Primary localized cutaneous nodular amyloidosis that appeared in a patient with severe atopic dermatitis


Tomomi Fujisawa, M.D., Department of Dermatology, Gifu University Graduate School of Medicine, 1-1 Yanagido, Gifu City 501-1194, Japan. Email:

Dear Editor,

Primary localized cutaneous nodular amyloidosis (PLCNA) can occur on the skin of any part of the body and is clinically characterized by ovoid, shiny, pink-yellow nodules or plaques with a glossy appearance. Histological findings show amyloid deposition in the papillary and reticular dermis.1,2 We present a patient with severe atopic dermatitis treated with cyclosporine, who noticed two nodules with amyloid deposition stained with anti-immunoglobulin (Ig)G kappa light chain antibody. Two factors, atopic dermatitis and treatment with cyclosporine, might have affected the patient’s immunity, resulting in the amyloid deposition in the dermis.

A 33-year-old Japanese man had a long history of severe atopic dermatitis, bronchial asthma and cataracts from childhood. There was no amyloidosis in his family history. He had been administrated 150 mg/day of cyclosporine for 7 years, to a total of approximately 330 g, which was effective against atopic dermatitis. He noticed a nodule on his left cheek in March 2008, which gradually enlarged. Another nodule appeared on the preauricular portion of his left cheek in September 2008. The first nodule was 2 cm × 2.5 cm in size, and the second one was 1.2 cm in diameter. The nodules were elastic, hard, glossy and orange-colored, but did not cause any pain or itching (Fig. 1).

Figure 1.

 Clinical findings on the left cheek (a). High magnification of the upper (b) and lower nodules (c).

The patient’s laboratory data were as follows: leukocyte count, 5670/μL; platelet count, 30.3 × 104/μL; aspartate aminotransferase, 22 U/L; alanine aminotransferase, 22 U/L; lactate dehydrogenase, 184 U/L; blood urea nitrogen, 8 mg/dL; creatinine, 0.56 mg/dL; Ca, 9.5 mg/dL; C-reactive protein, 0.09 mg/dL; IgG, 1697 mg/dL; IgM, 86 mg/dL; IgA, 413 mg/dL; IgD, 0.9 mg/dL; and IgE radioimmunosorbent test (RIST), more than 16 000 IU/mL. Antinuclear antibody, anti-SS-A/Ro antibody and anti-SS-B/La antibody were negative. We determined the serum light chain concentration using high-performance liquid chromatography, which demonstrated the following findings: kappa chain, 22.7 mg/L (standard range: 3.3–19.4); lambda chain, 24.7 mg/L (5.7–26.3); and kappa/lambda ratio, 0.92 (0.26–1.65). The β2-microglobulin concentration in his urine was within the normal range, and no Bence–Jones protein was detected. The patient’s serum protein profile, an electrocardiogram and chest X-ray were also normal. Neither laboratory data nor X-ray supported the diagnosis of myeloma or gammopathy.

Histological examination in hematoxylin–eosin stain showed that eosinophilic amorphous material was deposited nodularly in the dermis (Fig. 2a). The deposits were found to be congophilic (Fig. 2b) and were seen around the skin appendage and just beneath the flattened epidermis. In the immunochemical staining, the deposits were positive for human kappa light chain (Fig. 2c) but negative for human lambda light chain (Fig. 2d). Dense dermal infiltration of plasma cells and lymphocytes was observed, usually adjacent to the deposits.

Figure 2.

 Histopathological examination (a–d) and electron microscopic observation (e–h) of a nodule. Hematoxylin–eosin staining (a). Eosinophilic amorphous material was deposited nodularly in the dermis. Congo red staining (b). Immunostaining with anti-human kappa light chain antibody (c) and anti-human lambda light chain antibody (d). Original magnification: (a–d) ×100. Panels (f) and (h) show high magnification of rectangular areas marked by red lines in panels (e) and (g), respectively. Amyloid islets were observed around the vessels (v) in the upper dermis (e,f). Small fibrils were found among the elastic fibers at the dermoepidermal junction. Fibroblasts (fb) were present in the amyloid deposition in the deep dermis (g,h). Fibrillar materials were seen among the collagen bundles. The fibrils were non-branched (g).

In the electron microscopic study, amyloid islets were observed around the vessels in the upper dermis (Fig. 2e), small fibrils were found among the elastic fibers at the dermoepidermal junction (Fig. 2f). Fibroblasts were observed in the amyloid deposition in the deep dermis (Fig. 2g), and fine fibrillar materials were seen among the collagen bundles. The fibrils were non-branched and consistent with amyloid fibrils (Fig. 2h).

On the basis of the above findings, we made a diagnosis of PLCNA with a background of atopic dermatitis treated with cyclosporine. We removed these two nodules surgically in December 2008 and February 2009, and no further nodules appeared after the operation. The dose of cyclosporine was gradually reduced because there was a possibility that this agent had caused the nodular amyloidosis.

We report herein a case of PLCNA that appeared in an atopic dermatitis patient receiving treatment with cyclosporine. It is difficult to distinguish PLCNA from an initial finding of systemic amyloidosis.3 The amyloid deposits in PLCNA are closely related to the immunoglobulin light chain,4 as those in systemic primary and myeloma-associated amyloidoses.3,5 Amyloid is thought to be produced by a localized aggregate of clonal plasma cells in PLCNA. Systemic amyloidosis was neglected by the laboratory data in the present case. On the other hand, the amyloid in lichen amyloidosis and macular amyloidosis was shown to have identical antigenicity to epidermal keratin.6 The deposition in the present case was positively stained with anti-IgG kappa light chain antibody, but the serum kappa light chain concentration was not increased. These results show that this case was not derived from excessive light chains in serum per se. There have been no reports of PLCNA either in atopic dermatitis patients and in cyclosporine-treated patients, at least since 1983, according to Medline. Amyloid deposition has been reported in lichen amyloidosis associated with several skin disorders including atopic dermatitis.7,8 However, lichen amyloidosis has not been reported in cyclosporine-treated patients.

It is well known that PLCNA is sometimes associated with autoimmune connective tissue disorders, especially Sjögren’s syndrome.9 It has been reported that five out of 14 patients with PLCNA are involved with Sjögren’s syndrome,4 suggesting that PLCNA is associated with B lymphocyte disorders in Sjögren’s syndrome.9 The amyloid in PLCNA is usually composed of monoclonal Ig light chains, which are likely produced by plasma cells infiltrated into the skin. The immunity of the present patient, who did not suffer from Sjögren’s syndrome, may have been affected by atopic dermatitis and/or the cyclosporine used for the treatment.

This is the first report of PLCNA in a patient with atopic dermatitis treated with cyclosporine. Although there is a possibility that these two factors affected the immunity of the patient, resulting in the formation of amyloid nodules, the mechanisms for amyloid deposition from IgG kappa light chain is still unknown.


We thank Dr Norihiro Suga and Professor Hirokazu Imai of the Department of Internal Medicine, Aichi Medical University School, for determining the serum light chain concentrations.