Pravit Asawanonda, MD, DSc, and Yaowalak Nateetongrungsak, MD Bangkok, Thailand

Methotrexate plus narrowband UVB phototherapy versus narrowband UVB phototherapy alone in the treatment of plaque-type psoriasis: A randomized, placebo-controlled study Pravit Asawanonda, MD, DSc, and Yaowalak Nateetongrungsak, MD Bangkok, Thailand Background: Combining phototherapy with topical and oral agents allows clinicians to treat recalcitrant psoriasis with reduced number of treatments and cumulative UV exposures. Objective: This study was designed to determine the number of treatments necessary to clear plaque-type psoriasis when narrowband (NB) UVB is administered with methotrexate (MTX) or placebo in a randomized, controlled fashion. Methods: MTX (15 mg/wk) or placebo was administered 3 weeks before standard NB UVB phototherapy was started. Treatments with the oral agent and phototherapy were continued until Psoriasis Area and Severity Index scores were reduced to less than 10% of the original scores or 24 weeks. Follow-up was performed until lesional scores returned to 50% of the original ones. Results: A total of 24 patients were enrolled and 19 patients completed the study. Kaplan-Meier analysis revealed that the median time to clear psoriasis in the MTX/NB UVB group was 4 weeks, which was significantly less than that for the placebo/nb UVB group. Limitations: Our sample size was relatively small (24 patients) with 5 dropouts. In addition, the study was conducted in skin types III to IV, Asian patients. Follow-up was limited to 4 to 6 months after completion of phototherapy. Conclusion: MTX pretreatment allows physicians to clear psoriasis in fewer phototherapy sessions than when phototherapy is administered alone. ( J Am Acad Dermatol 2006;54:1013-8.) Treatment for psoriasis is not totally harmless. The use of methotrexate (MTX) is notoriously associated with liver toxicity especially when long-term treatment is used. Retinoids are teratogenic and cause a plethora of adverse effects, although mostly pharmacologic and dose-related. Abbreviations used: DLQI: Dermatology Life Quality Index MED: minimal erythema dose MTX: methotrexate NB: narrowband PASI: Psoriasis Area and Severity Index PUVA: psoraleneuva From the Division of Dermatology, Department of Medicine, Chulalongkorn University. Funding sources: None. Conflicts of interest: None identified. Accepted for publication January 6, 2006. Reprint requests: Pravit Asawanonda, MD, DSc, Division of Dermatology, Department of Medicine, King Chulalongkorn Memorial Hospital, Rama 4 Rd, Bangkok 10330 Thailand. E-mail: pravit@adsl.loxinfo.com. 0190-9622/$32.00 ª 2006 by the American Academy of Dermatology, Inc. doi:10.1016/j.jaad.2006.01.004 Cyclosporin A is nephrotoxic. UV therapies can cause skin to age prematurely and, more significantly, skin cancers. Because most of the adverse effects from the above-mentioned modalities occur after prolonged treatments, rotational, 1 sequential, 2 and combination therapies are often recommended. Combining two or more modalities often allows treating physicians to clear psoriasis faster, thereby exposing patients to fewer doses of each treatment. 1013

1014 Asawanonda and Nateetongrungsak JAM ACAD DERMATOL JUNE 2006 Table I. Baseline characteristics of study patients Characteristics MTX/NB UVB (n = 11) Placebo/NB UVB (n = 13) P value Age, y Range 19-73 30-62 Mean 6 SD 40.36 6 16.46 47.62 6 11.74.24* Sex Male 6 9.46 y Female 5 4 Skin type III 3 4.85 y IV 8 9 Mean duration of disease (mo) 6 SD 77 6 69.69 106.15 6 89.38.38* PASI score at baseline Range 7.6-37.4 5.7-25.2 Mean 6 SD 18.05 6 10.39 14.61 6 7.23.35* PASI score after 3 wk of MTX or placebo Range 1.6-22.5 5.6-25.2 Mean 6 SD 9.16 6 6.08 12.78 6 6.37.17* PASI score at the end of NB UVB treatment Range 0-1.4 1.2-13.8 Mean 6 SD 0.31 6 0.44 4.62 6 3.57.002* Mean DLQI score 6 SD 9.91 6 5.05 10.27 6 5.35.87* DLQI, Dermatology Life Quality Index; MTX, methotrexate; NB, narrowband; PASI, Psoriasis Area and Severity Index. *Student t test. y Chi-square test. In the developing countries, retinoids, cyclosporine, let alone biologics, are often times prohibitively expensive. MTX is, thus, the mainstay when costs are taken into consideration. Phototherapy is a relatively safe and very effective treatment modality. However, several weeks are often required to achieve satisfactory clearing of lesions. Our study was designed to evaluate the efficacy of MTX plus narrowband (NB) UVB phototherapy, the combination that has never been tested to clear plaque-type psoriasis. METHODS The study protocol was approved by the our ethics committee. Informed consent was obtained from