Atypical cutaneous lymphocytic infiltrate and a role for quantitative immunohistochemistry and gene rearrangement studies

Defining early mycosis fungoides

This editorial review summarizes the results of 5 meetings sponsored by the International Society for Cutaneous Lymphoma at which the clinicopathologic and ancillary features of early mycosis fungoides were critically examined. Based on this analysis, an algorithm was developed for the diagnosis of early mycosis fungoides involving a holistic integration of clinical, histopathologic, immunopathologic, and molecular biological characteristics. A novel aspect of this algorithm is that it relies on multiple types of criteria rather than just one, for example, histopathology. Before its finalization, the proposed diagnostic algorithm will require validation and possibly further refinement at multiple centers during the next several years. It is anticipated that a more standardized approach to the diagnosis of early mycosis fungoides will have a beneficial impact on the epidemiology, prognostication, treatment, and analysis of clinical trials pertaining to this most common type of cutaneous lymphoma.

Histopathologic Features of Early (Patch) Lesions of Mycosis Fungoides

The histologic diagnosis of early mycosis fungoides (MF) is one of the most vexing problems in dermatopathology. We reviewed the histopathologic features of 745 biopsy specimens from 427 patients (male:female = 277:150; median age, 52 years; range, 3-95 years) with early (patch) lesions of MF collected from the lymphoma database of the Department of Dermatology of the Medical University of Graz (Austria). In all patients, the diagnosis was established by clinicopathologic correlation. The most common histopathologic pattern consisted of a band-like or patchy lichenoid infiltrate admixed with coarse bundles of collagen in the superficial dermis. Epidermotropism of lymphocytes was observed in most cases in one or more forms (single lymphocyte epidermotropism, 22%; basilar lymphocytes, 23%; Pautrier's microabscesses, 19%; "haloed" lymphocytes, 40%; disproportionate exocytosis, 17%; pagetoid epidermotropism, 3%). In 4% of cases, epidermotropism was completely missing. Atypical lymphocytes were present only in 9% of cases. Features of interface dermatitis were observed in 59% of cases. Other unusual findings were the presence of necrotic keratinocytes (23%), melanophages (8%), and extravasated erythrocytes (4%). In 28 patients, two or more biopsies taken on the same day at different body sites showed different histopathologic aspects, underlying the protean features of MF even in a single patient at a given time. Our study expands previous observations on histopathologic features of early lesions of MF. Although sometimes the histopathologic features are not diagnostic, they should be considered consistent with MF and do not rule out the diagnosis.

Low CD7 expression in benign and malignant cutaneous lymphocytic infiltrates: experience with an antibody reactive with paraffin-embedded tissue

Loss of CD7 expression by neoplastic lymphocytes is considered a distinguishing characteristic of mycosis fungoides (MF) and cutaneous T-cell lymphoma. Reports to date examining for the CD7 immunophenotype in MF have been performed on fresh-frozen tissue. In this study, we used a paraffin-reactive antibody directed against CD7 to determine its range of expression in MF and to compare these results with those in controls. Examining 22 cases of MF and 61 controls, we found minimal CD7 expression by lymphocytes in MF and in a few cases of benign inflammatory dermatosis (BID). The lowest mean CD7 counts (as a percentage of total lymphocytes) were found in MF (patch stage: 5% +/- 5%, range: 0-10; plaque and tumor stages: 15% +/- 5%, range: 5-25), and these counts were significantly lower than those for BID (35% +/- 20%, range: 5-80; p = 0.001). By logistic regression analysis, low CD7 expression (<10% lymphocytes labeling) had sensitivity and positive predictive values of 80% and 72%, respectively, and specificity and negative predictive values of 93% and 96%, respectively, for the diagnosis of patch stage MF. False-positive results were found for spongiotic dermatitis. Moreover, spongiotic dermatitides exhibited a progressive decrease in mean CD7 counts from acute to subacute to chronic stages (50% versus 35% versus 30%, respectively). In conclusion, minimal CD7 expression is a specific finding for MF. Benign inflammatory infiltrates can also show low CD7 expression, however, which rarely matches that of patch stage MF. Progressive loss of CD7 expression in BID is the likely consequence of expansion of antigen-selected CD3+CD4+CD7- T cells. These inflammatory CD4+CD7- T cells may represent the physiologic counterpart to the neoplastic lymphocyte of MF.

A new monoclonal anti-CD7 antibody reactive on paraffin sections

CBC.37 monoclonal antibody (mAb) was generated using balb/c mice immunized with CEM T cell line. It was selected because of its strong reactivity on T lymphocytes on paraffin tissue sections. The anti-CD7 specificity of CBC.37 mAb was assessed by immunohistochemistry, cross-blocking, and cross-immunoprecipitation experiments using CBC.37 and the anti-CD7 mAb DK24. CBC.37 mAb immunoprecipitated a 40-kDa protein. Cross-blocking and cross-immunoprecipitation experiments demonstrated that the two antibodies recognized the same molecule. Immunostaining of a large number of reactive lymph nodes and B and T cell lymphomas confirms that CBC.37 mAb was directed against T cells. As expected, on reactive lymph nodes the staining pattern was comparable to that of CD3. Among the 110 T cell lymphomas examined, all T lymphoblastic lymphomas were positive (15+/15; 100%). As a result of the frequent loss of CD7 antigen, only 25+/95 (26%) of peripheral T cell neoplasms were found to be positive for CBC.37. A marked reduction in the number of CBC.37-positive T cells was observed in 7 of the 60 cases of benign inflammatory dermatoses studied (approximately 12%). CBC.37 was unreactive with all healthy and neoplastic non-lymphoid samples examined. Because the lack of CD7 expression in T cell lymphomas is of diagnostic value, CBC.37 mAb in association with other anti-T cell antibodies working on paraffin sections could be of particular value in asserting the diagnosis of T cell lymphomas in routine histopathology.

p53 tumor suppressor gene mutations in transformed cutaneous T-cell lymphoma: a study of 12 cases

The transformation of cutaneous T-cell lymphoma (t-CTCL) is an uncommon phenomenon that is associated with histopathologic changes and follows an aggressive course. The factors contributing to this transformation are poorly understood. The aim of this study was to analyze the p53 status in t-CTCL and to correlate it with disease outcome. The p53 status was investigated by immunohistochemistry, single-strand conformation polymorphism (SSCP) and DNA sequencing in 12 patients with t-CTCL. Eight mutations were detected; including four in exon 5, one in exon 6 and three in exon 7. Five were point mutations and three were deletions. Paired samples from nontransformed patch and plaque lesions showed no p53 over-expression. Eight disease-related deaths were reported, six to 23 months after transformation, all of which had p53 mutations. Three other patients with wild phenotype (WT-p53) were last reported alive with the disease 19-33 months after transformation (p < 0.0002). One other case had a p53 mutation but a short period of follow-up. Our results suggest that phenotypic changes of t-CTCL are frequently associated with genotype alterations in the p53 gene. Because 70% of the mutations detected were either G to C transversions or deletions, nucleotide-pairing mismatch and not DNA damage by UVB represents a likely mechanism for mutagenesis. Furthermore, the data may help in the design of gene transfer therapies that target the p53 molecule.