Effectiveness and side effects of UVB‐phototherapy, dithranol inpatient therapy and a care instruction programme of short contact dithranol in moderate to severe psoriasis

ARTICLE

Auteur(s) : O.Q.J. SWINKELS¹, M. PRINS¹, R.T. VEENHUIS¹, T. DE BOO², M.J.P. GERRITSEN¹, G.J. VAN DER WILT³, P.C.M. VAN DE KERKHOF¹, P.G.M. VAN DER VALK¹

¹ Department of Dermatology, University Medical Centre Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. 
² Department of Epidemiology and Biostatistics, University Medical Centre Nijmegen, The Netherlands. 
³ Department of Medical Technology Assessment, University Medical Centre Nijmegen, The Netherlands.

Article accepted on 26/02/2004

Psoriasis is a chronic skin disease with a prevalence of 1-2% in Caucasian populations [1]. Although the severity of psoriasis varies and is seldom disabling, psoriasis has a significant impact on social life in many patients, making treatment and temporary relief of symptoms sought by many patients [2]. Mild to moderate psoriasis can be treated topically with vitamin D preparations and/or dermatocorticosteroids. For moderate to severe psoriasis not sufficiently responding to the previously mentioned topical treatments, UVB-phototherapy (UVB) and dithranol therapy are first choice alternatives. Severe psoriasis resistant to topical therapy or UVB or fast relapsing psoriasis can be treated systemically with for instance methotrexate, acitretin or cyclosporin [3, 4]. A future perspective is that patients with moderate to severe psoriasis may benefit from single target immunotherapies [5-9]. 

UVB is effective but its use is limited because of the dose-related risk of skin cancer [10, 11]. Dithranol therapy is considered as highly effective, and no side effects have been observed with the exception of therapy-related skin irritation and permanent discoloration of clothing, furniture and sanitary equipment. Because of these side effects dithranol treatment was exclusively used in a hospital setting in the past [12]. Apart from the expensive treatment setting, the disadvantage of treatment at an inpatient department is the disturbance of the social life of the otherwise healthy patient. 

Dithranol exerts its action mainly in the first minutes to hours after application. The principle of short contact exposure to dithranol in easily removable formulations makes use of those early pharmacological effects [13, 14]. This approach considerably restricts the time needed for the treatment, permitting the patients to continue their daily activities. 

An optimal biological effect of dithranol just below the level of irritation of the surrounding uninvolved skin depends on careful tuning of exposure time and concentration increments. The patients must be well guided and instructed to guarantee optimal compliance and treatment schedules. It has been shown that the effectiveness of dithranol treatment schedules is related to the frequency and intensity of supervision and instruction [15]. 
In 1991 we started a day care instruction programme for patients with moderate to severe psoriasis focusing on optimal time and concentration adjustment, and patient compliance. In the present multicentre open randomised trial we studied the effectiveness and side effects of UVB, dithranol inpatient therapy and an outpatient care instruction programme with short contact dithranol in moderate to severe psoriasis. The influence of prognostic factors (psoriasis severity, demographic data and compliance) was also studied. In a previous publication we reported a cost effectiveness analysis concerning those treatment modalities [16].

Methods

Patients that met the selection criteria were randomised, making use of a computer programme, using envelopes, within three parallel randomisation strata over the three treatments under study. Patients with no contra-indications for either of the three treatment options (UVB, short contact dithranol treatment and inpatient dithranol treatment) were randomised in group I. Patients were randomised with a contra-indication for UVB, or who refused this therapy in group II (short contact treatment versus inpatient treatment). Patients who rejected inpatient treatment were randomised in group III, short contact treatment versus UVB. The stratification into three treatment options (based on contraindications and patients’personal wishes) allowed the inclusion of many patients, however with a certain risk of bias. Two extramural day-care centres, and four university centres with day-care facilities participated in the study. The inclusion, exclusion and stop criteria are summarised in Table I.