each participant. Patients with plaque-type psoriasis, with at least 20% of body surface area involvement, whose disease activity had been stable in the 3 months before entering the study were eligible. Patients had to have discontinued systemic treatments, including psoraleneuva (PUVA), for the past 8 weeks, UVB phototherapy for 4 weeks, and all topical treatments for 2 weeks before entering study. Patients with known history of MTX intolerance, photosensitivity, immunosuppression, and alcohol abuse, and those who were pregnant or lactating were excluded. Patients were randomized by way of randomization cards to receive either MTX or placebo, which were identical in appearance. MTX (15 mg) or placebo was given in 3 divided weekly doses for 3 weeks before starting NB UVB phototherapy. The oral medications were given on weekends and continued until clearance or end of 24-week study period. At the end of the 3-week run-in period and before starting phototherapy, minimal erythema dose (MED) was determined in all patients. The fluences used at MED testing were 0.28, 0.40, 0.56, 0.80, and 1.12 J/cm 2. Total body irradiation was administered thrice weekly using stand-up cabinets (Daavlin Spectra 311, Bryan, Ohio; TL-01 100-W lamps, Philips, Eindhoven, The Netherlands). Before each phototherapy session, liberal use of mineral oil was encouraged. Starting UVB dose and dose escalations were standard as used in many treatment centers, ie, initial UVB dose was 70% of the MED then increased by 20% if no reactions developed from previous treatment, 10% if minimal reactions occurred, and 0% if erythema lasting longer than 24 hours developed. If no erythema developed from MED determination, UVB fluence was started at 0.70 J/cm 2 (rounding down from 0.78 J/cm 2, which was 70% of the highest fluence given at testing) for safety reasons. Dose increments continued until lesion clearance, which was defined as 90% reduction in Psoriasis Area and Severity Index (PASI) scores. At clearance or 24 weeks, all forms of therapy were discontinued without tapering or maintenance.

JAM ACAD DERMATOL VOLUME 54, NUMBER 6 Asawanonda and Nateetongrungsak 1015 Table II. Time to achieve Psoriasis Area and Severity Index 90 and time to relapse after clearance Patient no. Time to clearance, wk No. of NB UVB treatments Cumulative UVB fluence, J/cm 2 Time to relapse, wk MTX/NB UVB 1 2 6 6.95 Not relapsed at 16 wk 2 6 15 27.09 Not relapsed at 16 wk 3 6 18 48.88 12 4 4 12 18.30 Not relapsed at 24 wk 5 6 15 35.26 4 6 2 6 8.53 16 7 6 17 54.15 Not relapsed at 16 wk 8 4 12 24.13 12 9 6 14 26.92 Not relapsed at 16 wk 10 4 12 19.00 Not relapsed at 16 wk 11 Lost to follow-up 0 0 Lost to follow-up Placebo/NB UVB 12 Not cleared at 24 wk 68 169.58 4 13 Not cleared at 24 wk 62 163.99 4 14 Not cleared at 24 wk 65 179.45 4 15 10 28 69.25 4 16 Not cleared at 24 wk 65 204.90 4 17 4 16 32.52 Not relapsed at 24 wk 18 8 22 73.33 Not relapsed at 4 wk 19 Erythroderma 1 0.70 N/A 20 Lost to follow-up 44 55.08 Lost to follow-up 21 Lost to follow-up 16 33.57 Lost to follow-up 22 Lost to follow-up 20 42.12 Lost to follow-up 23 8 24 53.48 4 24 8 23 67.67 16 MTX, Methotrexate; NB, narrowband. PASI scores were given by an investigator blinded to the treatment assignment (P. A.) at baseline, after the 3-week pretreatment period, and every 2 weeks thereafter. Follow-up was done every 4 weeks after clearance or end of 24-week phototherapy period until relapse, which was defined as PASI scores returning to 50% of original scores or 6 months. Dermatology Life Quality Index (DLQI) questionnaire, which was translated to Thai with the permission of Dr Andrew Finlay and validated in more than 200 patients with variable dermatologic conditions, was used to evaluate the quality of life of patients every week until clearance. Complete blood cell counts and blood chemistry were performed every 4 weeks while patients were taking oral treatment. Statistical analysis Data from all randomized patients were included on an intent-to-treat basis. We used the t test, where equal variance was demonstrated, and the chi-square test in the initial exploration of the data. Otherwise, equivalent nonparametric statistics (Wilcoxon rank sum test) was used. Time-to-event analyses were performed for clearing and relapse of lesions by using Kaplan-Meier method to account for truncated observations. We analyzed differences between event curves by means of the log rank test. RESULTS Twenty-four patients were enrolled in the study. Of these, 11 were randomized to receive MTX plus NB UVB and 13 were in the placebo/nb UVB group. Table I illustrates the demographic data of the participants. Age, sex, skin