy - sis of randomized controlled trials (RCTs), which found lower response rates and higher discontinuation rates when treated with TNFi and IL17i in females. On the other hand, for Janus kinase inhibitors (JAKi), the effi - cacy between the sexes was similar, but females experi - enced a higher incidence of adverse events, which could contribute to their lower therapy persistence [23]. These sex-based differences in therapy persistence may be further elucidated by examining the reason for therapy discontinuation, where we found notable differences. While the most common reason for discontinuation among both sexes was lack of efficacy, interestingly, males were three times more likely to discontinue therapy due to remission, further supporting the findings related to treatment response and effectiveness. Females, on the other hand, were more likely to discontinue treatment because of safety events. Our safety data showed that, during the initial b/tsDMARD intervention, females had more safety events overall, but SAEs were reported twice as often in males. This finding aligns with a worldwide Table 2 First-Line Therapies and Concomitant Therapies at Treatment Initiation in Females and Males with PsA Females (n = 457) Males (n = 343) Total (n = 800) First-Line Therapy, n (%) First-Line TNFi 247 (54) 197 (57) 444 (56) First-Line IL17i 119 (26) 107 (31) 226 (28) First-Line Other Modes of Action 91 (20) 39 (11) 130 (16) IL23i 13 (14) 3 (8) 16 (12) IL12i/IL23i 13 (14) 9 (23) 22 (17) IL6i 1 (1) 0 (0) 1 (1) JAKi 19 (21) 8 (21) 27 (21) PDE4-Inhibitor 44 (48) 19 (49) 63 (48) T-Cell Costimulation Inhibitors 1 (1) 0 (0) 1 (1) Concomitant Therapies, n (%) Current Glucocorticoids, n (%) 156 (34) 99 (29) 255 (32) Glucocorticoid Dose, mean (SD) 7 (4.8) 8.8 (7.2) 7.7 (5.9) Current Non-Opioid Analgesics Therapy 87 (26) 47 (19) 134 (23) Current Opioid Therapy 30 (11) 17 (8) 47 (9) Current NSAID Therapy 194 (43) 119 (35) 313 (39) Current csDMARD Therapy 197 (43) 154 (45) 351 (44) Page 7 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 Fig. 1 Therapy Persistence Rates per Group Across All Therapies, TNFi and IL17i Page 8 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 study on differences in reporting drug events in which male reports were more frequently classified as SAE [ 30]. The reasons are mainly unclear and could be related to actual physiological differences in therapy response or gendered reporting practices and healthcare-seeking behaviors. For example, women utilize healthcare pro - gram offerings more regularly [ 31]. They also tend to report more events than men, which can possibly lead to an earlier detection and intervention on these potential serious impairments. As expected, the most common treatment-related events overall were infections, neurologic/psychiatric and other events, with more non-serious events reported in females. Most of the infectious events were attribut - able to flu-like infections. In a study on patients with psoriasis receiving bDMARDs, females had significantly more fungal and herpes simplex infections compared to males [32]. In this study, the authors report that females experienced more side effects and had a lower overall sat- isfaction rate, which they suggest may explain the lower therapy persistence rates observed in females. A post- hoc analysis of a phase 3 trial with tofacitinib showed similar proportions among both sexes who experienced AEs. However, females treated for up to 12 months had a higher incidence of SAEs [ 23]. Gosselt et al. also found that compared to men, women report more and a wider variety of adverse drug reactions (ADR) in patients receiving adalimumab and etanercept [ 33]. They also Table 3 Reasons for Therapy Discontinuation as Recorded by the Treating Physician Females (n = 218) Males (n = 110) Total (n = 328) Reason for Therapy Discontinuation (Serious) Adverse Events, n (%) 49 (22) 14 (13) 63 (19) Lack of Efficacy, n (%) 83 (38) 45 (41) 128 (39) Remission, n (%) 1 (0) 4 (4) 5 (2) Other Reasona, n (%) 12 (6) 5 (5) 17 (5) Unknown (%) 73 (33) 42 (38) 115 (35) ae.g. non-compliance, patient decision, other medical interventions Table 4 Identification of Events as Treatment-Related Reasons for Therapy Discontinuation Females Males Total Patients Who Discontinued