Immunohistochemical study before and after narrow band (311 nm) UVB treatment in vitiligo

ARTICLE

Auteur(s) : Vincenzo De Francesco¹, Giuseppe Stinco¹, Sebastian Laspina¹, Maria Elena Parlangeli¹, Laura Mariuzzi², Pasquale Patrone¹

¹Institute of Dermatology, Department of Clinical and Experimental Pathology and Medicine, University of Udine, Italy
²Institute of Pathology, Department of Morphological and Medical Research, University of Udine, Italy

accepté le 4 Decembre 2007

Vitiligo is an acquired multifactorial disease characterised by the appearance of achromic skin patches, occasionally associated with thyroid disorders (hyper or hypothyroidism), pernicious anaemia, Addison’s disease, juvenile diabetes as well as other diseases of suspected or certain autoimmune pathogenesis [1-3]. The pathogenetic mechanisms of vitiligo are unclear. Currently, research hypotheses go in three main directions: an inherent defect in the melanocyte, an alteration in the development of the peripheral nervous system damaging the neural-crest derived melanocytes, a dysregulation of the immune response. Histologically vitiligo is characterised by a reduction of melanocytes until their complete loss and by the disappearance of melanin granules in basal and spinous layer keratinocytes.

The goal of the therapy is to obtain a repigmentation of the achromic areas, to restore chromatic skin integrity for aesthetic and functional purposes. Narrow band-UVB 311 nm (nb-UVB) phototherapy promises to be an effective and well-tolerated therapy for vitiligo [4-8].

We studied the changes occurring in the vitiligo patches before and after nb-UVB treatment with histochemical and immunohistochemical methods to identify possible markers predicting therapy outcome.

Materials and methods

Patients underwent nb-UVB phototherapy using Spectra 724 UVB lamps with a digital timer, of the FS 72 T 12/HO Daavlin type. Initially, the UVB dose was defined according to the minimal erythema dose (MED). The treatment started with an initial dose equal to 70% of the patient’s MED and incremented by 20% in the following sessions in order to maintain a minimal erythema in the irradiated vitiligo areas, up to a maximum dosage of 500 mJ per session. When the patient reported the appearance of moderate erythema or itching, the dose was kept constant in the following sessions until complete resolution of the symptoms. In the event of severe erythema, the dose was either reduced by 20% or the treatment was discontinued for one or more sessions. Patients were subjected to phototherapy 3 times a week, on alternate days, for a period of 9 months. The skin areas affected by vitiligo were photographed before, during and after treatment.

Before starting the nb-UVB treatment (T0) three biopsies were carried out on each patient: on lesional, perilesional and healthy skin, always excluding the face. After 9 months of treatment (T1) a control biopsy was carried out on repigmented skin located in the same area where the first biopses had been performed. In the absence of repigmentation, a control biopsy was repeated corresponding to the first biopsy site.

For the histochemical study, a Masson-Fontana (MF) argentaffin reaction was carried out, based on the ability of melanocytes to act as silver reducing agents in ammoniacal reactions. For the immunohistochemical study, two antigens were looked for: tyrosinase (Tyr), an enzyme necessary for the transformation of tyrosine into DOPA and HMB45, a protein present on the membrane of melanosomes and premelanosomes. The following mouse monoclonal antibodies were employed: clone T311 for Tyr and clone HMB45 for HMB45 (Ventana). For the evaluation of Tyr, a quantitative method was used which measured the number of positive cells in comparison to the number of basal cells present in four high power fields. The MF histochemical analysis and the immunohistochemical staining for HMB45 were evaluated by a semiquantitative method, whereby the staining was defined as negative (–), positive (+), and strongly positive (++).

Clinical improvement was evaluated by the photographic documentation considering the mathematical percentage of repigmented areas after nb-UVB phototherapy versus the initial clinical picture. The patients were classified in the following groups: repigmentation > 75%, repigmentation between 40% and 75%, repigmentation < 40%.

Results

In six of the 18 patients recruited (33.3%) the hematochemical tests showed low-titre anti-nucleus antibodies. Anti-thyroid peroxidase antibodies were detected in four patients (22.2%), in two of whom, in association with anti-thyroglobulin antibodies. One patient (5.5%) presented anti-gastric mucosa antibodies and another one anti-smooth muscle antibodies. Four patients (22.2%) were affected by thyroid disease, and one patient by colitis and gastritis.

At the end of the treatment with nb-UVB, six (42.8%) of the 14 patients obtained > 75% repigmentation of the vitiligo areas; seven patients (50%) displayed moderate repigmentation, ranging from 40% to 75% of the lesional areas; one patient (7.2%) showed poor clinical response, with repigmentation attaining only 15% of the total vitiligo surface and being limited to face lesions (table 1). In six patients the mode of pigmentation was perifollicular, in six marginal, in one diffuse. One patient showed a combined perifollicular and marginal pigmentation. Side effects observed during nb-UVB therapy were limited to the appearance of erythema in lesional skin in four patients which required a 20% reduction in the dose at the following session and therapy discontinuation for one session on two occasions.