types, baseline PASI scores, and baseline DLQI scores did not differ significantly between patients in the two treatment groups (Table I). In the MTX-NB UVB group the mean MED was 0.96 J/cm 2 6 0.23 whereas that of the placebo-nb UVB group was 0.80 J/cm 2 6 0. In all, 9 patients in the MTX-NB UVB group and 6 patients in the placebo-nb UVB group had MED beyond 1.12 J/cm 2. In our experience, the starting fluence larger than 0.70 J/cm 2 occasionally resulted in unexpected phototoxicity, even when MED testing yielded negative results; thus, in these patients, the starting NB UVB fluence was 0.70 J/cm 2 for safety reasons. Of 11 patients in the MTX/NB UVB group, 10 completed the study whereas 3 of 13 patients in the placebo/nb UVB dropped out. All 4 dropouts were a result of conflicts in work schedule. One patient

1016 Asawanonda and Nateetongrungsak JAM ACAD DERMATOL JUNE 2006 Fig 1. Kaplan-Meier analysis of time to clearance. MTX, Methotrexate; NB, narrowband. Fig 2. Kaplan-Meier analysis of time to relapse. MTX, Methotrexate; NB, narrowband. (No. 19) had erythroderma early in the course of phototherapy and had to discontinue the treatment. Nineteen patients, 10 male and 9 female, completed the study (Table II). Both pretreatments with MTX and placebo resulted in reduction of PASI scores. As expected, the score reduction was greater in the MTX group. The median difference in score reduction in the MTX/NB UVB versus placebo group was 5.6 (1.1, 9.7) (P =.013, Wilcoxon rank sum test). Of 11 patients who received MTX/NB UVB, 10 achieved clearance, again defined as PASI 90, whereas only 5 of 13 patients in the placebo-nb UVB did (P =.01). At the end of the treatment period, in the MTX-NB UVB group the median PASI score was 0.15. In the placebo-nb UVB group the median PASI was 3.15 (P \.001, Wilcoxon rank sum test). The median time to clear in the MTX/NB UVB group was 4 weeks whereas that in the placebo/nb UVB group could not be determined because at the end of the 24-week study period, more than half of the patients were still not cleared or lost to follow-up (Fig 1). The difference in median time to clear in the two groups was statistically significant (P\.0001, log rank test). The mean cumulative MTX dose received was 114.0 6 24.7. The mean cumulative UVB dose received in those who cleared in the MTX/NB UVB group was 26.92 6 15.54 J/cm 2 (6.95-54.15 J/cm 2 ) whereas in the placebo/nb UVB group this dose was 59.25 6 16.71 J/cm 2 (32.52-73.33 J/cm 2 )(Table II). The difference in cumulative UVB dose received in the two groups was statistically significant (P =.0026, Student t test). The median time to relapse in the placebo/nb UVB group was 4 weeks whereas that of the MTX/NB UVB group could not be evaluated because more than half of the patients still maintained their PASI scores below 50% of the original ones (Fig 2). The difference was marginally nonsignificant (P =.078, log rank test). The results obtained from DLQI assessment revealed that both groups had improvement in their quality of life with mean score reduction of 5.10 6 4.72 in the MTX/NB UVB group and 4.43 6 3.55 in the placebo/nb UVB group. This difference in score reduction was, however, not statistically significant (P =.74, Student t test). The number of adverse events in both groups did not vary with 4 episodes of mild erythema, two in each treatment group, and generalized hyperpigmentation in all patients. Lesional postinflammatory hyperpigmentation occurred in 5 and 7 patients in the MTX/NB UVB and placebo/nb UVB groups, respectively. Erythroderma developed in one patient in the placebo/nb UVB group. Treatments were otherwise well tolerated by all patients. Complete blood cell counts and blood chemistry findings were all within normal limits in all patients tested. DISCUSSION In modern phototherapy, UV is almost always administered with at least one other agent, topical or systemic, to minimize the dose of light necessary to clear lesions. Among the systemic agents used, retinoids are favored by many authors. 1,3-5 Retinoids have been shown to enhance the therapeutic

