Correspondence
549
using Mac 387 antibody and found an identical pattern of expression as for CD36, albeit cytoplasmic, including uninvolved keratinocytes adjacent to SCC and BCC in the absence of an inflammatory infiltrate. In this regard, our results are consistent with those of Gabrielsen et al.^ We have also attempted to induce keratinocyte CD36 and Li antigen expression in vitro by incubating cultured, normal human keratinocytes with a range of cytokines including IFN-y, tumour necrosis factor-a and interleukin i. In all experiments, expression of these antigens could not be induced, despite strong expression of ICAM-i and HLA-DR.' In vivo, intradermal administration of IFN-y* failed to induce keratinocyte expression of CD36. Thus, it appears that rather than being specific for infiammatory cutaneous disease, keratinocyte expression of these two myelomonocytic antigens is widespread in diseases involving the epidermis even in the absence of a conspicuous mononuclear cell infiltrate. Furthermore, since expression is confined to suprabasal keratinocytes in areas of epidermal damage and is unrelated to the underlying infiammatory infiltrate, we suggest that Li and CD36 expression, rather than being directly associated with cytokines produced by infiltrating mononuclear cells in the papillary dermis, may represent a direct response of keratinocytes, depending on their state of differentiation, to an assortment of environmental cues (injurious stimuli). In support of this hypothesis, keratinocytes respond in vitro to a variety of low molecular weight substances including phorbol esters, retinoids and urushiol (the catechol responsible for poison ivy dermatitis) to produce a range of pro-infiammatory cytokines including tumour necrosis factor-a and interleukin 8. This hypothesis would explain the absence of Li expression in normal skin, but positive L I staining of normal-appearing oral mucosa where repeated trauma probably exists.' University of Michigan,
J.N.W.N.BARKER
Ann Arbor,
M.H.ALLEN*
U.S.A. and *Guy's Hospital
C.E.M.GRIFFITHS B.J.NICKOLOFF
London, U.K.
D.M.MACDONALD* REFERENCES
1 Kirkham N, Peacock SJ, Jones DB. Monoclonal antibody Mac 387 recognizes a myelomonocytic antigen shared by epithelial cells in inflammatory skin diseases. Br J Dermatol 1990; 122: 61-9. 2 Gabrielsen T-0, Dale I, Brandtzaeg P et al. Epidermal and dermal distribution of a myelomonocytic antigen (Li) shared by epithelial cells in various inflammatory skin diseases. J Am Acad Dermatol 1986; 15: 173-9. 3 Kelly SE, Jones DB, Fleming S. Calgranulin expression in inflammatory dermaioses. jf Pathol 1989; 159: 17-21. 4 Barker JNWN, Markey AC, Allen MH, MacDonald DM. Keratinocyte expression of OKM5 antigen in inflammatory cutaneous disease. Br J Dermatol 1989; I2O: 613-18. 5 Hunyadi J, Simon M. Expression of OKM5 antigen on human keratinocytes in vitro upon stimulation with yinterferon. Acta Derm Venerol (Stockh) 1986; 66: 527-30. 6 Gabrielsen T-0, Brandtzaeg P, Hoel PS, Dale I. Epithelial distribution of a myelomonocytic antigen Li in relation to cutaneous malignancies and melanocytic naevi. Br J Dermatol 1988; n 8 : 59-67. 7 GriflRths OEM, Voorhees JJ, Nickoloff BJ. Gamma-interferon induces different keratinocyte cellular patterns of HLA-DR and -DQ and intercellular adhesion molecule-i (ICAM-i) antigens. Br J Dermatol 1989; 120: 1-8. 8 Barker JNWN, Allen MH, MacDonald DM. Alterations induced in normal human skin by in vivo interferon-gamma. Br J Dermatol 1990; 122; 451-8. 9 Brandtzaeg P, Dale I, Fagerhol MK. Distribution of a formalin-resistant myelomonocytic antigen (Li) in human tissues. 2. Normal and aberrant occurrence in various epithelia. AmJ Clin Pathol 1987; 87: 700-7.
