Clinical Report

Accepted Oct 3, 2018; Epub ahead of print Oct 3, 2018

Acta Derm Venereol 2019; 99: XX–XX.

Corr: Hyun-sun Park, SMG-SNU Boramae Medical Center, Boramae-ro-5-gil, Dongjak-gu, Seoul 07061, Korea. E-mail: snuhdm@gmail.com

Treatment of severe alopecia areata is often challenging and unsatisfactory. However, oral tofacitinib, which is approved for treatment of rheumatoid arthritis, has had promising results. This is the first study to compare the treatment outcomes of tofacitinib with other conventional therapies in patients with refractory severe alopecia areata. After 6 months of treatment, patients on tofacitinib had a higher response rate and greater tolerability than those on topical immunotherapy and oral steroid ± immunosuppressant, respectively. This study will help clinicians to better determine treatment options for severe refractory alopecia areata.

Alopecia totalis (AT) and alopecia universalis (AU) are rare, extensive subtypes of alopecia areata (AA). AA is a common autoimmune dermatological disease characterized by non-scarring hairless patches on the scalp, often resulting in significant psychological distress for patients (1–3). The extent of alopecia may predict the degree of psychological distress experienced (2). Hence, given that AA can negatively influence the quality of life of patients, striving for a treatment with greater efficacy and safety is paramount. To date, despite numerous proposed treatment methods, there is no reliable and efficacious therapy for AA. Moreover, extensive forms of AA, such as AT and AU, are known to have a poorer treatment response with greater probability of developing chronic diseases (4).

It has been shown recently that blocking the common signalling pathways downstream of cytokine receptors, particularly Janus kinase (JAK), may reverse AA in mice (5). Based on this information, JAK inhibitors have been proposed as a novel promising therapeutic option for AA. In several previous studies, tofacitinib, one of the JAK inhibitors, has been shown to have promising results in managing severe AA (6–8). However, lack of evidence, safety uncertainties, and high associated costs compared with other conventional methods have limited its application in clinical practice. Thus far, to the best of our knowledge, there is no report comparing the treatment outcomes between tofacitinib and other conventional therapies. Hence, this study retrospectively compared the efficacy and tolerability of oral tofacitinib with conventional oral treatment and topical immunotherapy in treating refractory AT/AU.

This study was approved by the Institutional Review Board of Seoul National University Bundang Hospital (B-1708/415-204) and SMG-SNU Boramae Medical Center (10-2017-24). Electronic medical records and clinical photographs were reviewed for all patients with AT or AU who were treated in a tertiary or a secondary hospital with oral tofacitinib, conventional oral treatment (steroid ± cyclosporine), or diphenylcyclopropenone (DPCP) immunotherapy during a period between August 2005 and August 2017. Inclusion criteria were: (i) ≥ 18 years old; (ii) AT (> 80% scalp hair loss) and AU (> 80% scalp hair loss with total body hair loss); (iii) ≥1 year of current episode duration; (iv) previous treatment failure of at least one conventional oral treatment or immunotherapy; and (v) at least 6-month follow-up period. Clinical and demographic information, including age, sex, age of disease onset, and duration of current episode of disease, were retrieved from the patients’ medical records. Disease severity was assessed with the Severity of Alopecia Tool (SALT) (9). SALT scores were measured using patients’ clinical photographs for those in the conventional oral treatment and DPCP groups; but for those in the tofacitinib group, the SALT scores recorded in the electronic chart were used.

Treatment response was evaluated by calculating the percentage regrowth of scalp hair.

SALT₅₀ was defined as 50% regrowth. The presence or absence of adverse reactions, types of adverse reactions, and permanent discontinuation of drugs due to adverse reactions, were evaluated by reviewing patients’ medical records.

Kruskal–Wallis test and χ² test were used to compare continuous and non-continuous variables among the 3 groups, respectively. If there was a significant difference, pair-wise comparisons using Mann-Whitney U test or χ² test with post hoc Bonferroni correction were performed. All analyses were performed using SPSS for Windows (Version 20, IBM, Chicago, IL, USA). p-values less than 0.05 were considered statistically significant.

