Cutaneous T-Cell Lymphoma: Pathogenesis and Treatment

Introduction

Cutaneous T-cell lymphoma (CTCL) is an umbrella term used to describe a spectrum of clinical and histologic variants characterized by a malignant population of “cutaneous T cells” that have a predilection for the skin. As such, CTCL may be viewed as analogous to another life-threatening cancer that commonly begins in the skin—malignant melanoma. Whereas malignant melanoma is a cancer of melanocytes, a cell population that normally resides in the skin to perform an important function (eg, to protect against ultraviolet insult), CTCL is a malignancy of a distinctive subset of T cells of the skin immune system that normally patrol the skin’s environment (eg, to protect against infection).

The primary dermal and epidermal infiltrating cell in the skin of CTCL patients was first identified 20 years ago as having a T-helper phenotype.[1] Since that time, the advancement and application of sensitive tools of immunology and molecular biology have allowed for the organization of CTCL variants into a more logical classification than had previously been possible. The use of polymerase chain reaction (PCR) to amplify T-cell receptor genes has greatly enhanced our ability to detect the presence of a clonal population of T cells.[2] This has allowed for an earlier and more sensitive means of diagnosing CTCL, as well as the identification of several previously poorly classified CTCL variants.

Here again, CTCL is analogous to malignant melanoma, which can present with different, important-to-recognize variations, all of which reflect the dominant cellular features of the malignant melanocytes. Similarly, CTCL can present with a spectrum of clinical variations—all manifestations of the features of the dominant set of malignant “cutaneous T cells.” Nevertheless, it should be emphasized that even the most innocuous-appearing of these clinical variations (for example, focal patch/plaque involvement) are malignancies of cutaneous T cells.

Despite the advent of PCR technology, it may be difficult at times to make the diagnosis of CTCL or to pigeonhole cases into a particular CTCL clinical variant, for several reasons. First, there exist several clinical presentations that may demonstrate a predominant T-cell clone by molecular analysis or immunoassay but that only rarely progress to clear-cut CTCL. These entities have been termed “premycotic” CTCL, and may represent a premalignant state of CTCL.

Second, there is often clinical overlap between variants of CTCL and/or transformation from one variant to another. This transformation between the different clinical forms of CTCL may represent either the natural evolution of more aggressive subclones of the original malignant cutaneous T cell or partial reversal of disease progression by therapeutic inhibition of the most aggressive subclone(s).

Thus, a definitive diagnosis of CTCL can be made when the lymphoma shows evidence of T-cell clonality in addition to a clinical and histologic picture that is highly suggestive of one or more of the CTCL variants.

Skin Immune System and CTCL Pathogenesis

Emergence of the concept of a skin immune system[3] has shed light on the pathogenesis of CTCL. It is now clear that, in the normal state, T cells can circulate from the peripheral blood, through the selective barrier of endothelial cells bearing adhesion molecules into the dermis, to regional lymph nodes via the lymphatic vessels, and back into the peripheral circulation through high endothelial venules in the lymph-node cortex (Figure 1).

The key to this skin-blood-lymph node–specific cycling appears to be T-cell expression of the cutaneous lymphocyte antigen (CLA) marker. Naive (ie, CD45Ra-positive) T cells that encounter their specific antigen in draining lymph nodes of the skin differentiate into activated/memory (ie, CD45Ro-positive) T cells and express CLA. Once returned to the blood, CLA-positive T cells can encounter their ligand E-selectin on dermal endothelial cells for recruitment back into the skin during an inflammatory response.

During normal homeostasis, this system performs an immunosurveillance function as part of the defense against cutaneous insults and infection. However, under abnormal circumstances, CLA-positive T cells mediate many inflammatory skin disorders, including psoriasis, atopic dermatitis, and cutaneous graft-vs-host disease.[4] Although often referred to as a “primary cutaneous lymphoma” to distinguish it from, eg, Hodgkin’s disease, which may eventually involve the skin, CTCL is perhaps more accurately viewed as a lymphoma arising from skin-homing, CLA-positive memory T cells of the skin immune system.

