Immunohistochemical differentiation between inflammatory linear verrucous epidermal nevus (ILVEN) and psoriasis

ARTICLE

Auteur(s) : W.H. Vissers, L. Muys, P.E. Van Erp, E.M. de Jong, P.C. Van de Kerkhof

Department of dermatology, University Medical Centre St Radboud. P.O. Box 9101 6500 HB Nijmegen. The Netherlands

Article accepted on 29/03/2004

Inflammatory linear verrucous epidermal nevus (ILVEN) is a rare skin disease characterized by unilateral lichenoid, verrucous or psoriasiform lesions. The lesions coalesce to form plaques and linear bands and may have a clinical resemblance to psoriasis [1, 2].
In 1971 Altman and Mehregan defined clinical criteria for the diagnosis ILVEN [3]: early age onset; 4:1 predominance in females; frequent involvement of the left lower extremity; pruritus; distinctive inflammatory and linear appearance, following the lines of Blaschko; persistent lesions showing marked refractoriness to treatment.
ILVEN and linear psoriasis have been compared and contrasted in the literature. A difficulty has resulted from the fact that linear psoriasiform lesions may appear in patients with psoriasis; the linear lesions may closely resemble ILVEN. In fact these linear lesions in psoriasis may be regarded as psoriasiform Blaschkitis, Koebnerized by a linear genetic defect. In patients with psoriasis who also have linear psoriasiform lesions the situation is complicated. Therefore, we defined for comparative studies three groups of patients:
(i) Chronic plaque psoriasis
(ii) ILVEN cum psoriasis; ILVEN in psoriatic patients
(iii) ILVEN sine psoriasis; ILVEN in patients without signs of psoriasis
Histopathological criteria for the diagnosis ILVEN were defined by Dupré and Cristol in 1977 [4]. In addition to the psoriasiform features, including epidermal acanthosis with elongated rete ridges, focal parakeratosis, elongation of dermal papillae and a mixed, predominantly lymphocytic infiltrate, they described distinct areas of hypergranulosis with overlying orthokeratosis to alternate with areas of agranulosis with overlying parakeratotic hyperkeratosis in ILVEN. However others stated that this is not a pathognomonic feature of ILVEN [5].
Because of the lack of generally accepted pathognomonic features of ILVEN it remains difficult to distinguish ILVEN from linear psoriasis both clinically and histopathologically [5]. Previously de Jong et al. concluded that the absence of neutrophils and a normal expression of keratin 10 are in favour for the diagnosis ILVEN [6]. Welch et al. stated that ILVEN shows a different pattern of clonal dysregulation from other linear epidermal naevi [7].
As psoriasis can be regarded as a T-lymphocyte driven disorder, it is attractive to speculate that analysis of T-lymphocyte subsets may help to unravel the ever-remaining question whether ILVEN can be regarded as a disease entity distinct from psoriasis [8, 9].
The aim of the present report is to compare by quantitative immunohistochemistry the number and composition of the T-lymphocyte infiltrate of patients with clinical and histological hallmarks of ILVEN with patients bearing clinically and histologically the characteristics of psoriasis. Furthermore, we compared the degree of proliferation and differentiation in these conditions.

Case-reports

Clinical characteristics of the studied cases are summarized in Table I. Furthermore the general histological observations are mentioned in Table II.

Table II. Histological characteristics of patients with ILVEN and psoriasis

Cases 1-4: Patients with psoriasis

Patients with classical plaque psoriasis, covering at least 15% of the body, participated in this study. Two male and two female patients, aged between 42 and 56 years had active psoriasis with a fairly symmetrical distribution, not arranged according to linear patterns. Case 1 and case 4 had lesions over the head, trunk and extremities. One of them (case 1) suffered in addition from lesions over the axillae and groins, had swellings of the distal interphalangeal joints, onycholysis of the nail plate and nail pitting. Furthermore case 1, case 3 and case 4 experienced substantial itch. Case 2 had widespread nummular lesions over her body and case 3 had also lesions over her body except for her back and extremities. Biopsies from all four patients had histopathologic features that could be designated to psoriasis. They all showed papillomatosis and acanthosis, and a dermal as well as pronounced epidermal infiltrate. Cases 1, 3 and 4 had epidermal polymorphonuclear leukocytes. Cases 1 and 4 showed also extensive hyperkeratosis. The four patients had responded well to topical or systemic treatment in the past. Case 3 and case 4 had received UVB-therapy. Case 1 had received methotrexate in the past. At the time of the biopsies patients were in a stable phase of their disease. During the last two weeks they had applied no topical treatments and had not taken any systemic treatments for at least 4 weeks.

