Oral Alitretinoin (9-cis-Retinoic Acid) Therapy for Chronic Hand Dermatitis in Patients Refractory to Standard Therapy: Results of a Randomized, Double-blind, Placebo-Controlled, Multicenter Trial

Objective  To assess the efficacy and safety of oral alitretinoin (9-cis-retinoic acid), 10 mg/d, 20 mg/d, and 40 mg/d, compared with placebo control, in the treatment of chronic hand dermatitis.

Design  Multicenter, randomized, double-blind, placebo-control, prospective trial.

Setting  A total of 43 outpatient clinics in 10 European countries.

Patients  Of 348 patients screened, 319 with moderate or severe refractory chronic hand dermatitis were randomized, in the ratio of 1:1:1:1, to 4 treatment groups and received allocated intervention. Of 75 patients who withdrew, 24 withdrew owing to adverse events.

Interventions  Placebo or 10 mg, 20 mg, or 40 mg of oral alitretinoin (9-cis-retinoic acid) taken once daily for 12 weeks. Safety was assessed for all patients during a follow-up period of 4 weeks, and responders were observed for a follow-up period of 3 months.

Main Outcome Measure  Physician’s global assessment of overall chronic hand dermatitis severity.

Results  Alitretinoin led to a significant and dose-dependent improvement in disease status, with responses in up to 53% of patients, and up to a 70% mean reduction in disease signs and symptoms. Treatment was generally well tolerated, with dose-dependent effects comprising headache, flushing, mucocutaneous events, hyperlipidemia, and decreased hemoglobin and decreased free thyroxin levels. Three months after discontinuation of treatment, the rate of relapse was 26%, independent of dose.

Conclusion  Alitretinoin given at well-tolerated doses induced substantial clearing of chronic hand dermatitis in patients refractory to conventional therapy.

Hand dermatitis is frequently encountered in clinical practice. Epidemiological studies conducted in northern Europe indicate that the overall prevalence of hand dermatitis is 6% to 11% of the population.^1-3 Clinical signs are highly variable but generally include 1 or more classic features of eczema localized to the hands, including erythema, edema, vesiculation/blistering, scaling/hyperkeratinization, fissures, pruritus, and pain.^4,5Hand dermatitis is often associated with atopic dermatitis and irritant or allergic contact dermatitis, but the disease etiology is often multifactorial, and it is rarely possible to identify and entirely remove a single causative factor.⁶

Because it is difficult to identify and avoid causative factors, hand dermatitis often develops into a chronic condition,^5,6 with significant social and economic impact.² Chronic hand dermatitis (CHaD) has been associated with psychological distress and is a frequent cause of temporary or permanent working disability, as well as change in occupation.⁵

Mild cases of CHaD are usually adequately responsive to emollients and topical corticosteroids, but the disease can be debilitating in patients with more severe disease if they are refractory to this standard regimen.⁵ Refractory patients include those who have utterly no response to topical corticosteroids. More commonly, such patients will improve to some degree but with persisting lesions that hamper normal activities, or patients will initially respond but soon relapse despite continued therapy or rapidly rebound following discontinuation of therapy.

Few effective treatment options are available for patients who do not respond adequately to topical corticosteroids and whose disease severely impairs social and professional life. Oral corticosteroids, cyclosporin, acitretin, psoralen–UV-A, or radiotherapy have been reported to be active, but their use is limited, either by adverse effects or because they are inconvenient.^4-6

In a published exploratory study in 38 patients with refractory CHaD, alitretinoin (9-cis-retinoic acid) given orally once daily at 20 mg or 40 mg induced clinically significant responses in 36 cases (95%), including very good response (>80% reduction in measurable disease signs and symptoms) in 21 patients (55%).⁷ Although this trial was not placebo-controlled, such a substantial improvement in most patients with refractory history was a remarkable finding. On this basis, we undertook a controlled, randomized dose-finding study of oral alitretinoin in patients with CHaD refractory to standard therapy, including at least 4 weeks’ use of topical corticosteroids. This article describes the final trial results.

Patients were eligible for enrollment in this study if they were aged 18 to 70 years and diagnosed as having “moderate” or “severe” CHaD (see “Efficacy Assessments”) of at least 3 months’ duration and refractory to standard therapy. Refractory status was defined as no response, or transient response to at least 4 weeks of topical corticosteroids, or intolerance to this regimen. Enrollment was open to patients with all types of chronic hand dermatitis, including hyperkeratotic, dyshidrotic (pompholyx), or fingertip dermatitis. Plantar involvement was neither required nor evaluated. Investigators were required to rule out alternative diagnoses, including infectious dermatoses or palmar psoriasis. Patients were also required to be either male or postmenopausal or surgically sterile female.