Due to an Event and Had (S)AE Reported 49 14 63 Identification of Events as Treatment-Related Reasons for Discontinuation Overall No. of Events Occurred, n 127 44 171 Causal relationship to therapya 61 (48) 17 (39) 78 (45) Time-based relationshipb 30 (24) 13 (29) 43 (25) Event out of range or missingc 36 (28) 14 (32) 50 (30) No. of Treatment-Related Events, n 91 30 121 Thereof SAEs 7 (8) 6 (20) 13 (11) Thereof AEs 84 (92) 24 (80) 108 (89) All data presented as n (%) aIndicated by physician as ‘definite,’ ‘probable,’ or ‘possible’ bEvents occurring no more than 90 days before or 14 days after the date of therapy discontinuation where eligible for selection and therefore matching to the therapy episode cIf an event could not be classified as either having a causal or time-based relationship to the first-line therapy or was missing, it was considered out of range and excluded from further analysis Table 5 Treatment-Related Serious (SAE) and Non-Serious Adverse Events (AE) Females (n = 49) Males (n = 14) n event / n patient* Non-serious adverse events (AE) Serious adverse events (SAE) Non-serious adverse events (AE) Serious adverse events (SAE) Infection 21/14 1/1 3/2 1/1 Flare 0/0 1/1 0/0 0/0 Cardiologic Events 0/0 0/0 0/0 1/1 Neurologic/Psychiatric Events 12/9 0/0 3/3 0/0 Hematologic Events 1/1 0/0 0/0 0/0 Neoplasia 3/3 0/0 0/0 0/0 Other Events 47/29 5/3 18/11 4/2 Total 84/56 7/5 24/16 6/4 *n event / n patient: represents the ratio of the number of events that occurred and the number of patients who experienced an event. Patients may be represented in multiple categories Page 9 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 examined the burden of ADR and found that there is no difference in reported burden between women and men. This could indicate, that although women are more likely to report events, they do not necessarily experience them in a more severe or burdensome way than men, which might simply reflect more frequent reporting rather than actual differences in severity or overall experience of ADR. Our findings underscore significant sex-specific differ - ences among patients with PsA in terms of baseline char- acteristics, treatment outcomes, and persistence with therapy. Moreover, we provide novel insights into the underlying reasons for treatment discontinuation, reveal- ing distinct patterns between female and male patients. Additionally, our analysis highlights sex-based disparities in the nature and incidence of adverse events. Strengths and limitations A strength of our study is the large prospective, observa - tional, real-world data source providing a close monitor - ing of efficacy and safety under all approved b/tsDMARD strategies within predefined follow up intervals. The sample size enabled sex-stratified analyses on reasons for therapy discontinuation, along with detailed information on adverse events.

Due to the observational nature of the registry, causal inferences cannot be drawn from our findings. As we only capture sex as a simple variable, we cannot separate sex and gender but recognize that both biological processes and gender roles likely influence outcomes. Due to the design of the questionnaires, we could not determine the exact event that led to therapy discontinu - ation, but had to make assumptions and assigned the events primarily via temporal context. In 35% of therapy discontinuations, the reason for discontinuation was not documented. This proportion of missing information may have influenced the observed distribution of discon - tinuation causes between sexes and should be considered when interpreting these results. While we adjusted only for age to retain the total observed effect of sex, other potential confounders (e.g., comorbidities, smoking, con- comitant drugs) were not included in the models. Thus, residual confounding may contribute to the observed sex differences and our findings should be interpreted accordingly. Differences in reporting behavior between sexes may have influenced our findings. In particular, females may be more likely to report less severe AE, potentially inflating AE-related discontinuation rates compared to males. This possible reporting bias could partly explain the observed sex differences and should be considered when interpreting the results.