Topical capsaicin for psoriasis SIR, The nervous system appears to affect psoriasis through peptides such as substance P (SP) which is released from cutaneous sensory nerve endings.' SP-immunoreactive nerve fibres are reported to be increased in the lesional skin in psoriasis,''^ and SP binds to receptors on mast cells, inducing degranulation.' In a developing lesion, mast cell degranulation is one of the early histological changes,^ and the release of infiammatory mediators may enhance epidermal and endothelial cell proliferation.^ SP induces local axon reflex vasodilatation though release of histamine from mast cells and released histamine
550
Correspondence
stimulates other sensory neurones.' SP can also increase inflammation through mediators released from macrophages and T cells. SP enhances the production of the keratinocyte-derived inflammatory mediators, interleukin i and granulocyte-macrophage colony-stimulating factor which are involved in the pathogenesis of psoriasis.* Capsaicin is derived from plants of the Solanaceae family and depletes SP through inducing degeneration of primary sensory neurones. We treated io patients with psoriasis vulgaris (seven men, three women; mean age 32 years) with topical capsaicin in an 8-week open trial. The patients had severe psoriasis with involvement of 15-30% of the total body surface area. The patients were instructed not to take any medication 4 weeks before the treatment and during the trial. The patients were provided with tubes of 0 025% capsaicin in a cream base (GenDerm Corp., Northbrook, IL, U.S.A.). The cream was applied to the psoriatic plaques on one side of the body, four times daily. Psoriatic lesions on the other side were treated with a petrolatum placebo. Evaluation was performed by the same physician before therapy, and then at weekly intervals. At each visit the itching, scaling and erythema were graded according to a four point scale (o, absent; i, trace; 2, mild; 3, moderate; 4, severe). The severity score was obtained by adding the scores. All of the patients completed the trial. At the end of the study, seven patients showed marked improvement with capsaicin as regards itching, scaling and erythema. In the non-responders, the itching disappeared. There was no significant change in the placebo-treated lesions. The patients tolerated the treatment well, and no side-effects were observed. The results of this preliminary study show that capsaicin therapy seems to be effective in the treatment of psoriasis, especially in relieving itching, by inhibiting the release of SP from cutaneous sensory neurones. Dermatology Unit, Memorial Ahmet Ors Hospital, 006510 Emek, Ankara, Turkey
N.KuRKguo6LU F.ALAYBEYI
REFERENCES
1 Naukkarinen A, Nickoloflf BJ, Farber EM. Quantification of cutaneous sensory nerves and their substance P content in psoriasis. J Invest Dermatol 1989; 92: 126-9. 2 Kurk^uoglu N, Qakar N. Substance P immunoreactivity in active edges of psoriatic plaques. Clin Exp Dermatol 1990, 15:000-000.
3 Toyry S, Fraki J, Tammi R. Mast cell density in psoriatic skin. The effect of PUVA and corticosteroid therapy. Arch Dermatol Res 1988; 280: 282-5. 4 Brown J, Perry P, Ansel J. Substance P induction of keratinocyte cytokines. J Invest Dermatol 1989; 92: 407.
Treatment of lichen planus with a short course of oral prednisolone SIR, A variety of systemic treatments have been suggested for lichen planus but corticosteroids remain the most widely used. There have been no clinical studies of the effectiveness of this therapy although a course of up to 6 weeks is recommended.' It is also stated that corticosteroid therapy does not affect the total duration of the disease.^ We therefore undertook a double-blind study of oral prednisolone in 38 patients with lichen planus. The short duration of the course of steroids allowed the dose not to be reduced gradually. The diagnosis of lichen planus was made on clinical grounds and contraindications to the study were a history of peptic ulceration, diabetes and previous treatment with corticosteroids. The patients were randomly allocated to either active or placebo therapy coded by the pharmacist. Active treatment consisted of prednisolone 30 mg daily for 10 days and placebo treatment consisted of a similar course of inactive tablets of identical appearance. Hydrocortisone-17-butyrate cream was given for topical treatment with instructions to use it twice daily. Patients were seen on the initial visit prior to entering the study, 6 weeks later then at intervals of 3 months until either the condition had cleared or 2 years had elapsed from the time of thefirstvisit. At each visit the observer recorded a score of severity of the lichen planus on a 10 cm linear analogue scale on which 0 represented 'much less severe' and 10 'very much worse'. A similar assessment was carried out by the patients themselves and there was a separate