Of 450 patients diagnosed as having AT or AU, a total of 74 patients (18 tofacitinib, 26 conventional oral treatment, 30 DPCP) were included in this study. The other 376 patients were excluded because they did not meet the inclusion criteria: treatment with other modalities (n = 11), age under 18 years (n = 43), initial SALT score ≤ 80 (n = 187), < 1 year of current episode duration (n = 57), no previous history of treatment failure (n = 32), or short follow-up duration (n = 46). A portion of patients in the tofacitinib group had been included in our previous study investigating the efficacy and tolerability of oral tofacitinib for Korean patients with moderate to severe AA (7). The demographic features of patients are summarized in Table I. The median age at onset of first episode was 18.5 years in the tofacitinib group, 23.5 years in the conventional oral treatment group, and 25 years in the DPCP group. The median duration of disease was 8 years in the tofacitinib group, 5 years in the conventional oral treatment and DPCP groups. The median initial SALT score was 100, 98.5, and 98.1 in the tofacitinib, conventional oral treatment, and DPCP groups, respectively. The median number of previously failed treatments for the current episode was 2, 1, and 1 in the tofacitinib, conventional oral treatment, and DPCP groups, respectively. Previous treatment methods are also summarized in Table I. Although most of the patients were switched to a different treatment regimen from the previous one, some patients who had long current episode duration were kept in the same treatment regimen with dose adjustment. Baseline characteristics in the tofacitinib group were not significantly different from the other 2 groups except for the number of previously failed treatments.

In the tofacitinib group, all patients received baseline laboratory tests, including complete blood cell count with differential, admission panel (aminotransferase, aspartate aminotransferase, gamma glutamyltransferase, total bilirubin, urea nitrogen, creatinine, calcium, phosphorus, total protein, albumin), lipid panel, interferon-γ release assay for Mycobacterium tuberculosis, and serology (HIV, hepatitis virus, syphilis). Three patients in the tofacitinib group had latent tuberculosis and received isoniazid 300 mg daily throughout the entire treatment period. Oral tofacitinib was prescribed as a monotherapy. Tofacitinib 5 mg was given twice daily to all patients; in the 4 patients without any sign of response, the dose was increased (to 5 mg 3 times a day after 2 months in one patient, to 5 mg 3 times a day after 4 months in 2 patients, and to 5 mg 3 times a day after 2 months, then increased again to 10 mg twice daily after 4 months in one patient). The median total tofacitinib dose was 1,680 mg (range 1,680–2,800 mg).

In the conventional oral treatment group, all patients received baseline laboratory tests, including complete blood cell count with differential, admission panel, lipid panel, and hepatitis virus serology test, before initiating treatment. In most cases, a total of 24 mg methylprednisolone was given per day for the first week; then reduced by 4 mg/day for each subsequent week, until the dose was 4 mg/day, which was maintained for 2 months. The dose was then reduced to 2 mg/day, which was maintained for 3 months. In cases without sufficient response, the initial dosing schedule was re-tried. The median total dose of methylprednisolone was 1,120 mg (840–1,330 g). In 22 patients, oral cyclosporine was combined with methylprednisolone. In 19 of these 22 patients, cyclosporine was combined from the beginning of the treatment, while in the remaining 3 patients it was added 3 months after the start of treatment. Cyclosporine was given 150–300 mg per day and the median total cyclosporine dose was 33,600 mg (25,200–33,600 g).

In the DPCP group, the initial concentration of DPCP used post-sensitization ranged from 0.0001 to 1%, with a median of 0.001%. The highest concentration of DPCP used ranged from 0.001 to 1%, with a median of 0.1%.

Treatment response in each group according to the treatment duration is shown in Fig. 1. In general, treatment response increased over time in all groups. After 6 months, patients in the tofacitinib group showed the highest treatment response (Fig. 1A). SALT₅₀ achievers in the tofacitinib group showed a sharper decline in the SALT score, showing more rapid hair regrowth than the other groups (Fig. 1B).