The results of the histochemical analysis of the skin sections stained according to the MF method were as follows (table 1):

• – In the skin sampled at the site of the lesion before starting therapy, the search for melanin was negative in all cases (figure 1E).
• – In the perilesional skin, melanin was present and the staining was strongly positive (++) in 10 cases, and positive (+) in four cases.
• – In biopsies carried out on pigmented (control) skin, melanin was present in all cases. In six patients the staining was strongly positive (++), and in the remaining eight cases it was positive (+).
• – In biopsies carried out on repigmented skin following therapy, melanin was present in 13 patients (3 patients with positivity +, 10 patients with positivity ++) (figure 1F), while the reaction proved negative in the single patient who showed poor clinical response.

The immunohistochemical studies showed the following findings (table 1):

• – Tyrosinase (Tyr). In the biopsies carried out before therapy, Tyr-positive cells were absent in 11 cases (figure 1A), and present in three cases in lesional skin, while there were numerous Tyr-positive cells in all 14 cases in perilesional skin where the number of Tyr-positive cells appeared comparable to normal control skin. After therapy, several Tyr-positive cells were detected in 13 patients, and in most of them their number appeared comparable or higher than in control skin (figure 1B). In the single patient who did not show any major clinical improvement, almost no Tyr-positive cells could be found.
• – HMB45. The staining for this antigen was undetectable (–) in the lesional skin biopsies in 11 cases (figure 1C) and present (+) in three patients. In perilesional samples, it was strongly positive (++) in three cases and positive (+) in 11 cases. In control pigmented skin biopsies, the staining was positive (+) in all cases. Finally, biopsies carried out following nb-UVB therapy showed a strongly positive labelling (++) in six patients, positive (+) in seven patients (figure 1D), and negative (–) in a single patient, i.e. the individual who presented a poor clinical response.

The comparative examination of the results obtained by means of the histochemical and the immunohistochemical studies showed that the lesional skin from 11 patients, in whom the melanogenesis process was inactive (i.e. Tyr-negative), did not contain HMB45-positive cells or melanin pigment (MF negative). After nb-UVB treatment, three out of these 11 patients presented a > 75% repigmentation (figures 1G and H), seven a repigmentation ranging from 40% to 75%, one patient showed a poor response with 15% repigmentation of the total vitiligo surface. At the end of the treatment period, all histochemical and immunohistochemical parameters became positive in 10 of these patients, while in one subject HMB45 staining remained negative, no melanin could be detected (MF negative) and exceedingly rare tyrosinase positive cells were observed. Finally, the three patients who showed Tyr-positive melanocytes and HMB45 positive epidermal labelling in the absence of detectable amounts of melanin (MF negative) in lesional skin before therapy, proved positive for all three parameters at the end of the treatment period. Clinically, these 3 patients displayed a > 75% repigmentation.
Table 1 Clinical, histochemical and immunohistochemical findings in the 14 vitiligo patients who completed the study.

^bTyrosinase was evaluated as the number of positive cells per number of basal epidermal cells in four high power fields.

^cMasson-Fontana histochemical analysis and HMB45 immunohistochemical staining were evaluated by a semiquantitative method and defined as negative (–), positive (+), strongly positive (++).

^dMasson-Fontana histochemical analysis and HMB45 immunohistochemical staining were evaluated by a semiquantitative method and defined as negative (–), positive (+), strongly positive (++).

^eRepigmentation expressed as percentage of the respective vitiligo areas and classified in four groups: > 75%, between 40% and 75%, and < 40%.

Discussion

The clinical result was confirmed by the histological examination which showed in all patients a clear recovery of melanogenesis with the presence of melanin in the epidermis. The histochemical and immunohistochemical analyses also confirmed the favourable evolution of the treatment. Before therapy initiation, Tyr was inactive and there were no detectable melanosomes and melanin in the epidermis, as assessed by HMB45 and MF staining respectively, in 11 patients; in the remaining 3 patients, though with evidence of Tyr activity and the presence of melanosomes, melanocytes turned out to be unable to produce melanin (MF–/Tyr+/HMB45+). Two of these three patients had recently completed a phototherapy treatment obtaining partial repigmentation. It its possible that the previous treatment had induced a, though partial, activation of melanogenesis leading to the positivisation of both immunohistochemical markers but not of the histochemical one. This observation is in agreement with a report by Im et al. [13] who showed an increase in the activity of melanocyte tyrosinase in vitiligo following a UVB treatment similar to that of melanocytes in the non-vitiligo skin. The third patient presented a localized vitiligo which appeared one month before recruitment. The recent disease onset may explain the presence of tyrosinase and melanosomes in the absence of detectable amounts of melanin. This patient responded very well to the treatment, obtaining a > 75% repigmentation and his response was more rapid than that observed in patients with longer-standing skin lesions. Although referring to a single patient, this outcome tends to underline that the best results can be obtained in patients with recent vitiligo lesions, as reported in the literature [10, 12]. Hence, once the diagnosis has been ascertained, the therapy should start as early as possible.

