Alefacept for Severe Alopecia Areata a Randomized, Double-Blind, Placebo-Controlled Study

STUDY Alefacept for Severe Alopecia Areata A Randomized, Double-blind, Placebo-Controlled Study

Bruce E. Strober, MD, PhD; Kavita Menon, MD; Amy McMichael, MD; Maria Hordinsky, MD; Gerald Krueger, MD; Jackie Panko, MD; Kimberly Siu, MD; Jonathan L. Lustgarten, PhD; Elizabeth K. Ross, MD; Jerry Shapiro, MD

Objective: To assess the efficacy of alefacept for the treat- istration–approved T-cell biologic inhibitor for the treatment of severe alopecia areata (AA). ment of moderate to severe plaque psoriasis.

Design: Multicenter, double-blind, randomized, placebo- Main Outcome Measure: Improved Severity of Alo- controlled clinical trial. pecia Tool (SALT) score over 24 weeks.

Results: Participants receiving alefacept for 12 consecu- Setting: Academic departments of dermatology in the tive weeks demonstrated no statistically significant im- United States. provement in AA when compared with a well-matched placebo-receiving group (P =.70). Participants: Forty-five individuals with chronic and severe AA affecting 50% to 95% of the scalp hair and re- Conclusion: Alefacept is ineffective for the treatment of sistant to previous therapies. severe AA.

Intervention: Alefacept, a US Food and Drug Admin- Arch Dermatol. 2009;145(11):1262-1266

LOPECIA AREATA (AA) IS A lesions of AA are transplanted into mice chronic, potentially revers- with severe combined immunodefi- ible autoimmune skin dis- ciency that lack T lymphocytes, hair ease characterized by non- growth may resume, further confirming scarring patchy hair loss the potential pathogenic role of T lym- involvingA any hair-bearing surface.1 Alo- phocytes.8 pecia areata often causes considerable emo- Treatment options for more severe pre- tional distress and has limited treatment sentations of AA are limited, and neither Author Affiliations: options. The presentation and course of a cure nor preventive treatment is avail- Departments of Dermatology, AA differ from patient to patient and are able. Most of the effective therapies for AA New York University School of unpredictable.2 Most often involving the are either immunosuppressive or immuno- Medicine, New York scalp, AA typically presents as well- modulatory. Of the therapeutic options, (Drs Strober, Menon, Siu, and demarcated patches of hair loss that may corticosteroids (topical and intrale- Shapiro), Wake Forest be either isolated or numerous.1 Patients sional) remain the most popular, al- University, Winston-Salem, North Carolina may develop total loss of scalp hair (alo- though other therapies, including anthra- (Dr McMichael), University of pecia totalis) or loss of all body hair (alo- lin, minoxidil, systemic corticosteroids, 1,3 Minnesota, Minneapolis pecia universalis). Alopecia areata has an topical immunotherapy, psoralen–UV-A (Dr Hordinsky), University of incidence rate of 0.1% to 0.2%, with a life- (PUVA), and cyclosporine are also com- Utah, Salt Lake City time risk of roughly 1.7%.1,4 Most often it monly used with varying success.1 Inter- (Drs Krueger and Panko), is those with AA of the scalp who present estingly, the biologic agents etanercept Biomedical Informatics, for evaluation. and efalizumab—both effective in treat- University of Pittsburgh, T lymphocytes seem to play a central ing psoriasis, a T-cell–mediated disease— Pittsburgh, Pennsylvania role in AA through a targeted immuno- demonstrate no efficacy in treating AA.3,9 (Dr Lustgarten), Dermatology logic attack on the hair follicle leading to Despite these failures, other biologic and Laser Center NW, 3,5-7 Bellingham, Washington anagen arrest. Histologically, lesions of agents with different mechanisms of ac- (Dr Ross), and Dermatology active AA reveal dense peribulbar lym- tion may be effective. Alefacept (Ame- and Skin Science, University of phocytic infiltration, consisting of acti- vive; Astellas Pharma US Inc, Deerfield, Il- British Columbia, Vancouver, vated T lymphocytes and antigen- linois) is a bioengineered lymphocyte Canada (Dr Shapiro). presenting Langerhans cells.5-7 When function–associated antigen-3/immuno-

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 11), NOV 2009 WWW.ARCHDERMATOL.COM 1262