After 3 months, 9 patients (50%) in the tofacitinib group, 7 patients (26.9%) in the conventional oral treatment group, and one patient (3.6%) in the DPCP group achieved SALT₅₀ (Fig. 2A). There was a significant difference in the proportion of patients who achieved SALT₅₀ among the 3 groups (p = 0.001). It was significantly higher in the tofacitinib group than in the DPCP group (corrected p = 0.000). After 6 months, 8 patients (44.4%) in the tofacitinib group, 9 patients (37.5%) in the conventional oral treatment group, and 3 patients (11.1%) in the DPCP group achieved SALT₅₀ (Fig. 2B). The proportion of patients who achieved SALT₅₀ was significantly different among the 3 groups (p = 0.028). It was significantly higher in the tofacitinib group than in the DPCP group (corrected p = 0.048). However, there was no significant difference in SALT₅₀ rate between the tofacitinib and conventional oral treatment groups at 3 and 6 months. The median value of SALT score change was 34.6 (range 0–80), 34.7 (0–89.2) and 0 (0–53.0) after 3 months; 36.5 (0–91.5), 39.9 (0–91.6), 0 (0–80) after 6 months in the tofacitinib, conventional oral treat-ment, and DPCP groups, respectively. Tofacitinib and conventional oral treatment groups were significantly different with the DPCP group after 3 and 6 months (all corrected p = 0.000).

Adverse reactions in each group are described in Table II. In the tofacitinib group, 6 (33.3%) patients experienced adverse reactions. Most of the adverse reactions were mild or transient, and no-one permanently discontinued treatment due to drug reactions. Meanwhile, 19 patients (73.1%) in the conventional oral treatment group experienced adverse reactions; among them, 6 patients (23.1%) permanently discontinued treatment. The most common adverse reactions were abdominal discomfort and acneiform eruption. In the DPCP group, 10 patients (33.3%) experienced adverse reactions; among them, 3 patients (10.0%) stopped treatment due to severe eczema and urticaria. The proportions of patients who experienced adverse drug reactions were significantly different between the 3 groups (p = 0.018), and the proportion in the tofacitinib group was significantly lower than that in the conventional oral treatment group (corrected p = 0.042).

There have been many proposed treatment modalities for AA; however, none are Food and Drug Administration (FDA)-approved and fully satisfactory. Since its efficacy was first demonstrated in 1952, systemic oral steroid has remained the most widely used method to treat extensive AA (10). Steroid treatment reduces inflammation around the hair follicle, allowing it to return to its normal growth cycle. However, despite its popularity, there is limited evidence to support the efficacy of oral steroids (11–16). The greatest barrier to using steroids is the associated adverse reactions, especially with prolonged treatment. Furthermore, no standard treatment dose and schedule for the use of steroids have been established. Cyclosporine is one of the options for treating chronic and severe AA. However, most previous studies have investigated the combined therapeutic effect of cyclosporine and systemic steroids rather than cyclosporine alone (17–21). Adverse reactions include increased blood pressure, liver toxicity, and acute renal failure. Clinical use of cyclosporine in combination with steroids is advised only in refractory cases, given the severity of systemic adverse effects and lack of evidence (4). Also, DPCP immunotherapy is an option for treating recalcitrant AT/AU (11). The exact therapeutic mechanism of DPCP has not yet been elucidated; however, antigenic competition and decreased production of anti-hair-follicle antibodies are thought to play important roles (22). The success rate of DPCP as a treatment of AT/AU ranges from 22% to 79% (11).

Recent experimental research has elucidated that the natural killer group, 2D-bearing CD8⁺ cytotoxic T lymphocytes, is the key player in the pathogenesis of AA; it has been shown, in a mouse model, to cause upregulation of interleukin-15, produce interferon-γ, and attack the hair follicle (5). JAK inhibitors, as downstream regulators of interleukin-15 and interferon-γ, have been demonstrated to reverse AA. Recently, an open-label clinical trial (23) and retrospective observational studies (6–8) with tofacitinib have revealed promising results for treating AA.