Patients were excluded from the study if they had active atopic dermatitis or unequivocal psoriasis not limited to the hands and requiring medicated treatment at the time of enrollment. Patients were also excluded if they had significant abnormalities in liver function (alanine aminotransferase and/or aspartate aminotransferase >150% of the upper limit of normal); triglyceridemia (>250% of the upper limit of normal); cholesterolemia (>150% of the upper limit of normal); a history of psychiatric disorders; active bacterial, fungal, or viral infection of the hands; clinically relevant allergic contact dermatitis of the hands and were unable to avoid exposure to the allergen; or any other skin disease likely to interfere with the conduct of the study. Other exclusion criteria comprised treatment with other investigational drugs within the previous 2 months; phototherapy (UV-B, psoralen–UV-A, or x-rays) or use of systemic corticosteroids, retinoids, or immunosuppressants within the previous 4 weeks; use of systemic ketoconazole, itraconazole, erythromycin, or clarithromycin within the previous 2 weeks; or concomitant use of retinoids (oral or topical) or vitamin supplements containing vitamin A (retinol). Known hypersensitivity to retinoids or to any component of the study drug formulations or known immunosuppression were also exclusion criteria.

The study was performed in accordance with the Declaration of Helsinki and all applicable amendments. All patients gave written informed consent prior to enrollment, and the study was reviewed and approved by the recognized ethics review committee at each individual center.

Eligible patients were randomized to treatment by center, in blocks of 4 without stratification, by use of computer-generated randomization codes provided by the study sponsor (Basilea Pharmaceutica Ltd, Basel, Switzerland) and incorporated into double-blind coded drug packaging. Placebo and active drug (as soft gelatin capsules) and packaging were indistinguishable.

Treatment with placebo or with alitretinoin (BAL4079; Basilea Pharmaceutica Ltd) at 10 mg, 20 mg, or 40 mg was given orally once daily after breakfast for 12 weeks, and no dose reductions were allowed. All patients were given an emollient (Bepanthol hand ointment; F. Hoffmann-La Roche Ltd, Basel) with instructions to apply it as frequently as required. Prohibited concomitant treatments during therapy comprised topical and systemic corticosteroids or other retinoids, any other systemic or topical antieczema therapy, phototherapy (UV-B, psoralen–UV-A, or x-rays), immunosuppressive or cytostatic drugs, vitamin A (retinol) supplements, tetracyclines, itraconazole, clarithromycin, systemic ketoconazole, and erythromycin.

The primary efficacy measure for therapeutic response was the physician’s global assessment (PGA) of overall CHaD severity, performed at screening and at week 12 (or at the last evaluation in case of premature withdrawal). The PGA criteria were defined as follows:

• Clear: no residual visible dermatitis.

•  Almost clear: minimal erythema and/or scaling.

•  Mild: clearly visible signs of dermatitis, with no hyperkeratosis, edema, fissures, or functional impact.

•  Moderate: moderately severe signs of dermatitis, with no edema or fissures, or functional impairment.

•  Severe: marked signs of dermatitis, or edema, fissures, or functional impairment.

Patients rated as clear or almost clear at week 12 (or last evaluation) were regarded as responders. Responding patients were followed up for 3 months for the occurrence of relapse, defined as the need for topical corticosteroids or other antieczema medicated therapy.

Secondary efficacy measures were the total lesion symptom score (TLSS), the patient’s global assessment (PaGA) of improvement, and the extent of disease. The TLSS⁷ was calculated as the sum of scores (0 = absent, 1 = mild, 2 = moderate, 3 = severe) assigned by the physician for the following 7 measures: erythema, edema, vesicles, desquamation, hyperkeratosis, fissures, and pruritus/pain. The PaGA was assigned by the patient as “clearing or almost clearing” (≥90% clearing of disease signs and symptoms compared with baseline), “marked improvement” (≥75% clearing), “moderate improvement” (≥50% clearing), “mild improvement” (≥25% clearing), “no change,” or “worsening.” The extent of disease was estimated by the physician as the total percentage involvement of the palms and dorsum of both hands. The Dermatology Life Quality Index, a widely applicable dermatology instrument previously used for eczematous skin conditions,^8,9 was administered at baseline and the end of treatment in centers where a validated translation in the local language was available.

Adverse events (AEs) were recorded throughout therapy and up to 4 weeks during follow-up. In monitoring AEs, the physician questioned patients at each assessment about the occurrence of unusual changes in mood, appetite, sleeping difficulties, and vision disturbances. Standard hematology, nonfasting blood chemistry, thyroid function, and urinalysis tests were performed before treatment, every 2 to 4 weeks during treatment, and after treatment.