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/24/2021 globulin fusion protein that interrupts T-lymphocyte ac- or Յ3 squamous cell carcinomas); a history of active tubercu- tivation by binding to CD2 on T lymphocytes and con- losis (TB) or were currently undergoing treatment for active sequently interrupting the costimulation between antigen- TB; or were pregnant or breastfeeding. Individuals with a his- presenting cells and T lymphocytes.10,11 Perhaps more tory of treatment with alefacept or treatment with another in- important, alefacept specifically induces the apoptosis of vestigational medication also were not allowed to participate ϩ ϩ in the study. Systemic therapies (eg, corticosteroids, fumaric both CD4 and CD8 memory effector T lymphocytes, acid derivatives, cyclosporine A, methotrexate, biologic medi- cells that comprise the peribulbar lymphocytic infiltrate cations, retinoids, azathioprine) and phototherapy (UV-B, 10-14 and have been implicated in the pathogenesis of AA. PUVA) were discontinued for 1 month prior and during the Case reports15,16 have shown that alefacept may be effec- entire study treatment period. Topical therapies (corticoste- tive in the treatment of AA. Herein, we report the re- roids, contact sensitizing agents, tacrolimus, or pimecroli- sults of a double-blind, randomized, placebo-controlled mus) were discontinued for 2 weeks prior to and during the study of alefacept in patients with severe AA. entire study treatment period. The use of prescription medi- cations for conditions unrelated to AA was permitted. Serum samples, measuring hepatic, renal and hematologic METHODS function, and CD4ϩ T-lymphocyte count, were drawn throughout the study, and participants were assessed for signs and symp- The primary objective of this study was to assess the safety and toms of infection. Study medication was withheld if the CD4ϩ therapeutic efficacy of a weekly regimen of intramuscular (IM) T-lymphocyte count fell below 250 cells/mm3. alefacept compared with placebo over a course of 12 weeks in pa- The severity of AA and clinical response were measured with tients with chronic, severe scalp AA. This was a double-blind, ran- the Severity of Alopecia Tool (SALT) as detailed in the Alope- domized, placebo-controlled, multicenter, investigator-initiated cia Areata Investigational Guidelines.17 The SALT score is com- study in a voluntary population of patients with chronic, severe puted by measuring the percentage of hair loss in each of 4 areas scalp AA. There were 5 participating study sites. The study was of the scalp—vertex (40%), right profile (18%), left profile (18%), approved by the institutional review boards at each of the study and posterior (24%)—and adding the total to achieve a com- sites and was conducted in accordance with Good Clinical Prac- posite score. Hair regrowth is reflected by a decrease in the SALT tice guidelines, the US Food and Drug Administration (FDA) score (eg, complete hair regrowth would confer a SALT score guidelines for clinical trials, and the principles set forth by the of 0). The primary efficacy end point was the percentage of par- Declaration of Helsinki. All participants provided written in- ticipants with at least a 50% improvement from the baseline in formed consent prior to the initiation of any study procedures. their SALT score at week 24. Treatment response was also mea- Eligible individuals were randomized to receive either weekly sured at week 12. The SALT score was measured visually by IM administration of placebo or alefacept,15 mg, for 12 weeks, the study physician and corroborated by photographic analy- followed by a 12-week, posttreatment