JAM ACAD DERMATOL VOLUME 54, NUMBER 6 Asawanonda and Nateetongrungsak 1017 response of broadband UVB, 6,7 NB UVB, and PUVA 8 treatments. This is reflected in the fewer numbers of treatments and cumulative UV doses required to clear lesions than when UV or the retinoid is used alone. 6,8-13 Despite the obvious efficacy, retinoids are prohibitively expensive in some developing countries, including Thailand where MTX is more widely used. The use of MTX in combination with phototherapy is not as popular as retinoids because of theoretic concern of increased carcinogenic risks. 14 This combination, although first reported in 1982, 15 is not mentioned even in some extensive reviews. 5 In their study, Paul et al 15 reported that all 26 patients were cleared after a mean of 12 broadband UVB treatments and 7 weeks of MTX intake. The mean cumulative MTX dose to achieve clearance was 112 mg (range 75-165 mg). The authors mentioned that with historic controls, the number of phototherapy sessions and the dose of UVB used in the combination study was less than half when broadband UVB alone was used. The use of MTX plus PUVA was also reported in 1982 16 by the same group of authors. They demonstrated that with the combination, the mean number of treatments required to clear psoriasis in 93% of the patients was 9.3 and the mean treatment time was 5.7 weeks. Mean cumulative MTX dose was 93.0 mg (67.5-127.5 mg). Again, when compared with prior treatments in the same patients, the number of treatments and treatment time and the cumulative UVA dose were less than half. To our knowledge this is the first randomized, placebo-controlled study of the combined use of MTX and NB UVB phototherapy. We have demonstrated that the combination of a 3-week course of MTX followed by NB UVB clears more patients with psoriasis (ie, 90.9% vs 38.5%), and in significantly fewer treatments than when NB UVB is used alone. Our results compare favorably with those obtained from similar studies. 15,16 The fact that NB UVB alone cleared only 38.5% of our patients warrants discussion as this clearance rate may not represent typical results normally obtained from this modality. Our explanations are 2-fold. First, our institution is a referral center where patients with psoriasis seen are relatively more recalcitrant to treatment. Secondly, PASI 90 represents an end point where lesions are almost totally cleared, thus, much more difficult to achieve than when less stringent end points are used. Most of our patients were of skin phototypes III and IV for which higher fluences are often necessary to clear the disease. Again, we were able to demonstrate that cumulative UVB doses can be significantly reduced with MTX administration. In addition, the cumulative MTX doses received were very low. With such low doses, MTX can be given for years before the total cumulative doses are high enough to pose any risks for hepatotoxicity. MTX, NB UVB, 17 and PUVA 18 are known to cause apoptosis in the infiltrating lymphocytes in psoriatic lesions. It is, thus, not surprising that the combination would be synergistic resulting in faster clearing of lesions. Moreover, MTX has been reported to result in reduction in the scaliness and thickness of the lesions, thereby explaining the improved skin optics when UVB phototherapy is given. 15,16 Our patients in both treatment groups reported significantly better quality of life compared with pretreatment baseline data as evident from the DLQI assessments. Unfortunately, our sample size was too small to demonstrate significant differences in DLQI scores between the two treatment groups despite the much better results with MTX-NB UVB treatment. The duration of remission obtained in our study is comparable with results obtained from other studies whereby patients maintained their remission for 3 to 6 months after NB UVB phototherapy with or without systemic therapeutic agents. 19 Short-term adverse effects associated with this treatment combination are not different from when either treatment is used alone. The MED obtained when patient is taking MTX is not at all different from unprimed skin. Erythroderma developed in one patient in the placebo/nb UVB group. The MED detected in this particular patient was more than 1.12 J/cm 2 and, thus, the initial UVB fluence administered was 0.70 J/cm 2. The erythema happened after the first phototherapy session, which might have resulted from too large incremental doses when MED was performed. The erythema recall phenomenon, although reported, 20 does not seem to occur so commonly as feared. 15 Subacute phototoxicity, reported when MTX was used in combination with PUVA, 16 was not seen in our study. In summary, we propose that the combination of MTX and NB UVB phototherapy is a useful and relatively safe treatment for plaque-type psoriasis. This combination could be especially useful in developing countries. The authors wish to thank Dr Andrew Finlay for his kind permission to use the DLQI questionnaire and Dr Chaichana Nimnuan for his assistance in statistical analysis. REFERENCES 1. Lebwohl M, Menter A, Koo J, Feldman SR. Combination therapy to treat moderate to severe psoriasis. J Am Acad Dermatol 2004;50:416-30.