This study observed sex differences in PsA, with women showing higher baseline disease activity and muscu - loskeletal burden, but lower therapy persistence, par - ticularly with TNFi and IL17i therapies. Male patients were more likely to discontinue treatment due to remis - sion, while females discontinued more often due to side effects, highlighting a key difference in treatment-related adverse events. These findings emphasize the need for sex-specific strategies in managing PsA to improve therapeutic out - comes and patient satisfaction. More nuanced research into both biological and socio - cultural factors influencing therapy outcomes in PsA is needed, especially in real-world settings. Observational studies provide valuable insights but the lack of stan - dardized reporting of sex data limits the ability to draw definitive conclusions. Capturing sex-related variables across all types of studies is an important future direction to broaden our knowledge and deepen our understand - ing of the underlying factors contributing to observed disparities. Supplementary Information The online version contains supplementary material available at h t t p s : / / d o i . o r g / 1 0 . 1 1 8 6 / s 1 3 0 7 5 - 0 2 5 - 0 3 6 5 0 - 4. Supplementary Material 1

The authors acknowledge the invaluable contributions of all participating consultant rheumatologists and their patients. In particular, we would like to thank those rheumatologists who enrolled the highest numbers of patients. Author contributions LL: Lisa Lindner; AW: Anja Weiß; AR: Andreas Reich; CB: Christine Baumann; FB: Frank Behrens; XB: Xenofon Baraliakos; ACR: Anne C. Regierer; N/A: Not applicable. Term; Definition; Author(s)• Conceptualization; Ideas; formulation or evolution of overarching research goals and aims; LL, ACR, AW, AR•Methodology; Development or design of methodology; creation of models; LL, AW, AR• Software; Programming, software development; designing computer programs; implementation of the computer code and supporting algorithms; testing of existing code components; LL, AW, AR• Validation; Verification, whether as a part of the activity or separate, of the overall replication/ reproducibility of results/experiments and other research outputs; LL, AR• Formal analysis; Application of statistical, mathematical, computational, or other formal techniques to analyze or synthesize study data; LL, AW, AR• Investigation; Conducting a research and investigation process, specifically performing the experiments, or data/evidence collection; LL, AW, AR, ACR• Resources; Provision of study materials, reagents, materials, patients, laboratory samples, animals, instrumentation, computing resources, or other analysis tools; N/A• Data Curation; Management activities to annotate (produce metadata), scrub data and maintain research data (including software code, where it is necessary for interpreting the data itself ) for initial use and later reuse; LL, AW• Writing - Original Draft; Preparation, creation and/or presentation of the published work, specifically writing the initial draft (including substantive translation); LL, ACR• Writing - Review & Editing; Preparation, creation and/or presentation of the published work by those from the original research group, specifically critical review, commentary or revision – including pre-or postpublication stages; LL, AW, AR, CB, FB, XB, ACR• Visualization; Preparation, creation and/or presentation of the published work, specifically visualization/ data presentation; LL• Supervision; Oversight Page 10 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 and leadership responsibility for the research activity planning and execution, including mentorship external to the core team; FB, XB•Project administration; Management and coordination responsibility for the research activity planning and execution; ACR• Funding acquisition; Acquisition of the financial support for the project leading to this publication; N/A. Funding Open Access funding enabled and organized by Projekt DEAL. RABBIT-SpA is supported by a joint, unconditional grant from AbbVie, Amgen, Biocon Biologics, Biogen, Celltrion, Johnson & Johnson, Lilly, Novartis, Pfizer, and UCB. Data availability The data that support the findings of this study are available from German Rheumatology Research Center but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the German Rheumatology Research Center. Declarations Ethics approval and consent to participate RABBIT-SpA is approved by the Ethics Committee of Charité University Medicine, Berlin (#EA1/246/16). Participants have to consent