observation period. Safety sis. The participant’s perception of the extent of scalp disease and efficacy were assessed throughout the 24-week study pe- on completion of the treatment phase and at the end of the post- riod. Randomization lists detailing the person’s identification treatment, observational phase was also assessed. (ID) number and medication allocation were provided by a co- ordinating monitor. An unblinded study coordinator or pharmacist maintained the randomization lists responsible for par- STATISTICAL ANALYSIS ticipant ID number assignment and drug shipment and management. Once an individual was deemed eligible for study Efficacy and safety analyses were based on the modified intent- entry, the unblinded study coordinator or pharmacist as- to-treat population, which included all randomized patients who signed the participant an ID number and dispensed study medi- received at least 1 dose of the study medication. For the popu- cation as listed in the randomization list. Participants were moni- lation and safety analyses, patients were separated based on the tored for safety and efficacy throughout the entire study by a treatment groups to which they were randomized. The pro- blinded investigator. Efforts were made to ensure that the same portion of patients who achieved a 50% or greater reduction investigator monitored any specific participant for the entire in their SALT scores in the alefacept and placebo groups was ␹2 study period. analyzed using the test. Participants’ assessment of disease was analyzed by the Wilcoxon rank-sum test. All statistical tests were 2-sided, and all tests for efficacy were performed at the ␣ PATIENT ELIGIBILITY =.05 level. The planned sample size was 76, with 38 patients expected in each treatment arm in order to detect a 30% difference between treatment and placebo groups with 80% power Eligible individuals were 18 to 65 years of age, with a diagno- and a 2-sided test at a significance level of P=0.05. Difficulty sis of chronic, severe, scalp AA defined as at least a 50% to 95% with participant enrollment necessitated reducing the study patchy scalp hair loss of at least 6 months’ duration. Patients population to a total of 45 individuals. We were unable to per- with alopecia totalis, alopecia universalis, or coexisting signifi- form statistical tests for those persons who achieved greater than cant androgenetic alopecia (Norwood-Hamilton stage IV or 50% improvement in SALT scores owing to the low sample size. greater in males, Ludwig stage III in females) were excluded. Individuals with CD4ϩ T-lymphocyte counts below the lower limit of the reference range at screening (as determined by the RESULTS local laboratory) were not included, nor were those with either abnormal hepatic function or hematologic test results. In ad- Forty-five participants were enrolled, of whom 23 (51%) dition, individuals were not enrolled if they had a known his- were randomized to receive alefacept and 22 (49%) to tory of unstable cardiovascular or pulmonary disease, or poorly controlled diabetes mellitus; were seropositive for human im- receive placebo (Figure 1). All participants received at munodeficiency, hepatitis C, or hepatitis B viruses; had a his- least 1 treatment with study medication. In the placebo- tory of recurrent bacterial, fungal, atypical mycobacterial, vi- treated group, 3 patients withdrew after day 29, and 2 ral or opportunistic infections; a history of lymphoproliferative patients withdrew after day 92. No patients from the alef- or malignant disease (other than treated basal cell carcinoma acept treatment group withdrew from the study. The alef-