Case 5: Patient with ILVEN cum Psoriasis

Case 5 was an 89 year-old female who had extensive erythematosquamous verrucous lesions over her body, according to linear patterns following the lines of Blaschko, since she was 5 years old (Fig. 1). The lesions were linearly arranged on her right arm, right leg and right half of the thorax. She suffered from severe pruritis at these sites. The biopsy material from the linearly arranged lesions was compatible with ILVEN. Extensive hyperkeratosis papillomatosis and acanthosis were seen. Parakeratosis with underlying agranulosis was also seen. Furthermore this patient had also developed over her body erythematosquamous lesions, not arranged according to linear patterns clinically compatible with psoriasis.
Because topical treatment and UVB therapy were failing, it was decided to perform multiple excisions of the linear lesions. The psoriatic lesions, not arranged according to a non-linear pattern, had responded well to dithranol and PUVA therapy in the past.

Case 6 and 7: Patients with ILVEN sine psoriasis

These cases had exclusively linear lesions without any sign of psoriasis on other parts of the body. Case 6 (Fig. 2) was a 16 year-old boy with a sharply demarcated erythematosquamous and verrucous skin lesion on the right leg and right buttock, following the lines of Blaschko. This skin lesion was present at birth and progressed over the years. Histopathologically the lesion was compatible with ILVEN. Acanthosis and papillomatosis could be observed, however, no pronounced epidermal infiltrate was noticed. There was alternation of parakeratosis with agranulosis and orthokeratosis with hypergranulosis. Therapy with topical corticosteroids, cryotherapy, vitamin D derivatives and dithranol were not successful Case 7 was a 49 year-old male with linear erythematosquamous lesions, which appeared at the age of 25. The lesions were localised on the left arm and axilla and followed the lines of Blaschko. They caused severe itch. Because cryotherapy and several topical therapies were not successful, the lesions were excised. Histopathology was compatible with ILVEN. Acanthosis and papillomatosis with sparse infiltrate and without a pronounced epidermal infiltrate were present. Parakeratosis with agranulosis could also be seen. Biopsies were taken from both patients, when the lesions were in a stable phase.

Material and methods

Biopsies were taken from lesional skin of four patients with active psoriasis and three patients with ILVEN. Biopsies were prepared for immunohistochemical staining of the following markers: cytokeratine 10 (marker for epidermal differentiation), Ki-67 (a marker for proliferation), CD2 (predominantly expressed on activated T-lymphocytes), CD4 (T-helper lymphocytes) CD8 (cytotoxic T-lymphocytes), CD25 (Il-2 receptor, expressed on activated T-lymphocytes), CD161 (NK-receptor, not bound to MHC), CD94 (NK-receptor bound to MHC), CD45RO (effector memory T-lymphocytes) CD45RA (naive T-lymphocytes).
Sections were sliced 6 um thick and were air-dried for 30 minutes. Then the sections were fixated in cold acetone for 10 minutes. After blocking for endogenous peroxidase (1% H2O2/Na-azide in 1% BSA/PBS), they were washed in PBS for 10 minutes and incubated with 20% normal horse serum (Vector laboratories, Burlingame, USA). Subsequently sections were incubated with the primary antibodies for 1 hour. The following primary antibodies (mouse anti-human) were used, diluted in 1% bovine serum albumin (Organon, Technika, Boxtel, Netherlands)/PBS, anti-CD2 (1:50) (clone MT910), anti-CD4 (clone MT310) (1:25), anti-CD8 (clone DK25) (1:25), anti-CD45RO (clone UCHL1) (1:25), anti-CD45RA (clone 4KB5) (1:25), anti-CD94 (clone HP-3D9) (1:25), anti-CD25 (clone ACT-1) (1:25), Ki-67 (clone MIB-1) (1:100), anti-HLA-DR (clone TAL.1B5) (1:50) (all obtained from DAKO, Copenhagen Denmark), keratine-10 (clone RKSE60) (1:200) (Monosan laboratories Uden Netherlands), anti-CD161 (clone IM3450) (1:25) (Immunotech Marseille France). Sections were washed in PBS for 15 minutes. Secondary IgG horse anti-mouse biotinylated antibody (ABC- kit-mouse, Vector laboratories Burlingame USA)(dilution 1:200 in 1% BSA/PBS) was added for 30 minutes. The sections were washed for 15 minutes in PBS and this step was followed by incubation with avidin-biotin complex (ABC kit-mouse, Vector Laboratories Burlingame USA) (avidin/biotin diluted 1:50 in 1% BSA/PBS) To visualize the staining we used enhanced DAB as chromogen (DAB + kit from DAKO). Counterstaining was performed with Mayer’s Haematoxylin (Sigma St Louis USA). Furthermore from each patient we performed a hematoxiline-eosine staining. After dehydration in alcohol and histosafe, sections were mounted in Permount.