to participate in the study. Consent for publication Not applicable. Competing interests LL: The author declares that they have no competing interests. AW: The author declares that they have no competing interests. AR: The author declares that they have no competing interests. CB: The author declares that they have no competing interests. FB: grants/research support from Bionorica, Bristol Myers Squibb, Chugai, Iron4u, Janssen-Cilag, LEO Pharma, Novartis, Pfizer, and Roche andmeeting support, honoraria, or fees for serving as a speaker, consultant, and/or advisory board member from AbbVie, Affibody, Amgen, Bristol Myers Squibb, Boehringer Ingelheim, Eli Lilly, Galapagos, GSK, Janssen- Cilag, MoonLake, Merck Sharp & Dohme, Novartis, Pfizer, Sandoz, Sanofi , and UCB. XB: Consultant, Scientific Advisory Board: Abbvie, Advanz, Alexion, Alphasigma, Amgen, AstraZeneca, BMS, Cesas, Celltrion, Clarivate, Galapagos, J&J, Lilly, Moonlake, Novartis, Peervoice, Pfizer, Roche, Sandoz, Springer, Stada, Takeda, UCB, Zuellig. Research Grants: Abbvie, Celltrion, Janssen, Moonlake, Novartis. Non-commercial disclosures: ASAS Past President, EULAR President- Elect. ACR: Speaker fees from Amgen, BMS, Novartis. Author details 1German Rheumatology Research Center (DRFZ Berlin), Epidemiology and Health Services Research, Research Charitéplatz 1, 10117 Berlin, Germany 2Private Practice for Rheumatology, Plauen, Germany 3Fraunhofer Institute for Translational Medicine and Pharmacology ITMP , Frankfurt am Main, Germany 4Rheumatology Center Ruhrgebiet, Ruhr University Bochum, Herne, Germany Received: 9 July 2025 / Accepted: 11 September 2025

1. Ogdie A, Langan S, Love T, Haynes K, Shin D, Seminara N, et al. Prevalence and treatment patterns of psoriatic arthritis in the UK. Rheumatology (Oxford). 2013;52(3):568–75. 2. Bragazzi NL, Bridgewood C, Watad A, Damiani G, McGonagle D. Sex-Based medicine Meets psoriatic arthritis: lessons learned and to learn. Front Immu- nol. 2022;13:849560. 3. Coates LC, van der Horst-Bruinsma IE, Lubrano E, Beaver S, Drane E, Ufuktepe B, et al. Sex-Specific differences in patients with psoriatic arthritis: A system- atic review. J Rheumatol. 2023;50(4):488–96. 4. Tarannum S, Leung YY, Johnson SR, Widdifield J, Strand V, Rochon P , et al. Sex- and gender-related differences in psoriatic arthritis. Nat Rev Rheumatol. 2022;18(9):513–26. 5. Eder L, Thavaneswaran A, Chandran V, Gladman DD. Gender difference in disease expression, radiographic damage and disability among patients with psoriatic arthritis. Ann Rheum Dis. 2013;72(4):578–82. 6. Lubrano E, Scriffignano S, Fatica M, Triggianese P , Conigliaro P , Perrotta FM, et al. Psoriatic arthritis in males and females: differences and similarities. Rheumatol Ther. 2023;10(3):589–99. 7. Passia E, Vis M, Coates LC, Soni A, Tchetverikov I, Gerards AH, et al. Sex-specific differences and how to handle them in early psoriatic arthritis. Arthritis Res Ther. 2022;24(1):22. 8. Eder L, Gladman DD, Mease P , Pollock RA, Luna R, Aydin SZ et al. Sex differ- ences in the efficacy, safety and persistence of patients with psoriatic arthritis treated with tofacitinib: a post-hoc analysis of phase 3 trials and long-term extension. RMD Open. 2023;9(1). 9. Hellamand P , van de Sande MGH, Ørnbjerg LM, Klausch T, Eklund KK, Relas H et al. Sex differences in the effectiveness of first-line tumor necrosis factor inhibitors in psoriatic arthritis; results from the EuroSpA Research Collabora- tion Network. Arthritis Rheumatol. 2023. 10. Saad AA, Ashcroft DM, Watson KD, Hyrich KL, Noyce PR, Symmons DP . Persis- tence with anti-tumour necrosis factor therapies in patients with psoriatic arthritis: observational study from the British society of rheumatology biolog- ics register. Arthritis Res Ther. 2009;11(2):R52. 11. Joven B, Manteca CF, Rubio E, Raya E, Pérez A, Hernández R, et al. Real- World persistence and treatment patterns in patients with psoriatic arthritis treated with Anti-IL17 therapy in spain: the PerfIL-17 study. Adv Ther. 2023;40(12):5415–31. 12. Koehm M, Foldenauer AC, Rossmanith T, Kellner H, Kiltz U, Burmester GR et al. Differences in treatment response between female and male psoriatic arthritis patients during IL-12/23 therapy with or without methotrexate: post hoc analysis of results from the randomised MUST trial. RMD Open. 2023;9(4). 13. Regierer AC, Weiß A, Baraliakos X, Zink A, Listing J, Strangfeld A. [RABBIT-SpA: a new disease register for axial spondyloarthritis and psoriatic arthritis]. Z Rheumatol. 2020;79(2):135–42. 