We started UVB treatment with 50% of the minimal erythemal dose (MED), increasing just below erythema three times a week [17]. The maximum UVB treatment duration was 12 weeks. Narrow-band TL01 tubes were used in one centre (Utrecht), in one centre high-pressure mercury lamps were used (Ede) and in the other centres broadband TL12 tubes. During short contact treatment a day care instruction programme was used. During the first week of treatment the patients visited the department daily. They were instructed how to perform self-treatment at home and what to do in case of irritation. After this week they were treated twice a week at the department, the other five days of the week they treated themselves at home. We used a three day short contact application scheme starting with 15 minutes of application, followed by 30 to 45 minutes application time. After three days of 45 minutes application time the therapy was continued with a 15 minutes application of a one step higher concentration. The concentrations ranged from 0.1%, 0.2%, 0.3%, 0.4%, 0.5%, 1.0%, 2.0%, 3.0% to 5.0%. In case of irritation, the treatment time was shortened or dithranol concentration lowered. The cream was prepared according to a formulation, known for its good stability, by the hospital pharmacist of the University Medical Centre St Radboud in Nijmegen [18]. The maximum short contact treatment duration in this study was 12 weeks. During inpatient treatment dithranol in petrolatum was applied diffusely for 24-hours. Treatment was started with 0.05% dithranol in petrolatum and if no irritation occurred the dithranol concentration was increased stepwise every three days (maximum 5.0%). Maximum inpatient treatment duration was 8 weeks.
To avoid lengthy treatment periods, stop criteria were formulated, based on preliminary investigations in our department. In UVB and short contact dithranol we stopped treatment if the PASI-score was either above 75% and 50% of the scores at the start of treatment after 3 and 6 weeks respectively or if the skin was not cleared of more than 50% of the original lesions (area) after 9 weeks. The maximum treatment duration was 12 weeks. In inpatient treatment we used the same stop criteria, however we used 2, 4, 6 and 8 weeks to evaluate treatment progress.
The adjuvant therapy during treatment is summarised in Table II. The corticosteroid containing preparations were only used during treatment if necessary to treat irritation of uninvolved skin or to treat psoriasis on areas that were not accessible to dithranol or UVB. During UVB and short contact treatment the patients visited the investigator every three weeks, and during inpatient treatment every two weeks. Successfully treated patients were seen monthly with a follow up of at least one year until a relapse occurred. During the follow-up period the use of topical anti-psoriatic treatments was not restricted. All patients provided written informed consent. The Review Board of the University Medical Centre St Radboud in Nijmegen approved the study protocol.

Table II. Adjuvant therapy.

To evaluate the clinical effectiveness of the three therapies we used the Psoriasis Area and Severity Index (PASI) and the total percentage of involved body surface (area) [19, 20]. During the intake and every control visit the PASI and area were assessed. The clinical response rate was defined as the rate of patients with clearance within the maximum treatment duration. Clearance was defined as the resolution of lesions for at least 90% of the baseline area. The number of days until successful treatment was determined, as was the number of days in remission (the number of days between clearance and relapse). Relapse was defined as a recurrence of lesions of at least 50% of the baseline area. The number of clearance days was determined for all patients, irrespective of the treatment result. For patients with clearance this was defined as the number of days from clearance until relapse, for patients with a therapy failure it was set to 0 days, and for patients without a relapse during follow-up, the total number of follow-up days (right-censored) was recorded. In summary, the following clinical effectiveness parameters were used 1:clinical response rate, 2: number of days until successful treatment, 3: mean % improvement baseline PASI and area, 4: number of days in remission (successfully treated patients), 5: number of clearance days (intention to treat population), 6: relapse rate after one year.
In the present study non-compliance was defined as all days without treatment, independent of the reason, including irritation, sickness, holiday, etc. All adverse events, their relation with the study treatment and the treatment of the adverse events were registered. Mild irritation was not recorded as this was regarded as a side effect. However moderate to severe irritation requiring temporarily discontinuation of treatment was recorded.The relationship with possible predictors/prognostic factors was investigated.
Although this study is descriptive in nature post-hoc analyses were done to compare the three treatments. The analysis was conducted on the intention-to-treat population, consisting of patients who were randomised, and appeared at least once after the baseline measurements. We investigated the possible relationship of clinical response rate, treatment duration and remission duration with centre, randomisation group and baseline values of PASI-score and area. We did so by comparing the model with only treatment as independent variable with the model where centre, randomisation group, their interaction and baseline values of PASI-score and area were added as independent variables (likelihood-ratio test or F-test). Differences between treatments were analysed with the uncorrected CHI² – test (clinical response rate, relapse rate), the two-sample t-test (treatment duration) and the log-rank test (number of days in remission, number of clearance days). As level of significance, 5% was used for all tests. Estimates of means or medians with 95% confidence intervals were computed using the assumed distributions or with nonparametric methods. For the number of days in remission after clearance, a lognormal distribution was assumed, since the data did not allow for the computation of a nonparametric interval. In further exploratory analyses the relationship with possible predictors/prognostic factors was investigated using stepwise regression techniques.Assuming comparability of the three treatments in this descriptive study, we calculated that with an accrual of 250 patients, the standard error of the mean number of days in remission would be approximately three weeks. This calculation was done using the data in the study performed by Velle Briffa et al. where a lognormally distributed remission time was assumed [21].