549
using Mac 387 antibody and found an identical pattern of expression as for CD36, albeit cytoplasmic, including uninvolved keratinocytes adjacent to SCC and BCC in the absence of an inflammatory infiltrate. In this regard, our results are consistent with those of Gabrielsen et al.^ We have also attempted to induce keratinocyte CD36 and Li antigen expression in vitro by incubating cultured, normal human keratinocytes with a range of cytokines including IFN-y, tumour necrosis factor-a and interleukin i. In all experiments, expression of these antigens could not be induced, despite strong expression of ICAM-i and HLA-DR.' In vivo, intradermal administration of IFN-y* failed to induce keratinocyte expression of CD36. Thus, it appears that rather than being specific for infiammatory cutaneous disease, keratinocyte expression of these two myelomonocytic antigens is widespread in diseases involving the epidermis even in the absence of a conspicuous mononuclear cell infiltrate. Furthermore, since expression is confined to suprabasal keratinocytes in areas of epidermal damage and is unrelated to the underlying infiammatory infiltrate, we suggest that Li and CD36 expression, rather than being directly associated with cytokines produced by infiltrating mononuclear cells in the papillary dermis, may represent a direct response of keratinocytes, depending on their state of differentiation, to an assortment of environmental cues (injurious stimuli). In support of this hypothesis, keratinocytes respond in vitro to a variety of low molecular weight substances including phorbol esters, retinoids and urushiol (the catechol responsible for poison ivy dermatitis) to produce a range of pro-infiammatory cytokines including tumour necrosis factor-a and interleukin 8. This hypothesis would explain the absence of Li expression in normal skin, but positive L I staining of normal-appearing oral mucosa where repeated trauma probably exists.' University of Michigan,
J.N.W.N.BARKER
Ann Arbor,
M.H.ALLEN*
U.S.A. and *Guy's Hospital
C.E.M.GRIFFITHS B.J.NICKOLOFF
London, U.K.
D.M.MACDONALD* REFERENCES
1 Kirkham N, Peacock SJ, Jones DB. Monoclonal antibody Mac 387 recognizes a myelomonocytic antigen shared by epithelial cells in inflammatory skin diseases. Br J Dermatol 1990; 122: 61-9. 2 Gabrielsen T-0, Dale I, Brandtzaeg P et al. Epidermal and dermal distribution of a myelomonocytic antigen (Li) shared by epithelial cells in various inflammatory skin diseases. J Am Acad Dermatol 1986; 15: 173-9. 3 Kelly SE, Jones DB, Fleming S. Calgranulin expression in inflammatory dermaioses. jf Pathol 1989; 159: 17-21. 4 Barker JNWN, Markey AC, Allen MH, MacDonald DM. Keratinocyte expression of OKM5 antigen in inflammatory cutaneous disease. Br J Dermatol 1989; I2O: 613-18. 5 Hunyadi J, Simon M. Expression of OKM5 antigen on human keratinocytes in vitro upon stimulation with yinterferon. Acta Derm Venerol (Stockh) 1986; 66: 527-30. 6 Gabrielsen T-0, Brandtzaeg P, Hoel PS, Dale I. Epithelial distribution of a myelomonocytic antigen Li in relation to cutaneous malignancies and melanocytic naevi. Br J Dermatol 1988; n 8 : 59-67. 7 GriflRths OEM, Voorhees JJ, Nickoloff BJ. Gamma-interferon induces different keratinocyte cellular patterns of HLA-DR and -DQ and intercellular adhesion molecule-i (ICAM-i) antigens. Br J Dermatol 1989; 120: 1-8. 8 Barker JNWN, Allen MH, MacDonald DM. Alterations induced in normal human skin by in vivo interferon-gamma. Br J Dermatol 1990; 122; 451-8. 9 Brandtzaeg P, Dale I, Fagerhol MK. Distribution of a formalin-resistant myelomonocytic antigen (Li) in human tissues. 2. Normal and aberrant occurrence in various epithelia. AmJ Clin Pathol 1987; 87: 700-7.