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 11), NOV 2009 WWW.ARCHDERMATOL.COM 1263

Alefacept Placebo 33 Excluded Treatment Treatment Group Group P 45 Randomized SALT Score (n=23) (n=22) Value SALT score, mean for entire group

23 Allocated to alefacept 22 Allocated to placebo Baseline 71.1 67.0 a 23 Received alefacept 22 Received placebo Week 12 61.8 65.7 .58 a 0 Did not receive alefacept 0 Did not receive placebo Week 24 67.3 64.4 .70 Participants who had Ն50% improvement in SALT score, No. 0 Lost to follow-up 5 Lost to follow-up Week 12 2 2 NA Week 24 1 3 NA 23 Analyzed 22 Analyzed Abbreviations: NA, not applicable; SALT, Severity of Alopecia Tool. aIntent to treat, last observation carried forward used for noncompleters. Figure 1. Disposition of screened patients.

Table 3. Participant Self-Assessment of Disease Table 1. Participant Characteristics Alefacept Treatment Placebo Treatment All Alefacept Placebo Distribution Group Group P Randomized Treatment Treatment of Alopecia (n=23) (n=22) Value Participants Group Group Characteristic (N=45) (n=23) (n=22) Scalp Screening Moderate to severe Moderate to severe Age, mean (SD), y 36.3 (11.1) 38.3 (10.3) 34.2 (11.8) Week 11 Moderate to severe Moderate to severe .26 Sex, No. Week 24 Moderate to severe Moderate to severe .13 Male 13 5 8 Nonscalp Female 32 18 14 Screening Mild to moderate Mild to moderate Ethnicity, No. Week 11 Mild to moderate Mild to moderate .43 White 25 13 12 Week 24 Mild to moderate Mild to moderate .54 Black 12 8 4 Hispanic 7 2 5 Asian 1 0 1 Age at onset, 21.0 (13.0) 22.5 (14.6) 19.4 (11.1) sponse (PϾ.05). Specifically, only 2 patients in the alef- mean (SD), y acept treatment group and 2 patients in the placebo group Duration of disease, 180.7 (154.7) 182.6 (175.6) 178.8 (133.4) achieved at least 50% improvement in their SALT scores mean (SD), mo at week 12. Of the 2 alefacept-treated patients who had Baseline SALT score, 69.1 (15.6) 71.1 (15.5) 67.0 (15.8) mean (SD) at least 50% improvement in their SALT scores at week Baseline CD4 count, 1132.8 (440.4) 1186.0 (459.5) 1077.2 (423.0) 12, only 1 patient maintained this improvement at week mean (SD), cells/mm3 24. We were unable to perform statistical tests for those patients who achieved greater than 50% improvement in Abbreviation: SALT, Severity of Alopecia Tool. SALT scores owing to the low sample size. Participant assessment of disease (Table 3) was ana- acept and placebo treatment groups had comparable base- lyzed by grading patient perception of the extent of hair line demographics and disease characteristics (Table 1). loss on a 7-point qualitative scale (none, trace, mild, mild Most of the participants were female (71%). Of the par- to moderate, moderate, moderate to severe, severe). Par- ticipants, 56% were white, 27% black, 16% Hispanic, and ticipants graded both scalp and nonscalp hair loss. The 2% Asian. The mean age of the participants was 36 years, mean participant assessment of disease scores was un- with a mean age at onset of AA of 21 years. All 45 indi- changed throughout the study duration, from screening viduals had received prior treatment for AA, including to study completion, for both the placebo and alefacept systemic and topical immunosuppressant therapies and treatment groups. In both treatment groups, the mean phototherapy. patient rating of scalp hair loss remained moderate to se- At baseline, the mean SALT scores were comparable: vere from screening to week 24. 71.1 in the alefacept treatment group and 67.0 in the pla- Adverse events (AEs) occurring in the group receiv- cebo treatment group. The mean percentage of hair re- ing alefacept were similar to those occurring in the pla- growth as measured by the SALT score was not signifi- cebo group and to those noted in clinical trials of alef- cantly different (PϾ.05) in either treatment group acept for moderate to severe psoriasis (Table 4). Most (Table 2). Neither group achieved the primary efficacy AEs were rated as mild and considered unrelated to study end point of at least a 50% improvement in SALT scores medication. The most frequently reported AEs in both at the end of the treatment and observation periods. By treatment groups were infections (upper respiratory in- the end of the treatment period at week 12, the alefacept- fections, influenza), headaches, and nasal congestion. treated group, when compared with the placebo group, Three serious adverse events (SAEs) occurred dur- did not demonstrate a statistically relevant treatment re- ing the course of the study. None of the SAEs had an in-

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 11), NOV 2009 WWW.ARCHDERMATOL.COM 1264

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/24/2021 fectious etiology or were thought to be related to the study medication. Of the 3 SAEs, 2 occurred in participants re- Table 4. Adverse Eventsa ceiving placebo—an episode of asthma exacerbation and an episode of vomiting and dehydration. The alefacept- Alefacept Treatment Placebo Treatment receiving individual was hospitalized for right elbow ar- Event Group Group thropathy without evidence of bacterial or viral infec- Any adverse event 20 20 tion. The patient’s symptoms were not considered by the Congestion 8 2 Infection 4 3 investigator to be related to the study medication. Cold 4 2 Laboratory tests revealed no clinically significant Headaches 4 3 ϩ changes in findings for serum chemical analysis. CD4 Nausea 2 1 T-lymphocyte counts were monitored throughout the study, and the mean CD4 counts remained higher than a Data are given as number of patients. 250 cells/mm3 (Figure 2). Only 1 person in the alefacept treatment group developed a CD4 count lower than 3 3 250 cells/mm (239 cells/mm ). Alefacept was held for 1 1400 dose, after which the patient’s CD4 count returned to nor- 3 1200

mal (658 cells/mm ). This individual experienced no fur- 3 ther declines in CD4 count. 1000