Results

Demographic data

Fig. 1 shows the patient flow. Overall, 250 patients were included in the study. 12 patients were excluded because they did not return after the baseline assessment (short contact treatment: 7, UVB: 4 and in patient: 1). The intention to treat group existed of 238 patients of whom 160 patients were male, and 78 patients were female. 78 Patients were treated with UVB, 100 patients with short contact dithranol, and 60 patients underwent inpatient dithranol treatment. The mean age was 46.7 with a standard deviation of 14.3 years. The mean baseline PASI was 15.3 ± 6.9. The mean baseline area was 21% ± 13.8. Table III shows the values for baseline and end of treatment PASI and area with the demographic data for the different treatments. Six patients stopped because of unacceptable irritation (short contact treatment: three, UVB: two and inpatient treatment: one).

Eight patients discontinued treatment (dropouts). In the short contact treated group one patient stopped because of compliance problems related to alcoholism, one patient because of compliance problems related to work, one patient because he wanted inpatient treatment and one patient for unknown reasons. In the UVB treated group one patient stopped because of the travelling distance and one patient because of compliance problems related to work. In the inpatient treated group one patient stopped because of compliance problems related to alcoholism and one patient was not willing to comply with the dithranol regimen.
Fourteen patients in the intention to treat group were protocol violators on final analysis. In the short contact treated group, three patients were protocol violators because of β-blocker use, one patient because of having instable plaque psoriasis and six patients proved to have baseline areas of < 10%. In the UVB treated group, two patients were protocol violators because of having baseline areas of < 10%. In the in patient dithranol treated group, one patient was protocol violator because of β-blocker use and one because of having < 10% baseline area. The protocol violators were included in the analysis. 155 Patients reached clearance (short contact dithranol: 59 of 100 patients, UVB: 44 of 78 patients, inpatient: 52 of 60 patients) and participated in the follow up.

Clinical effectiveness

There was no statistically significant influence of the centre, the randomisation group or the baseline values of PASI and area on the treatment results: clinical response rate, treatment duration, and remission period (p > 0.10 in all cases). Consequently, in further analyses these variables were dropped from the models.

Clinical response rate

Table IV shows the clinical response rate of the three treatment groups. The percentage of successful treatments was significantly higher in the inpatient dithranol treated group compared with the UVB (p = 0.001) and short contact dithranol (p = 0.001) treated groups, while there was no significant difference between the response rates of the latter two treatments.

Number of days until successful treatment

Summary statistics for the number of days until successful treatment are shown in Table IV. It clearly shows that the treatment duration in the inpatient group is significantly shorter compared with the other two treatments (p < 0.001). The latter two did not show a significant difference in treatment duration.

