[Skip to Navigation]
Sign In
August 18, 2008

CD8+ Epidermotropic Cytotoxic T-Cell Lymphoma With Peripheral Blood and Central Nervous System Involvement

Author Affiliations

Author Affiliations: Department of Dermatology, Hospital of the University of Pennsylvania, Philadelphia.

Arch Dermatol. 2008;144(8):1027-1029. doi:10.1001/archderm.144.8.1027

Background  Most cutaneous T-cell lymphomas demonstrate a malignant population with a CD4+ phenotype. In rare cases, CD8+ phenotypes have been described based on immunostaining of skin specimens. Although some CD8+ lymphomas have an indolent course, others, such as CD8+ epidermotropic cytotoxic T-cell lymphomas, are typically more aggressive. To our knowledge, involvement of peripheral blood or cerebrospinal fluid with a malignant population of CD8+ cells demonstrated by flow cytometry and T-cell receptor gene rearrangement has not been previously described.

Observations  We describe a patient with a CD8+ cutaneous T-cell lymphoma with an initially indolent course and early stage diagnosed on the basis of a skin biopsy specimen. However, when flow cytometry was performed looking specifically at CD8+/CD4 cells in the peripheral blood and cerebrospinal fluid, a malignant population of CD8+/CD4/CD26/CD7 cells was discovered.

Conclusions  It is important for prognosis and treatment to be able to identify CD8+ epidermotropic cytotoxic T-cell lymphoma and separate it from other relatively indolent CD8+ lymphomas. Furthermore, detection of an abnormal CD8+/CD26/CD7 T-cell population within the peripheral blood has important prognostic and therapeutic implications. The use of flow cytometry looking for abnormal CD8+ populations in the peripheral blood or cerebrospinal fluid can assist with this critical information.

Cutaneous T-cell lymphomas (CTCLs) represent a group of primary malignant neoplasms of skin-homing T cells. The malignant clone in most of these lymphomas is a population derived from CD4+ or helper T (TH) cells. In fact, an increase in the ratio of CD4+:CD8+ cells in the peripheral blood is part of the definition of Sézary syndrome, a leukemic CTCL subtype.1,2 In addition, the CD4+ T cells exhibit the loss of 1 or more phenotypic markers, including CD26 and CD7, in many cases.

Several previously published reports3-5 have described patients with CTCL who exhibited a CD8+ or cytotoxic malignant population demonstrated by immunostaining and molecular analysis of skin biopsy specimens. Four distinct subtypes of CD8+ CTCLs have been identified to date. However, to our knowledge, there are no published reports of patients with a CD8+ CTCL and flow cytometric evidence of abnormal populations of CD8+/CD7 or CD8+/CD26 cells in the peripheral blood. We report an example of such a case, highlighting the need for recognition of the primary phenotype of the malignant cells to perform appropriate flow cytometry to detect peripheral blood involvement.

Report of a case

A 65-year-old woman who had previously been in excellent health developed constant pruritus over her back, followed several months later by pruritic, erythematous, discrete annular patches over her chest, back, abdomen, arms, and lower legs. At patient presentation, these lesions involved 10% of her body surface area and no lymphadenopathy or hepatosplenomegaly. Initial skin biopsy specimens demonstrated a markedly epidermotropic CTCL with atypical cells bearing a CD30+ phenotype and expression of CD4 and CD8. Molecular genetic testing using polymerase chain reaction (PCR) demonstrated a T-cell clonal population in her skin and a matching clone in her peripheral blood. Flow cytometric analysis of her peripheral blood revealed a decreased CD4:CD8 ratio of 0.6:1 and no populations of CD4+/CD7 or CD4+/CD26 cells.