800 COMMENT 600

Anecdotal experience and case reports15,16 have noted an 400

Mean CD4 Count, Cells/mm Alefacept group improvement of AA following alefacept treatment. With this 200 Placebo group randomized, double-blind, placebo-controlled trial, we dem- 0 onstrate that a 12-week course of IM alefacept, 15 mg, does CD4 CD4 CD4 CD4 CD4 CD4 CD4 CD4 not effectively treat severe AA. In the enrolled population (Screening) (wk 1) (wk 3) (wk 5) (wk 7) (wk 9) (wk 12) (wk 24) of patients, treatment with alefacept was well tolerated. Most Week of Measurement AEs were rated as mild, and the AE profile was comparable with those from previous clinical studies of alef- Figure 2. Mean CD4 counts of alefacept- and placebo-treated individuals acept.11,13,18 Serious infections in persons receiving alef- throughout the treatment and posttreatment observation periods. acept were not observed. Alefacept may reduce CD4ϩ T-lymphocyte counts, but most participants who re- pies,includingefalizumab,alsoaninhibitorofT-lymphocyte ceived alefacept demonstrated a CD4ϩ count that re- function, are ineffective in treating AA.3 mained higher than 250 cells/mm3. In the 1 patient whose Because AA may have multiple effectors, successful CD4ϩ count transiently fell below 250 cells/mm3,noin- treatment for AA may require combination therapy with fections or other adverse events were observed. mechanistically different agents; perhaps alefacept in com- This study was limited by the number of patients ran- bination with other modalities would be effective. Be- domized to receive treatment. The planned sample size cause active AA is characterized by a type-1 immune re- was 76 participants, with 38 patients in each group to sponse, future use of therapies that inhibit this response detect at least a 30% difference at the primary end point may prove successful.15,20 Also, successful therapy may between the treatment and placebo groups. Owing to slow involve the shifting from a type 1 to a type 2 immune enrollment, only 45 participants were enrolled. Conse- response.15,20 Clearly, AA is a complex disease, and fur- quently, the study may be underpowered to detect a sta- ther research into both understanding its pathophysi- tistically significant treatment effect of alefacept. How- ologic mechanism and the development of effective thera- ever, given the negligible response rate in either group pies is needed. of this study, a larger sample size likely would not sig- nificantly change the nature of the results. Accepted for Publication: February 1, 2009. It is unclear why alefacept failed to treat AA. Alefacept Correspondence: Bruce E. Strober, MD, PhD, Depart- is an FDA-approved therapy for psoriasis, which also is a ment of Dermatology, New York University Medical Cen- T-lymphocyte–mediated disease. The failure of alefacept ter, 550 First Ave, TCH-158, New York, NY 10016 (strober to treat AA in this setting suggests that AA has a more com- @nyc.rr.com). plex pathophysiologic mechanism that may not be driven Author Contributions: Drs Strober and Menon had full by memory-effector T lymphocytes.3,15 Studies from ani- access to all of the data in the study and take responsi- mal models indicate that the inhibition of T-lymphocyte bility for the integrity of the data and the accuracy of the activation may not completely treat AA.15,19 Blocking co- data analysis. Study concept and design: Strober, stimulatory signals and thereby inhibiting T-lymphocyte McMichael, Krueger, and Shapiro. Acquisition of data: activation in mice with chronic AA had little effect on the Strober, Menon, McMichael, Hordinsky, Krueger, Panko, course of AA.15,19 However, inhibiting costimulation did and Lustgarten. Analysis and interpretation of data: Strober, prevent the development of AA in the skin-grafted mouse Menon, Hordinsky, Siu, and Ross. Drafting of the manu- model, suggesting that the maintenance of chronic AA re- script: Strober, Menon, and Hordinsky. Critical revision quires multiple mechanisms.15,19 Indeed, single-agent thera- of the manuscript for important intellectual content:

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 11), NOV 2009 WWW.ARCHDERMATOL.COM 1265

