Franziska C. Eberle1, Kamran Ghoreschi2 and Michael Hertl1

1Department of Dermatology and Allergology, Philipp University, Deutschhausstr. 9, DE-35037 Marburg, and 2Department of Dermatology, Eberhard Karls University, Tuebingen, Germany. E-mail: hautklinik@med.uni-marburg.de

Accepted June 20, 2009.

Sir,

Necrobiosis lipoidica (NL) was first described by Oppenheim in 1929. NL is a granulomatous skin disease with association with diabetes mellitus (DM) in 10–40% of cases, but less than 1% of patients with DM develop NL. Additional risk factors for NL include other granulomatous diseases, such as granuloma anulare (GA), sarcoidosis, colitis ulcerosa and Crohn’s disease. Middle-aged women are mainly affected. The lesions typically present as irregular or ovoid plaques with a central yellow atrophic area and a violaceous border; telangiectasia is often present. Ulcerations occur in approximately one-third of patients and complicate the clinical course. The most common sites of NL manifestation are the lower legs. Topical glucocorticosteroids represent the standard treatment of NL. In more complicated cases, glucocorticosteroids are given systemically. Several other therapies, including phototherapy, as well as systemic treatment with antimalarial drugs, cyclosporine and tumour necrosis factor (TNF) antagonists have been reported (Table I). However, a constantly effective and safe treatment has not yet been established. Especially in patients with steroid-refractory severe ulcerative NL, safe and effective therapies are needed. We report here the first case of a young woman with long-standing ulcerative NL who did not respond to topical and systemic glucocorticosteroids or phototherapy, but was treated successfully with fumaric acid esters (FAE).

CASE REPORT

A 42-year-old Caucasian woman presented with a 7-year history of NL and a 2-year history of severe ulcerations of the lesional skin of both lower legs accompanied by dysesthesia and pain (Fig. 1A). A skin biopsy from the lower left leg showed typical features of NL: a superficial and deep perivascular infiltrate of lymphocytes and plasma cells as well as epithelioid histiocytes, arranged in clusters and in a palisade shape surrounding areas of thickened bundles of collagen and degenerated collagen. Diabetes screening, X-ray of the chest and ultrasound of the abdomen revealed no signs of NL-associated diseases. Therapy with topical glucocorticosteroids, topical tacrolimus (0.1%) and compression did not lead to any improvement. Phototherapy with topical psoralen plus ultraviolet A (PUVA) over a period of 12 weeks (total cumulative dose: 77.4 J/cm2) and subsequent UVA-1 phototherapy for an additional 12 weeks (total cumulative dose: 1535 J/cm2) showed only minimal and transient improvement, but no healing. Systemic therapy with prednisolone up to 50 mg daily over 4 months was not helpful.

Fig. 1. Ulcerative necrobiosis lipoidica (NL) treated with fumaric acid esters (FAE): Lower legs presenting with irregularly shaped painful ulcers with violaceous borders (A) before treatment and (B) 6 months after treatment with FAE showing complete healing. (C) Potential interactions of FAE with cellular and molecular pathways in NL pathogenesis. Macrophage (Ф), multinucleated giant cell (MNG). DC: dendritic cell; Th: T-helper cell; TNF: tumour necrosis factor.

Since FAE have been reported to be efficient in patients with NL and other granulomatous diseases, we decided to treat the patient with FAE (1, 2). Two formulations of FAE in capsule form exist: Fumaderm® Initial and Fumaderm® (Almirall Hermal GmbH, Reinbek, Germany). The capsules differ in the composition of their active compound dimethylfumarate (DMF). The low-dose capsule Fumaderm® Initial contains 30 mg DMF, whereas the high-dose Fumaderm® formulation consists of 120 mg DMF. Dosage of FAE was administrated following the standard therapy regimen for patients with psoriasis. During therapy with FAE, the patient was regularly monitored for clinical outcome, adverse effects and laboratory tests. Laboratory tests showed no abnormalities in the differential blood count, transaminases or creatinine. Only a slight increase in the blood cholesterol level was noticed, which was unlikely to be related to the medication with FAE. At a dosage of 240 mg DMF per day, the patient reported mild gastrointestinal adverse effects, which disappeared after reducing the dosage to 120 mg DMF per day. Six weeks after the therapy was started, a major clinical improvement was noticed. After 6 months of therapy with 120 mg DMF per day, the ulcerations had completely healed (Fig. 1B). At the same time, dysesthesia and pain disappeared. This therapy was continued for a further 6 months. During follow-up no signs of newly developing NL lesions or ulcerations were observed and no adverse effects or laboratory abnormalities were noticed.