Image analysis

Of each section digital photographs were made at 50X magnification, except for Ki67 staining when we used 100X magnification. We made three representative photographs of the sections per marker. Each photograph was analysed using IP-lab software. For quantification of the number of cells positive for CD2, CD4, CD8, CD45RO, CD45RA, CD25, CD161 and CD94 we used the following procedure: After choosing a representative “region of interest”(ROI), all positive segments in the ROI are marked with a colour and counted. The ROI is chosen in the centre of the section. Epidermis and a zone of dermis up to 100 um under the basement membrane were taken within the ROI. Dermal and epidermal infiltrate were counted separately. Quantification was measured as unit: Positive cells per mm².
For quantification of Ki67-antigen positive nuclei we used the following procedure: A line following the stratum basale was drawn and all positive cells above this line were counted. Quantification was done in the unit: Positive cells per mm length of basement membrane.
For quantification of keratine-10 positive cells and HLA-DR positive cells we used the following procedure: Counting keratine-10 surface we only took the epidermal compartment as ROI. We subtracted dermal surface if it was present in the epidermal compartment. For HLA-DR we took the same ROI as for the other inflammatory markers. Quantification was measured as a % of ROI as unit.

Immunohistochemical results

The psoriatic lesions as compared to the lesions in ILVEN showed the well-established immunohistochemical features characterized by elevated numbers of Ki-67 positive nuclei, a reduced expression of K10 positive cells, HLA-DR expression and a marked expression of T-lymphocyte subsets (Table III, figure 3).

Comparison of psoriasis vulgaris (cases 1-4) with ILVEN cum psoriasis (case 5) and ILVEN sine psoriasis (cases 6-7) revealed that the number of Ki-67 positive nuclei tended to be lower in ILVEN sine psoriasis and ILVEN cum psoriasis, as compared to psoriasis; The percentage of keratin-10 positive epidermal surface was markedly lower in psoriasis as compared to both ILVEN phenotypes; HLA-DR expression tended to be more prominent in ILVEN sine psoriasis and ILVEN cum psoriasis as compared to psoriasis.
The number of T-cell subsets and cells expressing NK receptors has been summarized in Fig. 3. It can be seen that in epidermis and dermis CD4+, CD8+, CD45RO+, CD2+, CD25+, CD94+ and CD161+ cells were lower in ILVEN sine psoriasis as compared to psoriasis. Reconciling exclusively the epidermis, a generally similar picture is seen. A remarkable reduction of CD8+, CD45RO+, CD2+, CD25+, CD94+ and CD161+ cells was found in ILVEN sine psoriasis as compared to psoriasis. Comparing ILVEN cum psoriasis with psoriasis on the one hand and ILVEN sine psoriasis on the other hand the T-cell counts were intermediary. Considering dermis and epidermis the number of all T-cell subsets in ILVEN cum psoriasis approached the density in ILVEN without psoriasis, with the exception of CD2+ cells, which approached the counts found in psoriasis. Considering the epidermis only, ILVEN cum psoriasis had an intermediary density of T-cells between psoriasis and ILVEN sine psoriasis, with the exception of CD45RO + cells and CD25+ cells which were higher in ILVEN cum psoriasis, as compared to psoriasis.

