Dyscrasias With “Undetermined Significance”

Sézary and Bouvrain,¹ at the Séance du 13 Février 1938 de la Société Française de Dermatologie et de Syphilographie, described a 58-year-old woman with “erythrodermie avec présence de cellules monstrueuses dans le derme et le sang circulant.” The patient also had pruritus, edema, lymphadenopathy, and hair loss.

Gniadecki and Lukowsky describe a group of patients with chronic recalcitrant erythroderma accompanied by a monoclonal expansion of CD4⁺7⁻26⁻29⁺ T-lymphocytes and designate them as having “erythroderma with monoclonal T-cell dyscrasia of undetermined significance (MTUS-E).”² These patients fulfill the criteria of the pre-Sezary syndrome, described by Winkelmann et al.³ Clinically, all patients had chronic recalcitrant erythroderma, but none developed any hematological malignancy during their lifetime or fulfilled the criteria of cutaneous lymphoma or Sezary syndrome.

The interesting hypothesis of MTUS-E being a nosologic entity raises several questions for discussion.

The concept of preneoplastic conditions showing clinical, histological, or laboratory findings of undetermined significance is not new.

These conditions point to the fact that distinct conditions exhibiting B-cell or T-cell clonality bridge the gap between polyclonal inflammatory reaction and monoclonal neoplastic proliferations. These conditions are the link between inflammation and cancer.^{11 - 12} The pathogenetic evolution has been best investigated for colon cancer, evolving from adenoma over a series of mutations under chronic stimulatory influence of known (microbial) or yet unknown pathogenetic factors,¹³ but also for other cancers.¹⁴

The clinical features of lymphoproliferative infiltrates are misleading in lymphadenosis benigna cutis, which may simulate a nodule in cutaneous B-cell lymphoma. Eczematous lesions in conditions referred to as parapsoriasis¹⁵ simulate the early eczematous stage of mycosis fungoides, and their significance in terms of a rarely ocurring transformation into overt lymphoma is undetermined. Erythroderma is not diagnostic per se and can be seen in psoriasis, lichen planus, drug eruption, atopic dermatitis, pityriasis rubra pilaris, and several other conditions.

Histological features, which are often considered the diagnostic gold standard, may include criteria that are insignificant and that, even though being highly specific in most cases, can be present in nonneoplastic conditions. Epidermotropism of single cells or of small accumulations of lymphoid cells (Pautrier microabscesses) can be found in eczematous, psoriasiform, and lichenoid conditions. The significance of this feature has been demonstrated in a study of early mycosis fungoides in which the most important diagnostic feature was the presence of lymphocytes with extremely convoluted medium to large nuclei, singly or clustered in the epidermis and in small sheets in the dermis.¹⁶ However, atypical lymphoid cells with a highly pathologic nuclear configuration, reflected also in the nuclear contour index, may be present in the cerebrospinal fluid of patients with Sjögren syndrome.¹⁷ The risk of developing non-Hodgkin lymphoma is highly increased in patients with Sjögren syndrome. These lymphomas are usually of B-cell origin.¹⁸ There are few reports of T-cell lymphoma associated with Sjögren syndrome.^{19 - 20} Sezary-type cells have been demonstrated in rheumatoid synovial fluid,²¹ and mutilating arthritis has been associated with CTCL.²² The finding of atypical lymphoid cells in these conditions also represents histological features of unknown significance.

Phenotypic and genotypic features per se are often of unknown significance but may be indicative of malignancy. Clonal expansions of T cells carrying identical T-cell receptor genes are the hallmark of T-cell malignancies, but they can also result from a strong immune reaction to a dominant epitope.²³ Dominant clonality has been detected in some cases of histologically nonspecific dermatitis (clonal dermatitis).²⁴ Magro et al²⁵ found that lymphoid infiltrates showed a reproducible CD7 and/or CD62 K deletion in conjunction with T-cell clonality and variable CD30 positivity findings. Rashes resolved or improved substantially after drug treatment. They refer to this clonal lymphomatoid dermatitis with expansion of CD7-negative and CD62 K-negative, activated memory T-lymphocytes as drug-induced, reversible lymphoid dyscrasia.

Clonality established by T-cell receptor gene analysis is sufficiently sensitive to be of value in routine diagnosis. However, immunophenotypic data showing an expanded CD4⁺/CD7^– population, an elevated CD4/CD8 ratio, or restricted Vβ expression are not specific to T-cell malignancy and are not recommended as the sole diagnostic criterion for the differentiation of Sezary syndrome from erythrodermas of nonneoplastic origin.²⁶

There is increasing evidence that cancerogenesis is not a big-bang event but a stepwise evolutionary process due to the accumulation of mutations in DNA repair genes, oncogenes, or tumor suppressor genes.

Modern concepts on the clonal evolution of CTCL starting from chronic dermatitis to clonal dermatitis, early CTCL, advanced CTCL, and transformed CTCL²⁷ further support early quantitative cytochemical studies of the characterization of the cellular infiltrate in the various stages of mycosis fungoides.²⁸ Cutaneous T-cell lymphoma may evolve by multilineage progression, and tumor subclones in mycosis fungoides can be detected in early disease stages by mutation analysis of microsatellite DNA obtained from multiple microdissected areas.²⁹

Provided that neoplastic disorders including malignant lymphoma evolve from a preneoplastic stage, the question is, Which mechanisms underlie the process of neoplastic transformation of cells and infiltrates? Most probably we are dealing with a multifactorial and multistep process due to the impact of various etiologic factors over a long period, starting as a hyperreactive inflammatory process. Deficits in cell proliferation regulation and defective oncogene and/or suppressor gene expression later promote transition from preneoplastic conditions to neoplasia. Genetic, environmental, infectious, and immunologic factors are potential initiators of this process.

For CTCL lymphomagenesis, we proposed the following hypothesis.³⁰ Abnormal but not primarily neoplastic lymphocytes showing genomic instability (genotraumatic lymphocytes)^{31 - 32} are driven into activation and reactive cell proliferation by antigenic or irritative³³ stimulation. Primary CD30-positive CTCL associated with chronic burn injury has been reported in a patient with long-standing psoriasis. The risk of mutation occurrence in the susceptible genotraumatic cell clone increases with every new cell division. The accumulation of mutations is usually limited by controlling mechanisms leading to cell death to prevent cells with chromosomal aberrations from unlimited neoplastic expansion. The most important controlling mechanism is programmed cell death (apoptosis), which can be blocked by increased Bcl-2 protein expression³⁴ due to Bcl-2 gene mutation or translocation.

The central role of nuclear factor–κB has been indicated by several studies³⁵ and has recently been elucidated as a tumor promoter in inflammation-associated cancer.^{11 - 12} The presence of neoplastic lymphoid stem cells in the bone marrow and the emergence of cutaneous lymphomas by spread of these cells into the skin have also recently been advocated based on a review of published data concerning this issue and hypothetical discussion of clinical and nosologic observations.³⁶ However, this concept, which explains some of the phenomena we see in CTCL (eg, multicentricity of skin lesions), is highly hypothetical and leaves some questions unanswered.

The term lymphoid dyscrasia is clinically relevant to designate conditions carrying features of malignancy but lacking the full-blown biological features of distinct malignant lymphoid neoplasias. It will be important to determine the long-term risk of patients with T-cell dyscrasias.

Correspondence: Dr Burg, Department of Dermatology, University Hospital Zürich, CH-8091 Zürich, Gloriastr 31, Switzerland (g.burg@usz.ch).

Accepted for Publication: November 15, 2004.

Financial Disclosure: None.