Long-Term Effectiveness and Safety of Abrocitinib in Patients with Atopic Dermatitis: A Real-World, Retrospective Study

Abstract

Introduction

Abrocitinib constitutes one of the most effective therapeutic agents for the treatment of atopic dermatitis (AD). However, real-world data validating its effectiveness outside randomized controlled trials remain limited and are primarily based on short-term follow-up studies. Herein, we aimed to summarize our 2-year experience with abrocitinib for AD.

Methods

In this retrospective cohort study, all patients with moderate-to-severe AD who received abrocitinib were included. The following physician- and patient-reported outcomes were collected at baseline and weeks 4, 12, 24, 52, and week 104: Eczema Area and Severity Index (EASI), SCORing Atopic Dermatitis (SCORAD), Investigator Global Assessment (IGA), Worst Pruritus Numeric Rating Scale (WP-NRS), Dermatology Life Quality Index (DLQI), and a stringent combined outcome (EASI90 and WP-NRS 0/1). Long-term safety was evaluated using clinical and laboratory data and patient-reported adverse events (AEs).

Results

A total of 23 patients [15 male patients/8 female patients; mean (SD) age 29.1 (11.5) years] were included. The mean (SD) EASI score significantly decreased from baseline to week 104 [from 25.0 (11.2) to 3.9 (7.7); p < 0.01], while mean (SD) SCORAD decreased from 67.2 (14.1) to 9.7 (19.3) (p < 0.01). Corresponding EASI75, EASI90, EASI100, and IGA 0/1 responses were achieved by 19 (83%), 13 (57%), 12 (52%), and 14 (61%) patients, respectively. Regarding patient-reported outcomes, QoL and itch severity improved significantly up to week 104, with 61% and 83% of patients achieving DLQI < 2 and WP-NRS 0/1, respectively. Three patients (13%) discontinued abrocitinib due to AEs, with the remaining AEs being mild in severity.

Conclusion

This real-world study supports clinical trial findings on the high effectiveness of abrocitinib in the management of AD. Moreover, it confirms that abrocitinib is generally well tolerated and can lead to early and sustained improvement in pruritus and overall QoL. Despite the small sample size, the results of the present real-world study seem to align with previous data on the effectiveness of abrocitinib also in the long-term management of AD, with an acceptable safety profile.

Key Summary Points

Why carry out this study?

To assess the long-term effectiveness and safety of abrocitinib for atopic dermatitis in a real-world setting.

What was learned from the study?

Overall, both physician- and patient-reported outcomes improved throughout the 2-year treatment period for most patients, supporting clinical trial data on the effectiveness of abrocitinib in moderate-to-severe atopic dermatitis.

Abrocitinib was found to be generally well tolerated, leading to early and sustained improvement in pruritus and overall quality of life during the 2-year treatment period.

Introduction

Atopic dermatitis (AD) is a chronic inflammatory skin disease that significantly impacts patients’ quality of life (QoL). Currently, biologics, including dupilumab, tralokinumab, lebrikizumab, and nemolizumab, as well as three Janus kinase (JAK) inhibitors, baricitinib, upadacitinib, and abrocitinib are available for the treatment of moderate-to-severe AD. Among these, abrocitinib, an oral selective JAK1 inhibitor, has demonstrated high effectiveness and safety in both the short and long term, leading to its approval for the treatment of moderate-to-severe AD in adults and adolescents aged 12 years and older [1, 2]. Its strong effectiveness is further reflected in significant improvements in patient-reported outcomes such as the Dermatology Life Quality Index (DLQI), SCORing Atopic Dermatitis (SCORAD), and pruritus assessments [3].

Despite these findings, real-world data validating the effectiveness of abrocitinib outside randomized controlled trials (RCTs) remain limited and are primarily based on short-term follow-up studies [4–6]. Therefore, we aimed to summarize our center’s 2-year experience with abrocitinib.