During the next year, the patient's skin lesions improved, but she continued to develop new lesions despite therapy with oral bexarotene, 225 mg/d, initially in combination with topical nitrogen mustard, 0.02%, and subsequently psoralen–UV-A light 2 to 3 times per week. Because of recalcitrant disease, skin biopsies were performed again, and the specimens demonstrated large, epidermotropic, atypical CD8+ cells (Figure 1 and Figure 2). The results of testing for human immunodeficiency viruses 1 and 2 and human T-cell lymphotrophic virus 1 with enzyme-linked immunosorbent assay were negative, and positron emission tomography and computed tomography revealed no abnormalities. Flow cytometric analysis was repeated, this time with a specific request to the laboratory to examine the CD8+ cells. The total white blood cell count was 44 000/μL (to convert to value ×109/L, multiply by 0.001) with 22% lymphocytes on differential. The CD4:CD8 ratio was again decreased at 0.7:1, and a discrete population of CD8+/CD26 cells composed 39% of the total peripheral blood lymphocytes, which had a dominant T-cell receptor gene rearrangement identical to that in her skin. No significant populations of CD8+/CD7, CD4+/CD26, or CD4+/CD7 cells were found.

Figure 1. 
Skin biopsy specimen demonstrating large, atypical lymphocytes with epidermotropism (hematoxylin-eosin, original magnification ×20).

Skin biopsy specimen demonstrating large, atypical lymphocytes with epidermotropism (hematoxylin-eosin, original magnification ×20).

Figure 2. 
Higher magnification of large, atypical lymphocytes demonstrating epidermotropism (hematoxylin-eosin, original magnification ×40).

Higher magnification of large, atypical lymphocytes demonstrating epidermotropism (hematoxylin-eosin, original magnification ×40).

Based on the presumption that these CD8+/CD26 cells represented leukemic involvement, the patient was administered extracorporeal photopheresis. Because the patient continued to develop new lesions, oral bexarotene and psoralen–UV-A therapies were discontinued and interferon alfa and topical nitrogen mustard, 0.03%, therapies were started. The active lesions became less inflammatory after a short course of local electron beam radiation. The patient was stable for approximately 6 months; however, she then developed sudden onset of headache and neck pain. A lumbar puncture demonstrated 61 leukocytes per deciliter of cerebrospinal fluid (CSF), with 96% of these being lymphocytes. Flow cytometric analysis of these lymphocytes, again with specific gating on CD8+ cells, demonstrated discrete populations of cells that were CD8+/CD7 (61% of the total lymphocytes) and CD8+/CD26 (67% of the total lymphocytes). Molecular analysis revealed that she had a T-cell clonal population in her CSF that matched the clonal population in her skin and blood. These results combined with the patient's clinical picture of meningismus were suggestive of CTCL that involved the central nervous system.


Four distinct subtypes of CD8+ CTCLs have been identified, the first of which has been well defined and is designated as primary cutaneous CD8+ epidermotropic cytotoxic T-cell lymphoma in the most recent World Health Organization–European Organisation for Research and Treatment of Cancer cutaneous lymphoma classification scheme.3,6 As the name suggests, this subtype exhibits marked epidermotropism on histopathologic analysis. It has a poor outcome and can involve the central nervous system, lungs, testis, and oral mucosa. The histopathologic features and course of our patient's disease place her in this category. The other 3 subtypes of CD8+ lymphomas recognized by the World Health Organization–European Organization for Research and Treatment of Cancer classification include subcutaneous panniculitic-like T-cell lymphoma; cutaneous CD8+ lymphomas associated with congenital or acquired immunodeficiencies, including human immunodeficiency virus; and a heterogeneous category that includes CD8+ variants of other well-defined CTCLs, including mycosis fungoides and anaplastic large T-cell lymphoma. This last group of CD8+ lymphomas has been described to have a similar clinical course as the CD4+ counterparts of each entity.3

Previous reports of CD8+ CTCLs have determined that the malignant cell is CD8+ based on immunophenotyping and PCR testing for T-cell gene rearrangements in skin biopsy specimens. In our patient, the presence of a clonal population of T cells was also confirmed by flow cytometry and T-cell gene rearrangement PCR to be in the peripheral blood and eventually in the CSF. It is a critical finding that the malignant CD8+ cells demonstrated loss of the same T-cell markers CD7 and CD26 that have been previously described as absent in CD4+ CTCLs.7,8 The ability to measure loss of CD26 and CD7 represents a powerful tool for diagnosis and monitoring of peripheral blood involvement in these patients, many of whom we know will eventually have aggressive disease. Identifying patients who have peripheral blood involvement allows physicians to make interventions with appropriate therapy at an earlier point. For example, institution of photopheresis, referral to clinical trials, or evaluation for a stem cell transplantation may all be undertaken at an earlier point if a patient is known to have a significant burden of peripheral blood disease or visceral involvement.