By analysing our immunohistochemical data, patients appear to belong to two groups, one including patients with MF–/Tyr–/HMB45–, the other comprising patients with MF-/Tyr+/HMB45+. It is our opinion that these two groups do not correspond to different vitiligo patterns, but rather to two evolution stages of the disease, and that belonging to the group showing Tyr activity and the presence of melanosomes in lesional skin represents a favourable prognostic index. In these patients, a good and rapid response to therapy can be expected. However, these findings need to be confirmed in a larger patient population.

The single patient who did not show any major improvement presented at the time of recruitment with diffuse vitiligo on the face as well as linear lesions in the left and right supraclavear regions, a likely expression of Koebner phenomenon (contact with brassiere straps). At the end of treatment, the lesions in the supraclavear regions remained clinically unchanged. The histological exam on the biopsy carried out on a vitiligo lesion of the supraclavear lesion showed, after UVB treatment, a weak positivisation for Tyr in rare epidermal cells and HMB45–, MF–. However, the appearance of Tyr activity suggests that, by continuing with phototherapy treatment, repigmentation might have occurred.

The repigmentation of amelanotic vitiligo patches induced by UVB is thought to be the result of the stabilisation of the depigmentation process and melanocyte stimulation. Thanks to their anti-inflammatory and immunosuppressive action capable of regulating the T-cytotoxic lymphocytes and, therefore, modulate the immune (both humoral and cellular) disorders which typically accompany vitiligo, UV rays can be assumed to be able to block the evolution of the disease [14, 15].

In conclusion, our findings confirm that nb-UVB phototherapy is currently an effective and safe therapy for the treatment of vitiligo, considering both the good clinical-histological results obtained and the absence of important side effects. Moreover, the results of the immunohistochemical analysis suggest that positivity for Tyr and HMB45 within lesional vitiligo skin might represent an index of favourable response to treatment. Further studies on larger patient cohorts are needed to assess the possible value of reactivity for Tyr and HMB45 in vitiligo lesions as positive prognostic markers.

Acknowledgements

References

2 Gopal KV, Rama Rao GR, Kumar YH, Appa Rao MV, Vasuder P. SriKant. Vitiligo: a part of a systemic autoimmune process. Indian J Dermatol Venereol Leprol 2007; 73: 162-5.

3 Dell’Anna ML, Picardo M. A review and a new hypothesis for non immunological pathogenetic mechanisms in vitiligo. Pigment Cell Res 2006; 19: 406-11.

4 Van Weelden H, Baart de la Faille H, Young E, van der Leun JC. Comparison of narrow band UVB phototherapy and PUVA photochemotherapy in the treatment of psoriasis. Acta Derm Venereol 1990; 70: 212-5.

5 Berneburg M, Rocken M, Benedix F. Phototherapy with narrowband vs broadband UVB. Acta Derm Venereol 2005; 85: 98-108.

6 Kim DY, Cho SB, Park YK. Various patterns of repigmentation after narrowband UVB monotherapy in patients with vitiligo. J Dermatol 2005; 32: 771-2.

7 Kanwar AJ, Dogra S, Parsad D, Kumar B. Narrowband UVB for the treatment of vitiligo: an emerging effective and well-tolerated therapy. Int J Dermatol 2005; 44: 57-60.

8 Forschener T, Buchholtz S, Stockfleth E. Current state of vitiligo therapy-evidence- based analysis of the literature. J Dtsch Dermatol Ges 2007; 5: 467-75.

9 Westerhof W, Nieuweboer-Krobotova L. Treatment of vitiligo with UVB radiation versus topical psoralen plus UVA. Arch Dermatol 1997; 133: 1525-8.

10 Scherschun L, Kim JJ, Lim HW. Narrow-band ultraviolet B is a useful and well-tolerated treatment for vitiligo. J Am Acad Dermatol 2001; 44: 999-1003.

11 Tjioe M, Gerritsen MJ, Juhlin L, Van de Kerkhof PC. Treatment of vitiligo vulgaris with narrow band UVB (311 nm) for one year and the effect of addition of folic acid and vitamin B12. Acta Derm Venereol 2002; 82: 369-72.

12 Brazzelli V, Prestinari F, Barbagallo T, Bellani E, Borroni G. Long-term narrowband UVB phototherapy in vitiligo: good results are correlated with a long period of continuous and constant therapy. G Ital Dermatol Venereol 2004; 139: 171-9.

13 Im S, Hann SK, Kim HI, Kim NS, Park YK. Biologic characteristics of cultured human vitiligo melanocytes. Int J Dermatol 1994; 33: 556-62.

14 Njoo MD, Westerhof W, Bos JD, Bossuyt PM. The development of guidelines for the treatment of vitiligo. Arch Dermatol 1999; 135: 1514-21.

15 Njoo MD, Spuls PI, Bos JD, Westerhof W, Bossuyt PM. Non surgical repigmentation therapies in vitiligo. Meta-analysis of the literature. Arch Dermatol 1998; 134: 1532-40.