This study included patients with AT/AU with a duration of ≥ 1 year of the current episode and at least one previous failed attempt of conventional oral treatment or immunotherapy, in order to collect refractory cases and to exclude those with acute diffuse and total alopecia, which can resolve spontaneously (24). There was no significant difference in the baseline characteristics between the 3 groups (Table I).

Tofacitinib was significantly more efficacious than DPCP after 3 and 6 months, whereas it did not reveal a significantly higher efficacy than conventional oral treat-ment. Nevertheless, considering that most patients in the tofacitinib group experienced multiple prior treatment failures to both conventional oral treatment and DPCP (Table I), tofacitinib appears to be a promising option for highly refractory cases. In general, the longer the treatment duration, the better response patients showed in all treatment groups (Fig. 1A). Interestingly, tofacitinib demonstrated a unique response pattern (Fig. 1B). Patients who achieved SALT₅₀ showed a sharp decline in the SALT score earlier in the treatment course. In contrast, most of those who failed to achieve SALT₅₀ showed a plateau or gradual decrease in the SALT score. Therefore, we believe that a treatment response during the first 3 months might be an important indicator for further treatment response to tofacitinib. One patient who showed an initial SALT₅₀ response within the first 3 months deteriorated after 6 months, but recovered again after 2 months with a dose increment of 10 mg twice a day.

Tofacitinib showed better tolerability compared with the other treatment groups (Table II). The proportion of patients who experienced adverse drug reactions was significantly lower (33.3%) in the tofacitinib group than in the conventional oral treatment group (73.1%). The proportion of adverse reactions in the conventional oral treatment group might be overestimated because the majority of patients were treated with steroid plus cyclosporine. Both drugs can cause adverse reactions and a combination of these might increase the frequency of adverse reactions compared with monotherapy with each agent. No patients in the tofacitinib group permanently discontinued the medication due to adverse reactions, while 6 patients (23.1%) in the conventional oral treatment group and 3 patients (10%) in the DPCP group did. Although there were no serious adverse reactions in the tofacitinib group, this only reflects the results from a 6-month follow-up period. In a previous study with a median follow-up of 12 months, approximately 38.9% of the patients experienced infections during treatment with tofacitinib (6). Because of the shorter follow-up period in the present study, the risk of infection might be underestimated. JAK inhibitors also may modulate anti-tumour inflammatory responses and increase the risk of malignancy. However, recent comprehensive pooled analysis of multiple clinical trials revealed that a standardized incidence rate of all malignancies of tofacitinib-treated rheumatoid arthritis patients was within the expected range of the general population with rheumatoid arthritis (25). Close monitoring and age-adequate cancer screening may be beneficial for those receiving tofacitinib for an extended period.

The present study has some limitations. First, it is a retrospective study with a small number of patients. Secondly, it is not a head-to-head comparison of treatment modalities. However, considering that AT/AU is a relatively rare disease and that it took more than 10 years to collect data from this number of patients, these results are noteworthy. Thirdly, we did not include a placebo group for proper control. However, since the spontaneous resolution rate of moderate-to-severe AA is low, such possibility in AT/AU is expected to be extremely low (26). Fourthly, the current study had a relatively short observational period. Although the majority of DPCP responders experienced their first regrowth within the first 6 months of treatment (27), DPCP is known to require extended treatment to show sufficient effects (28–31). Certainly, large scale long-term prospective head-to-head comparative randomized controlled trials are required to confirm the results of the current study.

Oral tofacitinib was more efficacious than DPCP immunotherapy and more tolerable than conventional oral treatment, at least during the first 6-month treatment period. To the best of our knowledge, this is the first investigation that comparatively analysed the treatment outcomes of tofacitinib with other conventional therapies in patients with refractory AT and AU. The present study will help clinicians to better determine their treatment options for severe refractory AA and will guide further clinical trials investigating the efficacy and safety of JAK inhibitors for AA.

The authors have no conflicts of interest to declare.