Methods

Study Design

This retrospective cohort study was approved by the Andreas Syggros Hospital (Athens, Greece) ethics committee (protocol number 3610/20-05-2025, retrospective approval). The study was conducted according to the principles of the Declaration of Helsinki, and all patients provided written informed consent for publication of their data.

Study Population

We included all patients who had been diagnosed with moderate-to-severe AD based on the Hanifin-Rajka criteria [7], who received at least one dose of abrocitinib between June 2022 and August 2023. Moderate-to-severe AD was defined according to validated severity measures, including an Eczema Area and Severity Index (EASI) score ≥ 16 and/or an Investigator’s Global Assessment (IGA) score ≥ 3, consistent with established clinical trial criteria. Exclusion criteria included concomitant systemic immunosuppressive therapy, incomplete baseline data, and absence of at least one follow-up assessment. Abrocitinib was initiated with 200 mg once daily in patients aged ≥ 18 years (21 patients) with subsequent dose tapering as needed and 100 mg once daily in those aged 12–18 years (2 patients). All patients applied topical steroids and calcineurin inhibitors as needed. Daily use of emollients was encouraged.

Clinical Outcomes

Patient baseline characteristics, including demographics, disease duration, comorbidities, and past treatments, were recorded. To assess long-term effectiveness, the following physician- and patient-reported outcomes were collected at baseline and weeks 4, 12, 24, 52, and 104: EASI (range 0–72), SCORAD (range 0–103), IGA (range 0–4), Worst Pruritus Numeric Rating Scale (WP-NRS, range 0–10) over the past 7 days, DLQI (score 0–30), a stringent combined outcome, defined as EASI90 and WP-NRS 0/1. Moreover, the concomitant topical treatment was recorded. To avoid overstating treatment effects due to the small sample size, a last-observation-carried-forward (LOCF) analysis was performed, where the last observed measurement from each patient who discontinued treatment before week 104, either due to treatment inefficacy or adverse events (AEs), was used to fill in all subsequent missing data points [8]. All patients were included in the analysis at each timepoint. Long-term safety was evaluated using clinical and laboratory findings recorded at each visit and patient-reported AEs.

Statistical Analysis

Statistical analysis was performed using SPSS (version 26.0; IBM Corp., Armonk, NY, USA). Categorical variables were expressed as frequencies and percentages, and quantitative variables with normal distribution were expressed as mean ± standard deviation (SD).

Results

A total of 23 patients [15 male/8 female patients; mean (SD) age 29.1 (11.5) years] were included, with their baseline characteristics summarized in Table 1. All patients had previously received conventional systemic immunosuppressants (systemic corticosteroids and/or cyclosporine or methotrexate), while four had received targeted therapies including dupilumab, baricitinib, and upadacitinib. Seventeen patients completed the 2-year follow-up, while six discontinued treatment due to AEs (three patients who discontinued at weeks 8, 12, and 16, respectively), inadequate disease control (two patients who discontinued at weeks 24 and 76, respectively), or self-discontinuation after complete AD resolution (one patient who discontinued at week 4) (Fig. 1).

Table 1.

Patient characteristics

Number of patients	23
Male/female patients, n (%)	15/8 (65/35)
Age (years), mean (SD)	29.1 (11.5)
Disease duration (years), mean (SD)	22 (11.7)
BMI (kg/m2), mean (SD)	24.3 (3.9)
Previous treatment, n (%)
Systemic corticosteroids	23 (100)
Cyclosporine	17 (74)
Methotrexate	3 (13)
Dupilumab	2 (9)
Upadacitinib	3 (13)
Baricitinib	1 (5)

N number, SD standard deviation

Fig. 1.