Mean % improvement baseline PASI and area

Table V shows the mean percentage baseline PASI and area improvement of the intention to treat population together with the percentage of patients with ≥ 75% baseline PASI reduction. Table VI shows the mean percentage baseline PASI and area improvement subdivided for therapy failures and successfully treated patients.

Table VI. Mean % improvement baseline PASI (SD) and area (SD) of therapy failures and successfully treated patients

Follow-up

Sixteen patients were lost during follow-up, while for another ten patients the follow-up period was shorter than one year because the study ended. For the computation of the relapse rate a follow-up of one year was defined as twelve months ± 1 month. 72 Patients had a relapse within one year of follow-up. The median follow up was 358 days with an interquartile range of 284 – 377 for patients without a relapse and 201 days with an interquartile range of 140-266 for patients with a relapse.

Number of days in remission (succesfully treated patients)

The median of the remission period for every therapy modality was estimated with 95% confidence interval. These results are listed in Table VII. Comparing short contact treatment to inpatient treatment the median number of days in remission is significantly longer after short contact treatment (p = 0.003).

Number of clearance days (intention to treat population)

Fig. 2 shows the Kaplan-Meier estimates for the number of clearance days after the three treatments. Table VII lists the estimated median number of clearance days with 95% confidence intervals for the three treatments. No significant difference in number of clearance days between the three treatments was found.

Table VII. Median days in remission, median number of clearance days and relapse rate after one year. 95% confidential interval in parentheses

The relapse rate after one year

Table VII shows the relapse rate after one year following the three evaluated treatments. The relapse rates are inversely related to the remission periods. The relapse rate after short contact treatment is lower compared to the relapse rate after inpatient treatment (p = 0.002).

Compliance

Inpatient treatment showed the best compliance. In the short contact treated group we often observed dithranol irritation leading to interruption of the treatment. In the UVB group all the other reasons for non-compliance were noticed more often. The observations have been summarised in Table VIII.

Adverse events

Irritation was observed more often in the short contact treated group, however this did not lead to a higher number of patients who stopped because of irritation in this group, compared to the other treatment groups. Table IX shows the adverse events.

Prognostic factors

We found a positive association between treatment duration and the sum of baseline thickness scores of the lesions on trunk, arms and legs registered in the PASI score table. A negative association between systemic treatment in the past and clinical response rate and number of days in remission was found. Furthermore, we found a positive association between the clinical response rate and compliance: non-compliance gives a lower probability of clearance. A larger residual area gave shorter remission duration, and a higher baseline PASI gave longer remission duration. We also found an association between the relapse rate with the end-treatment PASI and baseline PASI in the short contact treated group. A higher PASI at the treatment end gave a higher chance on relapse and a higher baseline PASI gave a lower chance on relapse within one year.