This trial was designed to demonstrate whether at least 1 dose of alitretinoin was more active than placebo in inducing responses after 12 weeks, with response defined as a PGA rating of clear or almost clear. Assuming that the percentage of responders would be 30% in the active group and 10% in the placebo group, a calculated sample size of 62 evaluable patients was expected to provide 80% power to reject the null hypothesis of no difference between active treatment and placebo at an α level of .05. Assuming a dropout rate of 10%, 70 patients per group and a total of 280 patients for the 4 treatment groups were planned.

All statistical tests were 2 sided and based on an α level of .05 and were carried out using SAS (version 8.1; SAS Institute Inc, Cary, NC) with the intention-to-treat population. For the primary efficacy measure, a closed test procedure based on the Cochran-Armitage trend test¹⁰ was used to compare the number of responders (PGA assessment) between treatment groups; comparisons always included the placebo group. For secondary efficacy measures, the percentage of patients rated as “clearing or almost clearing,” based on PaGA, was compared using a continuity-corrected χ² test. The percentage change from baseline in TLSS in the active and placebo groups was compared by use of a closed test procedure based on the Kruskal-Wallis test and always included the placebo group. The percentage change from baseline in extent of disease in the active and placebo groups was compared using the Wilcoxon-Mann-Whitney test. Descriptive statistics were used to analyze Dermatology Life Quality Index results, and safety data included AE frequencies and changes in laboratory test values.

Of 348 patients screened, 319 were enrolled between December 2001 and May 2002 at 43 outpatient clinics in Belgium, Denmark, Finland, France, Germany, Holland, Hungary, Poland, Switzerland, and the United Kingdom. Of these 319 patients, 244 (76%) completed the 12-week treatment course and 75 (24%) withdrew before completion. The most frequent reasons for withdrawal in the placebo and the 10-mg/d, 20-mg/d, and 40-mg/d groups were insufficient therapeutic response (8, 7, 4, and 6 patients, respectively), AEs (4, 5, 5, and 10 patients, respectively), and failure to return (4, 6, 4, and 2 patients, respectively). All enrolled patients received treatment, and all were included in the intent-to-treat analysis of efficacy as well as safety.

Patients in each group had similar demographic and disease characteristics (Table 1). The enrollment of more men than women was consistent with the exclusion of women of childbearing potential. Most patients were diagnosed with hyperkeratotic eczema, and their initial diagnosis of CHaD had been made several years before enrollment. Most patients had experienced a transient response to previous topical therapy, but approximately 21% were reported to have had no response. No definite etiological factors were identified for approximately 45% of patients.

Table 2 presents the results of the PGA of CHaD severity, the PaGA of improvement, and the median percentage change in TLSSs after 12 weeks of therapy or at the time of last treatment.

For the PGA, the closed Cochran-Armitage trend test revealed a significant and dose-dependent effect (P<.001). For the PaGA, the numbers of patients rating their response as “clear or almost clear” were significantly higher with all doses of alitretinoin than with placebo (χ² test, P = .01 for 10 mg/d, P = .002 for 20 mg/d, and P<.001 for 40 mg/d). The TLSS reductions were significantly more marked with all doses of alitretinoin than with placebo (P<.001, overall Kruskal-Wallis test), and this effect was dose dependent (P<.001, Jonckheere-Terpstra trend test).

Correlation analysis revealed good agreement between the primary efficacy end point (PGA) and the secondary efficacy parameters (PaGA) (correlation coefficient 0.80, Kendall τ), and TLSS (correlation coefficient 0.86, Spearman coefficient).

Dose-dependent increases in response rates were observed for all types of CHaD, regardless of whether topical therapy had been wholly ineffective or had induced a transient response or whether disease was moderate or severe at baseline (Table 3).

Extent of disease scores decreased from baseline to week 12 in all groups, and the difference between placebo and the 40-mg group was statistically significant (P = .02; analysis of covariance of ranked data). Dermatology Life Quality Index scores also improved during treatment in all groups, with a general trend to more positive effects with higher drug doses without reaching the α level of statistical significance of.05.

Among the 127 responders in all groups, 117 were followed up for 12 weeks after the end of treatment. Of these 117 responders, 31 (26%) required prescription therapy for their CHaD at some time during this period and were judged to have relapsed. The percentage of relapse during this posttreatment period was similar in all groups.