Patient flowchart with information regarding inclusion, discontinuations, and number of patients reaching the 104-week timepoint

All physician- and patient-reported outcomes are presented in Table 2 and Fig. 2. Regarding short-term effectiveness, the mean (SD) EASI score significantly decreased from baseline to week 12 [25.0 (11.2) to 7.4 (12.3); p < 0.01, confidence interval (CI) 12.6, 21.7], while mean (SD) SCORAD decreased from 67.2 (14.1) to 16.9 (23.2) (p < 0.01. CI 38.1, 58.1). Similar improvements were seen in QoL and itch intensity, as reflected by DLQI and WP-NRS reductions. The EASI response analysis at week 12 revealed 75%, 90%, and 100% reductions (EASI75, EASI90, EASI100) in 15 (65%), 10 (43%), and 8 (35%) patients, respectively. A significant proportion [11 (49%)] achieved an IGA score of 0/1 at week 12. At week 104, mean (SD) EASI and SCORAD scores further decreased to 3.9 (7.7) (p < 0.001, CI 15.2, 25.9) and 9.7 (19.3) (p < 0.001, CI 45.4, 65.3), respectively. Corresponding EASI75, EASI90, EASI100, and IGA 0/1 responses were achieved by 19 (83%), 13 (57%), 12 (52%), and 14 (61%) patients, respectively. As summarized in Table 2, QoL and itch severity improved significantly up to week 104, with 61% and 83% of patients achieving DLQI < 2 and WP-NRS 0/1, respectively.

Table 2.

Clinical outcomes at each follow-up visit

	Baseline	Week 4	Week 12	Week 24	Week 52	Week 104
Physician-reported outcomes (n = 23)
EASI, mean (SD; 95% CIa)	25.0 (11.2)	11.9 (13.2; 8.3, 17.3)	7.4 (12.3; 12.6, 21.7)	4.9 (8.0; 14.4, 24.9)	5.1 (8.3; 14.3, 24.7)	3.9 (7.7; 15.2, 25.9)
EASI75, n (%)	NA	9 (39)	15 (65)	16 (70)	18 (78)	19 (83)
EASI90, n (%)	NA	5 (22)	10 (43)	13 (57)	13 (57)	13 (57)
EASI100, n (%)	NA	5 (22)	8 (35)	10 (43)	10 (43)	12 (52)
SCORAD, mean (SD; 95% CIa)	67.2 (14.1)	26.5 (25.1; 28.6, 49.1)	16.9 (23.2; 38.1, 58.1)	12.1 (19.3; 43.1, 63.1)	12.8 (20.6; 62.3, 11.1)	9.7 (19.3; 45.4, 65.3)
IGA 0, n (%)	0 (0)	3 (13)	8 (35)	10 (43)	10 (43)	12 (52)
IGA 1, n (%)	0 (0)	2 (9)	3 (13)	4 (17)	4 (17)	2 (9)
IGA 0/1, n (%)	0 (0)	5 (22)	11 (49)	14 (61)	14 (61)	14 (61)
IGA 2, n (%)	0 (0)	9 (39)	5 (22)	5 (22)	4 (17)	5 (22)
IGA 3, n (%)	12 (52)	5 (22)	4 (17)	2 (9)	3 (13)	2 (9)
IGA 4, n (%)	11 (48)	4 (17)	3 (13)	2 (9)	2 (9)	2 (9)
Patient-reported outcomes (n = 23)
DLQI, mean (SD, 95% CIa)	19.4 (5.1)	6.3 (7.7; 9.4, 15.5)	4.0 (7.2; 11.5, 17.6)	3.0 (6.5; 12.5, 18.9)	3.8 (7.5; 11.9, 18.1)	2.6 (6.5; 12.9, 19.3)
DLQI < 2, n (%)	0 (0)	7 (30)	10 (43)	13 (57)	13 (57)	14 (61)
WP-NRS, mean (SD, 95% CIa)	8.8 (2.5)	1.9 (2.1; 5.3, 7.9)	1.2 (1.9; 5.9, 8.6)	0.9 (1.8; 6.1, 8.9)	1.4 (2.8; 5.6, 8.5)	0.9 (0.9; 6.2, 9.0)
WP-NRS 0/1, n (%)	0 (0)	11 (48)	16 (70)	19 (83)	18 (78)	19 (83)
EASI90 and WP-NRS 0/1 (%)	0 (0)	5 (22)	10 (43)	13 (57)	13 (57)	13 (57)

BMI body mass index, CI confidence interval, DLQI Dermatology Life Quality Index, EASI Eczema Area Severity Index, IGA Investigator Global Assessment, n number, NA not applicable, SCORAD Scoring Atopic Dermatitis, SD standard deviation, WP-NRS Worst Pruritus Numerical Rating Scale

^a95% confidence intervals for each value refer to the mean difference at each timepoint from baseline

Fig. 2.