Topical capsaicin for psoriasis SIR, The nervous system appears to affect psoriasis through peptides such as substance P (SP) which is released from cutaneous sensory nerve endings.' SP-immunoreactive nerve fibres are reported to be increased in the lesional skin in psoriasis,''^ and SP binds to receptors on mast cells, inducing degranulation.' In a developing lesion, mast cell degranulation is one of the early histological changes,^ and the release of infiammatory mediators may enhance epidermal and endothelial cell proliferation.^ SP induces local axon reflex vasodilatation though release of histamine from mast cells and released histamine
550
Correspondence
stimulates other sensory neurones.' SP can also increase inflammation through mediators released from macrophages and T cells. SP enhances the production of the keratinocyte-derived inflammatory mediators, interleukin i and granulocyte-macrophage colony-stimulating factor which are involved in the pathogenesis of psoriasis.* Capsaicin is derived from plants of the Solanaceae family and depletes SP through inducing degeneration of primary sensory neurones. We treated io patients with psoriasis vulgaris (seven men, three women; mean age 32 years) with topical capsaicin in an 8-week open trial. The patients had severe psoriasis with involvement of 15-30% of the total body surface area. The patients were instructed not to take any medication 4 weeks before the treatment and during the trial. The patients were provided with tubes of 0 025% capsaicin in a cream base (GenDerm Corp., Northbrook, IL, U.S.A.). The cream was applied to the psoriatic plaques on one side of the body, four times daily. Psoriatic lesions on the other side were treated with a petrolatum placebo. Evaluation was performed by the same physician before therapy, and then at weekly intervals. At each visit the itching, scaling and erythema were graded according to a four point scale (o, absent; i, trace; 2, mild; 3, moderate; 4, severe). The severity score was obtained by adding the scores. All of the patients completed the trial. At the end of the study, seven patients showed marked improvement with capsaicin as regards itching, scaling and erythema. In the non-responders, the itching disappeared. There was no significant change in the placebo-treated lesions. The patients tolerated the treatment well, and no side-effects were observed. The results of this preliminary study show that capsaicin therapy seems to be effective in the treatment of psoriasis, especially in relieving itching, by inhibiting the release of SP from cutaneous sensory neurones. Dermatology Unit, Memorial Ahmet Ors Hospital, 006510 Emek, Ankara, Turkey
N.KuRKguo6LU F.ALAYBEYI
REFERENCES
1 Naukkarinen A, Nickoloflf BJ, Farber EM. Quantification of cutaneous sensory nerves and their substance P content in psoriasis. J Invest Dermatol 1989; 92: 126-9. 2 Kurk^uoglu N, Qakar N. Substance P immunoreactivity in active edges of psoriatic plaques. Clin Exp Dermatol 1990, 15:000-000.
3 Toyry S, Fraki J, Tammi R. Mast cell density in psoriatic skin. The effect of PUVA and corticosteroid therapy. Arch Dermatol Res 1988; 280: 282-5. 4 Brown J, Perry P, Ansel J. Substance P induction of keratinocyte cytokines. J Invest Dermatol 1989; 92: 407.
Treatment of lichen planus with a short course of oral prednisolone SIR, A variety of systemic treatments have been suggested for lichen planus but corticosteroids remain the most widely used. There have been no clinical studies of the effectiveness of this therapy although a course of up to 6 weeks is recommended.' It is also stated that corticosteroid therapy does not affect the total duration of the disease.^ We therefore undertook a double-blind study of oral prednisolone in 38 patients with lichen planus. The short duration of the course of steroids allowed the dose not to be reduced gradually. The diagnosis of lichen planus was made on clinical grounds and contraindications to the study were a history of peptic ulceration, diabetes and previous treatment with corticosteroids. The patients were randomly allocated to either active or placebo therapy coded by the pharmacist. Active treatment consisted of prednisolone 30 mg daily for 10 days and placebo treatment consisted of a similar course of inactive tablets of identical appearance. Hydrocortisone-17-butyrate cream was given for topical treatment with instructions to use it twice daily. Patients were seen on the initial visit prior to entering the study, 6 weeks later then at intervals of 3 months until either the condition had cleared or 2 years had elapsed from the time of thefirstvisit. At each visit the observer recorded a score of severity of the lichen planus on a 10 cm linear analogue scale on which 0 represented 'much less severe' and 10 'very much worse'. A similar assessment was carried out by the patients themselves and there was a separate