Discussion

The present study reconfirms the validity of clinical and histopathological hallmarks for ILVEN, as defined previously. Indeed 2 out of 3 patients had an early onset and all had longstanding disease, which proved to be refractory to antipsoriatic treatments without any indication of spontaneous improvement and all patients suffered from severe itch. The alternating pattern with ortho- and parakeratosis was observed exclusively in case 6 and was not seen in cases 5 and 7.
In an earlier report the normal expression of keratin 10 was an important feature of ILVEN [6] and such was reconfirmed in the present study. (table III) Indeed a marked difference between psoriasis and ILVEN (cum and sine psoriasis) is evident. The proliferation marker Ki-67 tended to be lower in ILVEN (cum and sine psoriasis) but showed a considerable overlap. HLA-DR expression did not provide a clear differentiation between psoriasis and both presentations of ILVEN.
In dermis and epidermis, CD4+, CD8+, CD45RO+, CD2+, CD25+, CD94+ and CD161+ cells were lower in patients with ILVEN sine psoriasis as compared to psoriasis. Reconciling the epidermis only, patients with ILVEN sine psoriasis had major reductions of CD8+, CD45RO+, CD2+, CD94+ and CD161+ cells as compared to psoriasis. It is intriguing that, as compared to psoriasis, patients with ILVEN sine psoriasis have a major reduction of these immunocytes which have been supposed to play a crucial role in the pathogenesis of psoriasis as well as being a primary target for antipsoriatic treatments [10, 11]. Indeed it was recently shown that a selective reduction of CD45RO+ T-cells could be induced by inhibition of the LFA3-CD2 interaction, which at the same time ameliorated the psoriatic lesions [12].
Case 5 is a presentation of ILVEN in a psoriatic patient. The linear lesions in this patient may well represent a Koebner phenomenon to a pre-existing genetic mosaicism. Although the present report describes only one single case, a preliminary conclusion might be that the ILVEN lesion in this patient is characterized grosso modo by densities of T-cell populations in between psoriasis and ILVEN.

Conclusion

The previously reported increased expression of Keratin 10 in ILVEN as compared to psoriasis was confirmed. In ILVEN sine psoriasis, T-cell subsets relevant in the pathogenesis of psoriasis are markedly reduced as compared to psoriasis. In ILVEN cum psoriasis T-cell subsets and cells expressing NK-receptors had an intermediary position. Therefore, T-cell targeted treatments are unlikely to be of benefit in patients with ILVEN. n

References

1. Cheesbrough MJ, Kilby PE. The inflammatory linear verrucous epidermal naevus–a case report. Clin Exp Dermatol 1978; 3: 293-8.

2. Altman J, Mehregan AH. Inflammatory linear verrucose epidermal nevus. Arch Dermatol 1971; 104: 385-9.

3. Rogers M, McCrossin I, Commens C. Epidermal nevi and the epidermal nevus syndrome. A review of 131 cases. J Am Acad Dermatol 1989; 20: 476-88.

4. Dupre A, Christol B. [Bilateral inflammatory linear verrucous epidermal nevus localized on the lip and with minimal histological lesions]. Ann Dermatol Venereol 1977; 104: 163-4.

5. Menni S, Restano L, Gianotti R et al. Inflammatory linear verrucous epidermal nevus (ILVEN) and psoriasis in a child? Int J Dermatol 2000; 39: 30-2.

6. de Jong E, Rulo HF, van de Kerkhof PC. Inflammatory linear verrucous epidermal naevus (ILVEN) versus linear psoriasis. A clinical, histological and immunohistochemical study. Acta Derm Venereol 1991; 71: 343-6.

7. Welch ML, Smith KJ, Skelton HG et al. Immunohistochemical features in inflammatory linear verrucous epidermal nevi suggest a distinctive pattern of clonal dysregulation of growth. Military Medical Consortium for the Advancement of Retroviral Research. J Am Acad Dermatol 1993; 29: 242-8.

8. Gatti S, Carrozzo AM, Orlandi A et al. Treatment of inflammatory linear verrucous epidermal naevus with calcipotriol. Br J Dermatol 1995; 132: 837-9.

9. de Mare S, van de Kerkhof PC, Happle R. Dithranol in the treatment of inflammatory linear verrucous epidermal nevus. Acta Derm Venereol 1989; 69: 77-80.

10. Nickoloff BJ. The immunologic and genetic basis of psoriasis. Arch Dermatol 1999; 135: 1104-10.

11. Gottlieb AB. Immunologic mechanisms in psoriasis. J Invest Dermatol 1990; 95: 18S-9S.

12. Krueger JG. The immunologic basis for the treatment of psoriasis with new biologic agents. J Am Acad Dermatol 2002; 46: 1-23.