DISCUSSION

In contrast to other inflammatory skin diseases, evidence- based therapeutic guidelines for NL are missing. This is mainly due to the relatively low prevalence of NL and also to our incomplete understanding of NL pathogenesis. It remains to be clarified whether vascular changes, collagen abnormalities or immune reactions are the primary cause of NL. Collagen degeneration may precede a granulomatous response, followed by thickening of blood vessel walls, thickening of the basement membrane and fat deposition. Vascular changes, such as antibody-mediated vasculitis, may also be responsible for disease development. The observation that NL is often associated with DM supports the contention that microangiopathy initiates NL pathogenesis. Alternatively, inflammation triggered by altered metabolism or trauma may be the primary cause of NL, since impaired neutrophil migration has been suggested to result in an increased number of macrophages, which are typical cellular components of granuloma. The specific antigenic stimuli and molecular pathways that initiate granuloma formation in granulomatous diseases are unknown. Stimulation of innate immune cells results in the release of TNF and interferon (IFN-)γ. TNF induces the expression of adhesion molecules, such as ICAM-1 and VCAM-1, facilitating tissue recruitment of leukocytes. IFN-γ seems to be responsible for the activation of resident dendritic cells (DC) and macrophages. Activated DC and macrophages traffic to draining lymph nodes and to the dermis, secrete inflammatory cytokines such as interleukin (IL-)12 and induce the differentiation of IFN-γ-producing T helper (Th) 1 cells (Fig. 1C).

The critical role of immune cells and inflammatory cytokines establishes the rationale for immunosuppressive therapy of NL. The most common treatments of NL are potent topical glucocorticosteroids or calcineurin inhibitors. Phototherapy is efficient, induces local immunosuppression and reduces sclerosis by inducing tissue collagenases (3–5). Systemic rheological therapy might be helpful in some patients (6, 7). In patients with NL resistant to topical treatment and phototherapy, systemic administration of glucocorticosteroids is recommended as first-line therapy (8). For non-responders, several second-line therapies with different efficacies and evidence levels have been reported (9-11) (Table I). Nonetheless, in a subset of patients none of the traditional treatments leads to satisfying clinical outcome, especially in cases with severe ulcerations.

The patient described here did not respond to any of the common topical or systemic therapies. Cyclosporine was not administered, since the patient was already on phototherapy. Only 6 weeks after the initiation of FAE therapy, our patient showed clinical improvement and subsequent complete healing of ulcerations. This corroborates the findings of Kreuter et al. (1). The exact mode of action of FAE is still unclear. In the case of granulomatous diseases, FAE could theoretically interfere with multiple steps of granuloma formation (Fig. 1C). Although a subset of patients treated with FAE develops lymphopenia, no lymphopenia occurred in our patient. Preceding therapies with immunosuppressants were ineffective. Thus, immunomodulatory effects rather than immunosuppression seem to be mediated by FAE (12). Interestingly, FAE have been reported to act on endothelial cells (EC) and also on DC, both assumed to be relevant cell types in NL pathogenesis. FAE have been shown to inhibit the activation of NFĸB as well as the expression of ICAM-1 and VCAM-1 in EC (13, 14). Moreover, FAE inhibit the production of IL-12 by activated DC and induce the production of anti-inflammatory cytokines such as IL-4 (12, 15). The anti-inflammatory effects of FAE in combination with its effects on EC may be responsible for the clearance of ulcerative NL, as seen in our patient.

The authors declare no conflict of interest.

REFERENCES