Disease assessment scores from baseline to week 104

Sensitivity analyses using a per-protocol population restricted to study completers yielded comparable results, with similar reductions in EASI and SCORAD scores at week 12 [25.0 (11.2) to 6.6 (11.7) and 67.2 (14.1) to 14.6 (19.2), respectively; both p < 0.01], and consistent responder rates (EASI75 75%, EASI90 50%, EASI100 45%; IGA 0/1 50%). At week 104, further improvements were observed (mean EASI 1.6 [2.6]; mean SCORAD 3.8 [6.4]), with high responder rates (EASI75 100%, EASI90 82%, EASI100 58%, IGA 0/1 88%), supporting the robustness of the primary LOCF analysis.

A stringent combined outcome (EASI90 and WP-NRS 0/1) was achieved by 22%, 43%, 57%, 57%, and 57% of patients at weeks 4, 12, 24, 52, and 104, respectively. Regarding concomitant topical therapies, all patients with IGA ≤ 1 reported daily use of emollients and occasional use of topical corticosteroids or calcineurin inhibitors across all timepoints, but this was not systematically recorded and analyzed.

Interestingly, one patient in our study self-discontinued abrocitinib during the second month of treatment due to complete disease clearance (EASI100), which has been maintained with emollients alone. Another patient interrupted treatment at week 52 for a scheduled operation, subsequently experienced AD exacerbation, and again achieved rapid and sustained improvement upon abrocitinib re-initiation, explaining the slight increase in disease scores at week 52. Three patients (13%) discontinued abrocitinib due to AEs, including diarrhoea, recurrent infections, and fivefold increase in liver enzymes. Other AEs included twofold increases in creatine phosphokinase (CPK) in three (13%) patients, cholesterol increase by 10% in two (9%) patients, and acne in two (9%) patients, which, however, were mild in severity and did not necessitate dose modification, further supporting the favorable safety profile of abrocitinib.

Discussion

The present study provides novel long-term data on the effectiveness and safety of abrocitinib in the treatment of moderate-to-severe AD in a real-world setting. Overall, both physician- and patient-reported outcomes showed substantial improvement throughout the 2-year treatment period for most patients, suggesting that abrocitinib is an effective and safe option for the long-term management of AD. Furthermore, high effectiveness rates were already observed at week 4, supporting the rapid onset of action of JAK inhibitors in AD and AD-related pruritus. Compared with the pivotal clinical trials JADE MONO-1 and JADE MONO-2, the IGA 0/1 response rate in our study was slightly higher at week 12 (49% vs. 44% and 38% in JADE MONO-1 and -2, respectively) [9, 10]. Our long-term findings demonstrated effectiveness rates similar to those reported in an integrated analysis of abrocitinib in adolescents with moderate-to-severe AD, in which comparable proportions of patients treated with abrocitinib (200 mg and 100 mg) across five RCTs achieved EASI75 (85%, 83%), EASI90 (62%, 60%), and IGA 0/1 (57%, 57%) at week 112 [1]. Available real-world studies have reported effectiveness data up to week 52 of abrocitinib treatment, showing comparable EASI, IGA and WP-NRS responses to those observed in the present study [5, 6, 11–14]. A recently published systematic review and meta-analysis of real-world evidence demonstrated similar response rates to those reported in clinical trials in the short-term without providing long-term data [15].