Discussion

UVB and short contact dithranol therapy show comparable treatment results in moderate to severe psoriasis. Short contact dithranol showed a lower clinical response rate and a longer treatment duration compared to inpatient dithranol therapy, whilst the remission time was longer.
The effectiveness of dithranol inpatient therapy, dithranol short contact therapy and UVB-phototherapy we present in this paper cannot easily be compared to results described elsewhere, because of differences in study population, study design but even more important the lack of comparable effectiveness parameters in former studies [21-26]. The remission times, however, seem to be longer in our study, which may partly be explained by the use of topical treatment during follow-up. We conclude that UVB is a good first choice treatment for patients with moderate to severe psoriasis. UVB is not time-consuming for the patient, the nursing and medical staff. UVB patient compliance is usually good. The use of UVB is theoretically limited because of the dose-related risk of skin cancer. A systematic review of the literature by Pasker et al. showed no excess incidence of skin cancer due to exposure of therapeutic UVB [27]. It must, however, be noted that available data are scarce and insufficient for proper analysis. Contra-indications, and non-response to UVB (in our study 40%), make alternatives like short contact dithranol therapy necessary. Compliance appeared to be an important factor for successful treatment. The clinical response rate decreases after interruption of the treatment, caused by irritation or other factors. In particular the success of short contact treatment depends on compliance, so this treatment should be reserved to centres with sufficient expertise to guarantee optimal instruction and motivation.
Inpatient dithranol treatment showed a higher relapse rate compared to short contact therapy and to UVB. It is remotely possible that, due to the stratification into three treatment options based on contraindications and patients’ personal wishes the inpatient dithranol treated group harbours a subgroup with more severe psoriasis that was well treated but had a faster relapse, while this subgroup failed to reach clearance after short contact therapy or UVB. An explanation for the longer lasting remission period in the group that was successfully treated with short contact dithranol might be that these patients gained more discipline and insight into how to treat themselves at home. This learning process is one of the objectives of this care instruction programme.
Remarkably, no associations between the clinical response rate and baseline PASI-score and baseline area were found, indicating that psoriasis responsiveness to therapy is not well reflected by those parameters. However, an association between treatment duration and the sum of thickness scores of the lesions on trunk, arms and legs registered in the baseline PASI score was found. This indicates that in our study the thickness of plaques is prognostic with respect to treatment duration. Patients with systemic therapy in the past are apparently more difficult to treat with dithranol and UVB-phototherapy and showed a shorter remission duration indicating that systemic treatment in the past may be regarded as a severity indicator.
We found an association between a larger residual area and shorter remission duration after short contact therapy. This phenomenon might indicate that a subgroup with probably more severe psoriasis can be recognised by the larger residual area prompting to optimal maintenance therapy to extend remission periods.
The results of the present study confirm that UVB and dithranol are highly effective treatments. At least 90% improvement of the baseline area was reached by 56% of the patients on UVB, by 59% of the patients on short contact dithranol and by 87% of the patients on classical inpatient dithranol treatment. The (estimated) mean remission periods, were 328 days for UVB-photothterapy, 584 days for short contact dithranol treatment and 277 days for inpatient treatment.
Comparison of the results of the present study on dithranol and UVB-phototherapy with efficacy data of other antipsoriatic treatments is hazardous as study populations of various studies and outcome criteria are different. However, the present study suggests that dithranol and UVB phototherapy approaches the efficacy of available [28-30] and experimental [5-7] systemic treatments.

Conclusion

UVB and dithranol treatments are very effective and safe therapies in moderate to severe psoriasis as shown in the present study using clear outcome criteria. These therapies can be considered as time-honoured standards in the treatment of moderate to severe psoriasis. Comparison between different antipsoriatic treatments should reconcile clearing capacity, duration of remission and safety on one hand and patient acceptability and costs on the other hand. n

References

1. Bhalerao J, Bowcock AM. The genetics of psoriasis: a complex disorder of the skin and immune system. Hum Mol Genet 1998; 7: 1537-45.

2. Poyner TF. A survey of patients with plaque psoriasis who had not consulted their doctor in the past year. British Journal of Clinical Research 1995; 6: 201-7.

3. Koo JY. Current consensus and update on psoriasis therapy: a perspective from the U.S. J Dermatol 1999; 26: 723-33.

4. van de Kerkhof PC. The management of psoriasis. Neth J Med 1998; 52: 40-5.

5. Chaudhari U, Romano P, Mulcahy LD et al. Efficacy and safety of infliximab monotherapy for plaque-type psoriasis: a randomised trial. Lancet 2001; 357: 1842-7.

6. Ellis CN, Krueger GG. Treatment of chronic plaque psoriasis by selective targeting of memory effector T lymphocytes. N Engl J Med 2001; 345: 248-55.

7. Mease PJ, Goffe BS, Metz J et al. Etanercept in the treatment of psoriatic arthritis and psoriasis: a randomised trial. Lancet 2000; 356: 385-90.

8. Kanitakis J, Butnaru AC, Claudy A. Novel biological immunotherapies for psoriasis. Expert Opin Investig Drugs 2003; 12: 1111-21.

