Early Mycosis Fungoides Lesions with Increased CD3-CD4+ Histiocytes Are Associated with Dyslipidemia and Use of Statins

Background

The term "histiocyte" includes cells of the monocyte/macrophage series as antigen processing cells and the Langerhans cell/DC series as antigen-presenting cells. At least three DC subsets exist in skin: two expressing either CD1a or CD14 are dermal and Langerhans cells expressing CD1a are epidermal. Since the phenotype of histiocytic cells is typically CD3-CD4+, an estimation of the CD4+ histiocytic population can be made by comparing the numbers of CD3+ T cells with CD4+ cells. Programmed Cell Death 1 (PD-1) is an inhibitory receptor expressed on T cells, B cells, and some myeloid cells. During chronic antigen exposure, expression of PD-1 is sustained. Statins, inhibitors of cholesterol biosynthesis, are immunomodulatory agents acting on T cells and DCs, but their effects on skin immunology are unknown.

Objectives

To investigate whether infiltrates of CD3-CD4+histiocytes in early mycosis fungoides (MF) lesional skin biopsies are associated with any other factors, including history of medication and to reveal their histopathological pattern.

Methods

From Jan to Dec 2014, we identified cases of early MF from the clinic in which CD4+ cells exceeded CD3+ cells with biopsies to identify increased histiocytic population. Exclusion criteria included Sézary syndrome, granulomatous MF, T cell receptor beta monoclonality, abnormal T cell populations by flow cytometry, retinoid treatment, and progression of disease after treatment (n=12). Clinical and laboratory findings were retrospectively reviewed. Skin biopsies stained for H&E, CD3, CD4, CD7, and CD8 were reviewed. In 3 cases with paraffin blocks available, immunohistochemical stains for CD68, CD1a, CD163, PD-1, and PD-1 ligand PD-L1 were done.

Results

Clinical manifestations of early MF were pink scaly patches (9/12), capillaritis (2/12), and annular erythema - like patches (1/12). Eleven also had an increased monocytes in peripheral blood. All cases had a medication history of taking statins (atorvastatin 5/12; simvastatin 2/12; rosuvastatin 1/12) for dyslipidemia (hypercholesterolemia 7/12; both hypercholesterolemia and hypertriglyceridemia 3/12). In 9/12, symptoms persisted after MF treatment.

A lichenoid or superficial perivascular lymphohistiocytic infiltration was observed in skin lesions. Focal basal vacuolization was found in all 12 patients. Upper dermal perivascular extravasation of RBCs suggesting vasculopathy was also found in 12/12 cases. All twelve cases showed predominant CD4+ T cells compared to CD8+ T cells in dermis and the CD4+ T cells were more prominent in dermis rather than in epidermis. CD7+ T cells were preserved (3/12) or partially lost (9/12). In all 3 cases, macrophage markers CD68 and CD163 were positive in dermal infiltrates. CD1a+ DCs were increased in both epidermis and dermis in all 3/3. Only one case of three showed PD1/PD-L1+ T cells in dermis.

Discussion and Conclusion

All our cases had a medication history of statins for dyslipidemia. Of interest, skin biopsies showed a vasculopathy previously reported during high-dose atorvastatin treatment (Tehrani et al, 2013) and infiltration of CD4/CD8+ T cells, CD1a+DCs and CD163/CD68+ macrophages. We hypothesize that statins or dyslipidemia in early MF were associated with cutaneous T cell immune reaction. In support of our hypothesis that dyslipidemia is associated with histiocytosis, we found a report of nine cases of granulomatous pigmented purpuric dermatosis with concurrent hyperlipidemia (Battle et al, 2015). Cholesterol induces monocytosis and M1 macrophages in mice. One study showed that predominant migration of mature CD1a+ DC is associated with release of IL-12p70 and efficient expansion of Th 1 cells and functional CD8+ T cells. On the contrary, IL-10 up-regulates migration of immature CD14+ DC, expression of the M2 macrophage marker CD163, poor expansion of CD4+ and CD8+ T cells, and skewing of Th responses conducive to expression of PD-L1. We cannot know whether skin lesions are secondary to hyperlipidemia or to treatment with statins. Although M1 and M2 macrophages can be distinguished by diverse markers, none of these antigens are suitable for single-marker identification by immunohistochemistry in paraffin embedded tissue blocks. Further study of the cutaneous effect and immunologic mechanisms leading to increased expression of DCs and T cell dysfunction after statin medication is necessary.