The safety analysis in our study revealed no new signals compared with previously published clinical trial data [1, 16, 17]. Similarly, a recent meta-analysis of real-world safety data identified acne (16%), elevated CPK levels (13%), and lipid abnormalities (12%) as the most common AEs [18]. In our cohort, three patients (13%) discontinued abrocitinib due to AEs, including diarrhea, recurrent infections, and elevated liver enzymes.

Previous literature supports the rapid onset of action of JAK inhibitors in AD, leading to early improvement in pruritus, QoL, sleep quality, and daily activity [19, 20]. Minimal disease activity, defined as concurrent EASI90 and WP-NRS 0/1, was achieved by 22% of patients as early as week 4 and 57% by week 24, aligning with data from the DARE study [21].

To date, several studies have demonstrated the effectiveness of JAK inhibitors in patients with inadequate response to biologics, particularly dupilumab [21–24]. However, data on the sequential use of JAK inhibitors in moderate-to-severe AD remain limited [25]. In our cohort, patients previously treated with systemic or targeted agents showed variable responses to abrocitinib. For instance, one patient who had discontinued dupilumab due to lack of effectiveness and upadacitinib due to recurrent diarrhea experienced the same AE with abrocitinib and discontinued it as well. Another patient who developed secondary failure with baricitinib discontinued abrocitinib after 18 months again due to secondary failure, while a third patient previously treated with upadacitinib achieved meaningful improvement under abrocitinib.

The main strength of this study is the long follow-up period of 104 weeks, providing real-world data on the long-term effectiveness of abrocitinib and the structured assessment of effectiveness outcomes at multiple predefined timepoints, consistently applied across all patients and aligned with our routine clinical follow-up practices. However, key limitations include the small sample size, the high discontinuation rate, and the retrospective design. To maintain sample size, increase statistical power, and avoid overstating treatment effects, we implemented a LOCF analysis. Thus, the week-104 data may not fully represent the initial cohort but align with published study and real-world data on safety and effectiveness of abrocitinib. Additionally, patients were allowed to use over-the-counter topical treatments which were not systematically recorded due to the retrospective study design and could possibly influence outcomes. However, their use was not found to differ significantly between patients and likely reflect real-world treatment patterns and patient behavior. Finally, as a result of the retrospective study design, the underreporting of minor AEs may have limited the accuracy of our safety observations.

Conclusion

The results of the present real-world study seem to align with previous data on the effectiveness of abrocitinib in the long-term management of AD, with an acceptable safety profile. However, further long-term, real-world studies are warranted to optimize the therapeutic management of AD and to evaluate the sequential use of JAK inhibitors in patients with difficult-to-treat disease.

Author Contributions

All authors (Aikaterini Tsiogka, Ioannis-Alexios Koumprentziotis, Ileana Afroditi Kleidona, Aristeidis Vaiopoulos, Michail Bakakis, Eleni Hatzidimitriou, Mariana Vasilopoulou, Theodora Douvali, Alexander Stratigos and Stamatios Gregoriou) contributed to the study conception and design. Material preparation, data collection and analysis were performed by Aikaterini Tsiogka, Ioannis-Alexios Koumprentziotis, Ileana-Afroditi Kleidona and Stamatios Gregoriou. The first draft of the manuscript was written by Aikaterini Tsiogka and Ioannis-Alexios Koumprentziotis and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

No funding or sponsorship was received for this study or publication of this article.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of Interest

Stamatios Gregoriou is Editorial Board member of Dermatology and Therapy. However, Stamatios Gregoriou was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decision. Tsiogka Aikaterini, Ioannis-Alexios Koumprentziotis, Ileana Afroditi Kleidona, Aristeidis Vaiopoulos, Michail Bakakis, Eleni Hatzidimitriou, Mariana Vasilopoulou, Theodora Douvali and Alexander Stratigos declare no conflict of interest.

Ethical Approval

This retrospective cohort study was approved by the Andreas Syggros Hospital (Athens, Greece) ethics committee (protocol number 3610/20-05-2025, retrospective approval). The study was conducted according to the principles of the Declaration of Helsinki, and all patients provided written informed consent for publication of their data.