1018 Asawanonda and Nateetongrungsak JAM ACAD DERMATOL JUNE 2006 2. Koo J. Systemic sequential therapy of psoriasis: a new paradigm for improved therapeutic results. J Am Acad Dermatol 1999;41(Suppl):S25-8. 3. Lebwohl M, Drake L, Menter A, Koo J, Gottlieb AB, Zanolli M, et al. Consensus conference: acitretin in combination with UVB or PUVA in the treatment of psoriasis. J Am Acad Dermatol 2001;45:544-53. 4. Zanolli M. Phototherapy arsenal in the treatment of psoriasis. Dermatol Clin 2004;22:397-406. 5. Honigsmann H. Phototherapy for psoriasis. Clin Exp Dermatol 2001;26:343-50. 6. Iest J, Boer J. Combined treatment of psoriasis with acitretin and UVB phototherapy compared with acitretin alone and UVB alone. Br J Dermatol 1989;120:665-70. 7. Lowe NJ, Prystowsky JH, Bourget T, Edelstein J, Nychay S, Armstrong R. Acitretin plus UVB therapy for psoriasis: comparisons with placebo plus UVB and acitretin alone. J Am Acad Dermatol 1991;24:591-4. 8. Saurat JH, Geiger JM, Amblard P, Beani JC, Boulanger A, Claudy A, et al. Randomized double-blind multicenter study comparing acitretin-puva, etretinate-puva and placebo-puva in the treatment of severe psoriasis. Dermatologica 1988; 177:218-24. 9. Green C, Lakshmipathi T, Johnson BE, Ferguson J. A comparison of the efficacy and relapse rates of narrowband UVB (TL-01) monotherapy vs etretinate (re-tl-01) vs etretinate- PUVA (re-puva) in the treatment of psoriasis patients. Br J Dermatol 1992;127:5-9. 10. Spuls PI, Rozenblit M, Lebwohl M. Retrospective study of the efficacy of narrowband UVB and acitretin. J Dermatol Treat 2003;14:17-20. 11. Ruzicka T, Sommerburg C, Braun-Falco O, Koster W, Lengen W, Lensing W, et al. Efficiency of acitretin in combination with UV-B in the treatment of severe psoriasis. Arch Dermatol 1990;126: 482-6. 12. Tanew A, Guggenbichler A, Honigsmann H, Geiger JM, Fritsch P. Photochemotherapy for severe psoriasis without or in combination with acitretin: a randomized, double-blind comparison study. J Am Acad Dermatol 1991;25:682-4. 13. Muchenberger S, Schopf E, Simon JC. The combination of oral acitretin and bath PUVA for the treatment of severe psoriasis. Br J Dermatol 1997;137:587-9. 14. MacKie RM, Fitzsimons MB. Risk of carcinogenicity in patients with psoriasis treated with methotrexate or PUVA singly or in combination. J Am Acad Dermatol 1983;9:467-9. 15. Paul BS, Momtaz K, Stern RS, Arndt KA, Parrish JA. Combined methotrexate-ultraviolet B therapy in the treatment of psoriasis. J Am Acad Dermatol 1982;7:758-62. 16. Morison WL, Momtaz K, Parrish JA, Fitzpatrick TB. Combined methotrexate-puva therapy in the treatment of psoriasis. J Am Acad Dermatol 1982;6:46-51. 17. Ozawa M, Ferenczi K, Kikuchi T, Cardinale I, Austin LM, Coven TR, et al. 312-Nanometer ultraviolet B light (narrow-band UVB) induces apoptosis of T cells within psoriatic lesions. J Exp Med 1999;189:711-8. 18. Coven TR, Walters IB, Cardinale I, Krueger JG. PUVA-induced lymphocyte apoptosis: mechanism of action in psoriasis. Photodermatol Photoimmunol Photomed 1999;15:22-7. 19. Koo J, Lebwohl M. Duration of remission of psoriasis therapies. J Am Acad Dermatol 1999;41:51-9. 20. Korossy KS, Hood AF. Methotrexate reactivation of sunburn reaction. Arch Dermatol 1981;117:310-1.