A CD8+ malignant neoplasm may be suggested by the patient's histopathologic findings or an inverted CD4:CD8 ratio on histopathologic analysis or peripheral blood examination. If so, it is important to request that the gating on the flow cytometry be performed specifically on CD8+ cells so that any abnormal populations can be detected. In our patient, the diagnosis of peripheral blood involvement may have been delayed because of an initial flow cytometry that focused only on CD4+ populations that found no CD7 or CD26 cells.

Correspondence: Camille E. Introcaso, MD, Department of Dermatology, Hospital of the University of Pennsylvania, 2 Maloney Bldg, 3400 Spruce St, Philadelphia, PA 19104 (camille.introcaso@uphs.upenn.edu).

Accepted for Publication: December 28, 2007.

Author Contributions: The authors are entirely responsible for this article and the acquisition of the data described within. Dr Rook had full access to all the data in the study and takes responsibility for the integrity of the data and accuracy of the data analysis. Study concept and design: Introcaso and Rook. Acquisition of data: Introcaso, Rook, Gardner, Junkins-Hopkins, Vittorio, and Kim. Drafting of the manuscript: Introcaso. Critical revision of the manuscript for important intellectual content: Rook, Gardner, Junkins-Hopkins, Vittorio, and Kim. Administrative, technical or material support: Junkins-Hopkins. Study supervision: Rook, Vittorio, and Kim.

Financial Disclosure: None reported.

Disclaimer: No other persons contributed to the preparation of this article.

Olsen  EVonderheid  EPimpinelli  N  et al. ISCL/EORTC, Revisions to the staging and classification of mycosis fungoides and Sézary syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the cutaneous lymphoma task force of the European Organization of Research and Treatment of Cancer (EORTC) [published online ahead of print May 31, 2007].  Blood 2007;110 (6) 1713- 172210.1182/blood-2007-03-055749PubMedGoogle Scholar
Kim  EJHess  SRichardson  S  et al.  Immunopathogenesis and therapy of cutaneous T cell lymphoma.  J Clin Invest 2005;115 (4) 798- 812PubMedGoogle Scholar
El-Shabrawi-Caelen  LCerroni  LMedeiros  L  et al.  Hypopigmented mycosis fungoides: frequent expression of a CD8+ T cell phenotype.  Am J Surg Pathol 2002;26 (4) 450- 457PubMedGoogle Scholar
Marzano  AVGhislanzoni  MGianelli  UCaputo  RAlessi  EBerti  R Fatal CD8+ epidermotropic cytotoxic primary cutaneous T-cell lymphoma with multiorgan involvement.  Dermatology 2005;211 (3) 281- 285PubMedGoogle Scholar
Berti  ETomasini  DVermeer  MHMeijer  CJAlessi  EWillemze  R Primary cutaneous CD8-positive epidermotropic cytotoxic T cell lymphomas: a distinct clinicopathological entity with an aggressive clinical behavior.  Am J Pathol 1999;155 (2) 483- 492PubMedGoogle Scholar
Willemze  RJaffe  EBurg  G  et al.  WHO-EORTC classification for cutaneous lymphomas.  Blood 2005;105 (10) 3768- 3785PubMedGoogle Scholar
Bernengo  MGNovelli  MQuaglino  P  et al.  The relevance of the CD4+ CD26− subset in the identification of circulating Sezary cell.  Br J Dermatol 2001;144 (1) 125- 135PubMedGoogle Scholar
Harmon  CBWitzig  TKatzmann  J  et al.  Detection of circulating T cells with CD4+ CD7− immunophenotype in patients with benign and malignant lymphoproliferative dermatoses.  J Am Acad Dermatol 1996;35 (3, pt 1) 404- 410PubMedGoogle Scholar