Understanding that CTCL is a malignancy of the skin immune system helps one to better understand the pathogenesis of this lymphoma. For example, while the insidious patch/plaque CTCL (discussed below) usually persists for years in patients in the absence of obvious leukemia (ie, with normal peripheral lymphocyte subset counts and without any abnormalities present by flow cytometry) or lymphadenopathy, it is incorrect to view this condition as a lymphoma confined to the skin.

It is not unusual for such patients to present with multiple cutaneous patch and/or plaque lesions. Immunohistochemical and molecular analysis of infiltrating T cells demonstrates a common clonal population of CD4-positive, CD45Ro-positive memory T cells in each of the discrete lesions. (Each patch/plaque lesion of mycosis fungoides has the same clone.) The presence of multiple lesions with intervening normal skin is indirect evidence that, even in this early-stage patch/plaque disease, CTCL is a lymphoma of T cells of the skin immune system (ie, cycling among the skin, peripheral blood, and skin-draining lymph nodes).

In patients with early patch/plaque CTCL, in whom the vast majority of malignant T cells home to and reside in the skin, it is possible, using PCR, to detect the same malignant clone in the peripheral blood and lymph nodes, thus confirming that this is a “systemic” lymphoma.[5] At this stage, skin-directed therapy (eg, psoralen plus ultraviolet A light [PUVA], electron-beam radiation, or topical carmustine(Drug information on carmustine) [BCNU]) may be sufficient to prevent disease progression or induce apparent cure. This suggests the strong tendency of the malignant T cells to home to the skin, as well as their dependence on the cutaneous environment to maintain their growth.

In more advanced stages of CTCL, the malignant T cells apparently lose their dependence on the skin environment, and more of them are found in the peripheral blood (ie, “leukemic” CTCL) and lymph nodes. Under such circumstances, skin-directed therapies alone are insufficient to control disease, and systemic therapy is warranted.

While the presence of a malignant clone of CLA-positive, CD4-positive cutaneous T cells can be demonstrated in the skin of patients with patch/plaque CTCL, it is important to note that these cells usually represent a minority of the T-cell dermal infiltrate, which includes other nonmalignant activated CD4-positive, as well as CD8-positive, T cells. Several clinical findings have supported the hypothesis that the malignant cells of CTCL are immunogenic. CD8-positive tumor-infiltrating lymphocytes present within CTCL lesions tend to be more plentiful in early-stage disease. Their proportion correlates positively with improved survival, suggesting that they may exert an antitumor host response.[6]

In addition, cyclosporine (Neoral, Sandimmune), a potent immunosuppressive agent that is specific for T cells, can promote rapid disease progression, possibly through the downregulation of antitumor T cells.[7,8] Perhaps most importantly, results of treatment of CTCL with photopheresis (discussed below) are consistent with the stimulation of an antitumor immune response.[9]

Clinical and Histologic Features

Parapsoriasis

The terminology and delineation of the various forms of parapsoriasis are unclear. A pattern of large, atrophic, erythematous plaques on the trunk

with histologic features that are nondiagnostic of CTCL has been called large-plaque parapsoriasis. Treatment of large-plaque parapsoriasis, most effectively with PUVA, will generally clear these lesions. The apparent progression over years of a portion of these cases to obvious CTCL, if left untreated, and the ability to detect clonal dominance by PCR, suggests that large-plaque parapsoriasis is, in fact, patch/plaque CTCL from the outset, rather than a “premaligant” condition.[10-12]

This fact notwithstanding, it appears that patients who present with much smaller lesions, as in so-called small-plaque parapsoriasis, do not readily progress to CTCL, although a dominant T-cell clone may be detectable by PCR.[5,13] Since T-cell clonality may be detectable in several other T-cell–mediated inflammatory conditions of the skin, such as pityriasis lichenoides et varioliformis acute (PLEVA; Mucha-Habermann disease)[14], without any evidence of progression to CTCL, molecular testing must be interpreted in light of the clinical and histologic findings.