9. Tutrone WD, Saini R, Weinberg JM. Biological therapy for psoriasis: An overview of infliximab, etanercept, efalizumab and alefacept. IDrugs 2004; 7: 45-9.

10. Ohnaka T. [Health effects of ultraviolet radiation]. Ann Physiol Anthropol 1993; 12: 1-10.

11. Wulf HC, Hansen AB, Bech-Thomsen N. Differences in narrow-band ultraviolet B and broad-spectrum ultraviolet photocarcinogenesis in lightly pigmented hairless mice. Photodermatol Photoimmunol Photomed 1994; 10: 192-7.

12. Mahrle G. Dithranol. Clin Dermatol 1997; 15: 723-37.

13. Runne U, Kunze J. Short-duration (‘minutes’) therapy with dithranol for psoriasis: a new out-patient regimen. Br J Dermatol 1982; 106: 135-9.

14. Schaefer H, Farber EM, Goldberg L et al. Limited application period for dithranol in psoriasis. Preliminary report on penetration and clinical efficacy. Br J Dermatol 1980; 102: 571-3.

15. de Mare S, Calis N, den Hartog G et al. Outpatient treatment with short-contact dithranol. The impact of frequent concentration adjustments. Acta Derm Venereol 1989; 69: 449-51.

16. Hartman M, Prins M,.Swinkels OQJ et al. Cost effectiveness analysis of a psoriasis care instruction programme with dithranol compared to UVB phototherapy and inpatient dithranol treatment. Br J Dermatol 2001.

17. 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.

18. Ros JJ. Preparation, analysis and stability of oil-in-water creams containing dithranol. Eur J Hosp Pharm 1991; 1: 77-84.

19. Fredriksson T, Pettersson U. Severe psoriasis-oral therapy with a new retinoid. Dermatologica 1978; 157: 238-44.

20. van de Kerkhof PC. On the limitations of the psoriasis area and severity index (PASI) [letter]. Br J Dermatol 1992; 126: 205.

21. Vella Briffa D, Greaves MW, Warin AP et al. Relapse rate and long-term management of plaque psoriasis after treatment with photochemotherapy and dithranol. Br Med J Clin Res Ed 1981; 282: 937-40.

22. Green C, Lakshmipathi T, Johnson BE et al. 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.

23. Karrer S, Eholzer C, Ackermann G et al. Phototherapy of psoriasis: comparative experience of different phototherapeutic approaches. Dermatology 2001; 202: 108-15.

24. Schauder S, Mahrle G. [One-hour-treatment of psoriasis with anthralin and UV-light (author’s transl)] Einstundentherapie der Psoriasis mit Cignolin und UV-licht. Z Hautkr 1982; 57: 861-6.

25. Spuls PI, Witkamp L, Bossuyt PM et al. A systematic review of five systemic treatments for severe psoriasis. Br J Dermatol 1997; 137: 943-9.

26. Yamamoto T, Matsuuchi M, Irimajiri J et al. Topical anthralin for psoriasis vulgaris: evaluation of 70 Japanese patients. J Dermatol 2000; 27: 482-5.

27. Pasker-de Jong PC, Wielink G, van der Valk PG et al. Treatment with UV-B for psoriasis and nonmelanoma skin cancer: a systematic review of the literature. Arch Dermatol 1999; 135: 834-40.

28. Ho VC, Griffiths CE, Albrecht G et al. Intermittent short courses of cyclosporin (Neoral(R)) for psoriasis unresponsive to topical therapy: a 1-year multicentre, randomized study. The PISCES Study Group. Br J Dermatol 1999; 141: 283-91.

29. Nyfors A, Brodthagen H. Methotrexate for psoriasis in weekly oral doses without any adjunctive therapy. Dermatologica 1970; 140: 345-55.

30. Timonen P, Friend D, Abeywickrama K et al. Efficacy of low-dose cyclosporin A in psoriasis: results of dose- finding studies. Br J Dermatol 1990; 122 Suppl. 36: 33-9.