Table 4 presents all AEs reported for at least 3% of patients in any treatment group. The most frequent AEs were headache and mucocutaneous events, which are considered to be typical AEs of oral retinoids. The higher frequency of AEs in the 40-mg/d group was largely owing to headache. Severe AEs comprised headache in 9 patients (1 in the 10-mg/d group, 3 in the 20-mg/d group, and 5 in the 40-mg/d group) and flushing (2 patients in the 40-mg/d group). No other severe event was reported in more than 1 patient per dose group.

Headache was the most frequent dosing-related event leading to withdrawal, reported for 4 patients in the 40-mg group and 1 patient each in the 10-mg/d and 20-mg/d group. Mucocutaneous AEs, including erythema, exacerbated eczema, rash, pruritus, edema, and photosensitivity, led to the withdrawal of 2 patients in the placebo group, 2 in the 10-mg/d group, 3 in the 20-mg/d group, and 1 in the 40-mg/d group. Except for headache, no particular AE led to more than 1 withdrawal in any group. Two serious AEs were reported. In the 10-mg/d alitretinoin group, 1 patient was hospitalized with erysipelas and erythroderma, which was attributed to exacerbation of dermatitis and superinfection and judged to be possibly related to treatment. One patient in the 20-mg/d alitretinoin group with a history of atopic dermatitis developed a photoallergic reaction, which was judged to be unrelated to treatment.

Results from laboratory tests (nonfasting) indicated that treatment with alitretinoin was associated with increases above the normal range in serum triglyceride, cholesterol, and creatine kinase levels and decreases in hemoglobin and free thyroxine levels (Table 5). For most of these measures, abnormalities were clinically insignificant, but a dose-related increase was observed in the number of patients with clinically significant elevated serum triglyceride level (>407 mg/dL [>4.6 mmol/L]). No medications were routinely used to manage triglyceride levels or other abnormal laboratory findings in this trial. Vital signs and results of physical examinations were clinically unremarkable.

In this study, oral alitretinoin induced clinically significant responses in a high percentage of patients with moderate or severe CHaD refractory to standard topical therapy. Response in this study was defined as the complete or nearly complete disappearance of disease signs and symptoms and was reported for 53% of patients treated with the highest alitretinoin dose of 40 mg/d. Responses were reported for all types of CHaD and in patients with no clinically relevant response to prior standard treatment.

Efficacy was dose-dependent over the range tested. Compared with placebo, all 3 doses led to better PGAs and greater improvement in TLSSs and PaGAs. Approximately 10% to 20% of patients had no improvement during therapy, even at the highest doses given. While this may represent a sector of nonresponsive patients with CHaD, it is also possible that this is partly owing to the enrollment of some patients with other inflammatory skin diseases that may resemble CHaD, such as palmar psoriasis.^6,11

The response rate observed in the placebo group (27%) may be related to factors such as the natural history of the disease or to the placebo effects associated with participation in a clinical trial, as well as changes in climate, occupation, and level of stress, which are expected to influence the course of the disease.^12,13 Because the placebo response rate was lower in patients with severe disease at baseline (15%) or no response to prior therapy (15%), it seems likely that substantial clinical improvement attributable to spontaneous variability in disease may be more frequently encountered in patients with less severe disease.

Treatment was generally well tolerated at all 3 doses. At the lower doses of 10 or 20 mg/d, the safety profile was similar in most ways to that of placebo. At the highest dose of 40 mg/d, AEs comprised headache, flushing, elevations in serum lipid level, slightly decreased hemoglobin level, and decreased free thyroxin level. In view of the nonfasting status of patients, interpretation of laboratory data is difficult, except to note the general tendency for typical retinoid effects such as hyperlipidemia. Much higher doses of oral alitretinoin have been given in previous studies to patients with solid tumors or as preventive anticancer therapy. For example, 34 patients received once-daily oral alitretinoin doses of 5 to 230 mg/m² for 4 weeks,¹⁴ 22 patients received twice-daily doses from 20 to 150 mg/m² per day for at least 4 weeks,¹⁵ 41 patients received once-daily (26 patients) or twice-daily (15 patients) doses up to 140 mg/m² for at least 4 weeks,¹⁶ 53 patients received once-daily doses from 60 to 100 mg/m² for a median of 15.1 weeks,¹⁷ and 74 former smokers received once-daily doses up to 100 mg for 3 months.¹⁸ Dose-limiting toxic effects reported at greater than 100 mg/m² per day were headache and diarrhea. Other common AEs included cheilitis, dry skin, fatigue, and hypertriglyceridemia.