©2009 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/24/2021 McMichael, Hordinsky, Krueger, Panko, Siu, Lustgar- 5. Lu W, Shapiro J, Yu M, et al. Alopecia areata: pathogenesis and potential for therapy. ten, Ross, and Shapiro. Statistical analysis: Strober, Me- Expert Rev Mol Med. 2006;8(14):1-19. 6. Christoph T, Müller-Röver S, Audring H, et al. The human hair follicle immune non, and Ross. Administrative, technical, and material sup- system: cellular composition and immune privilege. Br J Dermatol. 2000;142 port: Menon, McMichael, Hordinsky, Krueger, Panko, and (5):862-873. Shapiro. Study supervision: Strober. Drafting and design 7. Whiting DA. Histopathologic features of alopecia areata: a new look. Arch Dermatol. of patient case report form files and protocol: Ross. 2003;139(12):1555-1559. Financial Disclosure: Dr Strober serves as a speaker for 8. Gilhar A, Ullmann Y, Berkutzki T, Assy B, Kalish RS. Autoimmune hair loss (alopecia areata) transferred by T lymphocytes to human scalp explants on SCID mice. Astellas Pharma US. Dr McMichael serves on the Scien- J Clin Invest. 1998;101(1):62-67. tific Advisory Council to the National Alopecia Areata 9. Strober BE, Siu K, Alexis AF, et al. Etanercept does not effectively treat moderate Foundation. to severe alopecia areata: an open-label study. J Am Acad Dermatol. 2005; Funding/Support: This research was supported by a grant 52(6):1082-1084. from the National Alopecia Areata Foundation, San Ra- 10. Amevive [package insert]. Cambridge, MA: Biogen Inc; 2003. fael, California, and by Biogen Idec and Astellas Pharma 11. Strober BE, Menon K. Alefacept for the treatment of psoriasis and other derma- tologic diseases. Dermatol Ther. 2007;20(4):270-276. US. 12. Vaishnaw AK, TenHoor CN. Pharmacokinetics, biologic activity, and tolerability Role of the Sponsors: The sponsors had no role in the of alefacept by intravenous and intramuscular administration. J Pharmacokinet design and conduct of the study; in the collection, analy- Pharmacodyn. 2002;29(5-6):415-426. sis, and interpretation of data; or in the preparation, re- 13. Gottlieb AB. Alefacept is well tolerated in patients with chronic plaque psoriasis. view, or approval of the manuscript. J Cutan Med Surg. 2004;8(suppl 2):14-19. 14. Lebwohl M, Christophers E, Langley RG, Ortonne JP, Roberts J, Griffiths CE; Alef- Additional Contributions: Vicki Kalabokes, BA, of the acept Clinical Study Group. An international, randomized, double-blind, placebo- National Alopecia Areata Foundation, provided encour- controlled phase 3 trial of intramuscular alefacept in patients with chronic plaque agement and tireless organizational support throughout psoriasis. Arch Dermatol. 2003;139(6):719-727. the duration of this study. 15. Heffernan MP, Hurley MY, Martin KS, Smith DI, Anadkat MJ. Alefacept for alopecia areata. Arch Dermatol. 2005;141(12):1513-1516. 16. Bui K, Polisetty S, Gilchrist H, Jackson SM, Frederic J. Successful treatment of REFERENCES alopecia universalis with alefacept: a case report and review of the literature. Cutis. 2008;81(5):431-434. 1. Wasserman D, Guzman-Sanchez DA, Scott K, McMichael A. Alopecia areata. Int 17. Olsen E, Hordinsky M, McDonald-Hull S, et al; National Alopecia Areata Foun- J Dermatol. 2007;46(2):121-131. dation. Alopecia areata investigational assessment guidelines. J Am Acad Dermatol. 2. Dudda-Subramanya R, Alexis AF, Siu K, Sinha AA. Alopecia areata: genetic com- 1999;40(2, pt 1):242-246. plexity underlies clinical heterogeneity. Eur J Dermatol. 2007;17(5):367-374. 18. Scheinfeld N. Alefacept: a safety profile. Expert Opin Drug Saf. 2005;4(6):975-985. 3. Price VH, Hordinsky MK, Olsen EA, et al. Subcutaneous efalizumab is not effec- 19. McElwee KJ, Freyschmidt-Paul P, Zöller M, Hoffmann R. Alopecia areata sus- tive in the treatment of alopecia areata. J Am Acad Dermatol. 2008;58(3):395- ceptibility in rodent models. J Investig Dermatol Symp Proc. 2003;8(2):182- 402. 187. 4. McDonagh AJ, Tazi-Ahnini R. Epidemiology and genetics of alopecia areata. Clin 20. Price VH. Therapy of alopecia areata: on the cusp and in the future. J Investig Exp Dermatol. 2002;27(5):405-409. Dermatol Symp Proc. 2003;8(2):207-211.

(REPRINTED) ARCH DERMATOL/ VOL 145 (NO. 11), NOV 2009 WWW.ARCHDERMATOL.COM 1266