14. Lindner L, Weiß A, Reich A, Kindler S, Behrens F, Braun J, et al. Implementing an automated monitoring process in a digital, longitudinal observational cohort study. Arthritis Res Ther. 2021;23(1):181. 15. Heidari S, Babor TF, De Castro P , Tort S, Curno M. Sex and gender equity in research: rationale for the SAGER guidelines and recommended use. Res Integr Peer Rev. 2016;1:2. 16. Schoels MM, Smolen JS, Aletaha D. To DAPSA or not to DAPSA? That is not the question. Ann Rheum Dis. 2019;78(7):e61. 17. England BR, Sayles H, Mikuls TR, Johnson DS, Michaud K. Validation of the rheumatic disease comorbidity index. Arthritis Care Res (Hoboken). 2015;67(6):865–72. 18. Stouten V, Pazmino S, Verschueren P , Mamouris P , Westhovens R, de Vlam K et al. Comorbidity burden in the first three years after diagnosis in patients with rheumatoid arthritis, psoriatic arthritis or spondyloarthritis: a general practice registry-based study. RMD Open. 2021;7(2). 19. Topp CW, Østergaard SD, Søndergaard S, Bech P . The WHO-5 Well-Being index: a systematic review of the literature. Psychother Psychosom. 2015;84(3):167–76. 20. Reich A, Weiß A, Lindner L, Baraliakos X, Poddubnyy D, Zinke S, et al. Depres- sive symptoms are associated with fatigue, poorer functional status and less engagement in sports in AxSpA and psa: an analysis from the RABBIT-SpA cohort. Arthritis Res Ther. 2023;25(1):136. 21. Tönnies T, Schlesinger S, Lang A, Kuss O. Mediation analysis in medical research. Dtsch Arztebl Int. 2023;120(41):681–7. 22. Miles JN, Kulesza M, Ewing B, Shih RA, Tucker JS, D’Amico EJ. Moderated mediation analysis: an illustration using the association of gender with delin- quency and mental health. J Crim Psychol. 2015;5(2):99–123. 23. Eder L, Mylvaganam S, Pardo Pardo J, Petkovic J, Strand V, Mease P , et al. Sex-related differences in patient characteristics, and efficacy and safety of advanced therapies in randomised clinical trials in psoriatic arthritis: a systematic literature review and meta-analysis. Lancet Rheumatol. 2023;5(12):e716–27. 24. Gonzalez-Cantero A, Constantin MM, Dattola A, Hillary T, Kleyn E, Magnolo N. Gender perspective in psoriasis: a scoping review and proposal of strategies for improved clinical practice by European dermatologists. Int J Womens Dermatol. 2023;9(4):e112. Page 11 of 11 Lindner et al. Arthritis Research & Therapy (2025) 27:188 25. Blauvelt A, Chiricozzi A. The Immunologic role of IL-17 in psoriasis and psori- atic arthritis pathogenesis. Clin Rev Allergy Immunol. 2018;55(3):379–90. 26. Gossec L, Kerschbaumer A, Ferreira RJO, Aletaha D, Baraliakos X, Bertheus- sen H et al. EULAR recommendations for the management of psoriatic arthritis with Pharmacological therapies: 2023 update. Ann Rheum Dis. 2024:ard–2024. 27. Højgaard P , Ballegaard C, Cordtz R, Zobbe K, Clausen M, Glintborg B, et al. Gender differences in biologic treatment outcomes-a study of 1750 patients with psoriatic arthritis using Danish health care registers. Rheumatology (Oxford). 2018;57(9):1651–60. 28. Navarini L, Costa L, Tasso M, Chimenti MS, Currado D, Fonti GL, et al. Retention rates and identification of factors associated with anti-TNFα, anti-IL17, and anti-IL12/23R agents discontinuation in psoriatic arthritis patients: results from a real-world clinical setting. Clin Rheumatol. 2020;39(9):2663–70. 29. Van Kuijk AWR, Nurmohamed MT, Siebert S, Bergmans P , de Vlam K, Gremese E, et al. Gender-specific differences in patients with psoriatic arthritis receiving ustekinumab or tumour necrosis factor inhibitor: real-world data. Rheumatology (Oxford). 2023;62(10):3382–90. 30. Watson S, Caster O, Rochon PA, den Ruijter H. Reported adverse drug reactions in women and men: aggregated evidence from globally collected individual case reports during half a century. EClinicalMedicine. 2019;17:100188. 31. Thompson AE, Anisimowicz Y, Miedema B, Hogg W, Wodchis WP , Aubrey- Bassler K. The influence of gender and other patient characteristics on health care-seeking behaviour: a QUALICOPC study. BMC Fam Pract. 2016;17:38. 32. van der Schoot LS, van den Reek J, Groenewoud JMM, Otero ME, Njoo MD, Ossenkoppele PM, et al. Female patients are less satisfied with biological treatment for psoriasis and experience more side-effects than male patients:

from the prospective biocapture registry. J Eur Acad Dermatol Vene- reol. 2019;33(10):1913–20. 33. Gosselt HR, van Lint JA, Kosse LJ, Spuls PI, Vonkeman HE, Tas SW, et al. Sex differences in adverse drug reactions from adalimumab and etanercept in patients with inflammatory rheumatic diseases. Expert Opin Drug Saf. 2023;22(6):501–7. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.