The mechanism of action of alitretinoin in CHaD is unknown. Alitretinoin has been described as a panagonist of retinoid receptors because it activates retinoid X receptors as well as all retinoic acid receptors. Retinoid X receptor–mediated stimulation of T_H2 immune function may prove to be related to alitretinoin activity.^7,19A hypothesis of retinoid × receptor–mediated activity is supported by the activity of bexarotene, a retinoid × receptor agonist, applied topically as a 1% gel twice daily or 3 times daily in patients with severe CHaD.²⁰However, at present, the therapeutic effect of alitretinoin cannot be explicitly linked to a particular pattern of receptor binding and consequent biological effects.⁷ The encouraging results of this trial support further clinical studies of alitretinoin in the treatment of patients with CHaD.

Article

Correspondence: Thomas Ruzicka, MD, Department of Dermatology, Heinrich-Heine University Hospital Dusseldorf, Moorenstr 5, 40221 Dusseldorf, Germany (ruzicka@uni-duesseldorf.de).

Accepted for Publication: July 22, 2004.

Funding/Support: This study was supported and funded by Basilea Pharmaceutica Ltd, Basel, Switzerland.

Previous Presentation: The results of this study were presented in part at the European Academy of Dermatology and Venereology (EADV) Annual Meeting; October 15-18, 2003; Barcelona, Spain.

Acknowledgment: We thank Jill Sauer, BSc (Basilea Pharmaceutica Ltd), for supervision of data management; Abdallah Ennaji, MSc (Hesperion AG, Allschwil, Switzerland), and Barbara Collins, PhD (Inveresk, Edinburgh, United Kingdom), for statistical analysis; and Peter Sonner, PhD (Inveresk, Munich, Germany), for assistance in the coordination and monitoring of the trial.

Financial Disclosure: Drs Ruzicka, Larsen, Galewicz, Horváth, Coenraads, Thestrup-Pedersen, Ortonne, and Zouboulis received grants from Basilea Pharmaceutica for the performance of this study. Drs Ruzicka, Coenraads, Thestrup-Pedersen, and Ortonne have received consultancy fees from Basilea Pharmaceutica. Drs Harsch, Brown, and Zultak are employees of Basilea Pharmaceutica.

Belgium: J. W. Lambert, University Hospital Antwerpen, Edegem; D. Roseeuw, University Hospital Brussels, Brussels; M. F. De La Brassinne, Centre Hospitalier Universitaire de Liège, Liège; M. Heenen, Université Libre de Bruxelles Erasme, Brussels; H. Degreef, U. Z. Sint-Rafaël, Leuven; Finland: E. Varjonen, Helsinki University Central Hospital, Helsinki; France: F. Cambazard, Hospital Nord, Saint Etienne; B. Dreno, Centre Hospitalier Universitaire de Nantes Hôtel Dieu, Nantes; L. Dubertret, Saint Louis Hospital, Paris; J. Revuz, Henri Mondor Hospital, Créteil; B. J. Halioua, Private Practice, Paris; Germany: H.-U. Peter, University Clinic Ulm, Ulm; D. Abeck, Clinic for Dermatology & Allergy, Munich; C. Willers, Bioskin Institute, Hamburg; R. Kaufmann, Johann Wolfgang Goethe University Clinic, Frankfurt; N. Hunzelmann, University Clinic for Dermatology, Cologne; T. Zuberbier, Humboldt University Clinic, Berlin; Hungary: A. Dobozy, Medical University, Szeged; J. Feldmann, Flór Ferenc Hospital, Kistarcsa; P. Hollo, Diagnostic Unit Hungary, Budapest; the Netherlands: T. Starink, Academic Hospital Free University, Amsterdam; H. B. van der Walle, Rijnstate Hospital, Arnhem; P. Van der Valk, Academic Hospital Nijmegen St Radboud, Nijmegen; Poland: J. Roszkiewicz, Dermatology Clinic Medical Academy, Gdansk; L. Rudnicka, Dermatology Clinic MSWiA, Warsaw; J. Bowszyc, Dermatology Clinic Medica Academy, Poznan; A. Wojas-Pelc, Dermatology Clinic Medica Academy, Crakow; A. Sysa-Jedrzejowska, Dermatology Clinic Medica Academy, Lodz; S. Zabielski, Medical Academy, Warsaw; Switzerland: G. Burg, University Clinic, Zurich; J.-H. Saurat, Cantonal Hospital, Geneva; United Kingdom: C. J. Fleming, MD, Ninewells Hospital and Medical School, Dundee; A. D. Burden, University Hospital, Glasgow; R. Ratnavel, Amersham Hospital, Amersham; R. D. Aldridge, Royal Infirmary, Edinburgh.