Value of Interphase FISH for the Diagnosis of t(11;14)(q13;q32) on Skin Lesions of Mantle Cell Lymphoma

Pierre Dubus, MD, PhD, Paul Young, MD, Marie Beylot-Barry, MD, PhD, Marc A. Belaud-Rotureau, PhD, Philippe Courville, MD, Béatrice Vergier, MD, PhD, Marie Parrens, MD, Bernard Lenormand, MD, PhD, Pascal Joly, MD, PhD, Jean P. Merlio, MD, PhD

Disclosures

Am J Clin Pathol. 2002;118(6) 

In This Article

Discussion

As recently pointed out by Marti et al,[6] cutaneous involvement by MCL is not well known by physicians and pathologists. Our report emphasizes the need to determine cutaneous MCL features better, as skin was involved initially in 2 of our 3 cases and in the 7 previously documented cases ( Table 1 ). Among the 10 patients, the mean age was 61 years (range, 22-89 years), and the sex ratio was 1.5. Skin lesions frequently were multiple (n = 8) and corresponded to nodules or tumoral or infiltrated plaques or papules. Most lesions were erythematous (n = 8), except in 2 patients with purpuric lesions ( Table 1 ). They almost invariably involved the thorax (n = 8) (breast or back), and less commonly the abdomen (n = 2), the face or scalp (n = 3), or the upper limbs (n = 3). Physical examination identified generalized lymphadenopathy in 7 patients, with hepatosplenomegaly in 2 of them. Therefore, MCL involvement of the skin cannot be considered as primary cutaneous lymphoma according to the EORTC criteria since most patients (n = 8) were at stage IV after initial staging. Two patients described by Bertero et al[4] had isolated skin lesions and were at stage IEN with 1 patient cured and the other developing a generalized MCL 2 years later. Similarly, one of our patients (case 1) had a 2-year history of recurrent maculopapular skin lesions, suggesting that skin was the primary site of MCL for an extended period.

No general or B symptoms have been observed in patients with skin MCL, except in 1 patient in our series, as in the majority of patients with noncutaneous or systemic MCL.[2,3] Interestingly, abnormal absolute PBL counts and/or leukemic forms were not identified in the 7 patients with skin MCL, which is in contrast with 2 of our patients. In noncutaneous MCL, a substantial proportion of patients with blood involvement did not have an elevated lymphocyte count, but the review of the peripheral blood smear generally led to detection of leukemic involvement.[2,3] In 2 of our cases, the parallel molecular study of PBLs revealed the same monoclonal IgH rearrangement as in the skin. Such parallel clonality analysis can be considered as an ancillary technique for the diagnosis of cutaneous B-cell lymphoproliferations as primary or secondary cutaneous disease.[13,31] In cutaneous MCL and systemic MCL, bone marrow infiltration is observed frequently.[2,3] Similarly, a lymph node biopsy was performed in 5 patients with cutaneous MCL, and the histopathologic examination was helpful in confirming the diagnosis of MCL.

On skin specimens, different patterns have been observed with diffuse dermal and subcutaneous infiltrate, as seen for our case 3 and in 2 other patients.[4,6] In cases 1 and 2, as in 5 reported cases,[4,5] a nodular infiltrate of various intensity was observed. In some patients, as in 2 of our patients (cases 1 and 2), the upper dermis contained lymphoid infiltrates surrounding the capillaries. Moreover, in case 1, medium to large lymphoid nodules were seen throughout the dermis and exhibited a bottom-heavy arrangement with deep dermal infiltration. Some of these nodules contained atrophic germinal centers outlined by immunostaining of dendritic cell networks, as described by Bertero et al.[4]

Whatever its abundance, a striking feature was that the infiltrate was monotonous in our cases, although reactive T cells and plasma cells have been observed by others.[4] Indeed, MCL has been described as a monotonous proliferation of small to medium-sized lymphocytes with irregular or cleaved nuclei, inconspicuous nucleoli, dispersed chromatin, and scant, pale cytoplasm.[32,33] Variant forms such as large cell, round cell, or blastoid cell may lead to confusion with other lymphomas. Blastoid morphologic features were observed in our case 2 and in the case reported by Marti et al.[6] Interestingly, the latter was found to express the homing receptor cutaneous lymphocyte-associated antigen, a T-cell ligand for E-selectin.[6] Immunohistochemical analysis is a major tool to confirm the diagnosis of MCL.[29,32,33] Typically, the phenotype is surface IgM+ and/or surface IgD+, CD19+, CD20+, CD5+, CD10-, CD23-, and CD43+. Absence of CD10 and CD23 is useful for distinguishing MCL from follicular and lymphocytic lymphoma, respectively. CD5 immunoreactivity is almost restricted to MCL and lymphocytic lymphoma among small B-cell lymphomas and is not observed in follicular or marginal zone lymphomas. However, CD5 staining may be weak or absent, as observed in skin and lymph node specimens in one of our cases and in up to 21% of systemic MCLs bearing the t(11;14).[34]

Therefore, direct or indirect detection of the t(11;14) has been progressively included among the diagnostic criteria for MCL.[24] Cyclin D1, a cell cycle protein, is not expressed by lymphoid cells at detectable levels, and its overexpression in MCL results from CCND1 deregulation by t(11;14).[28] Although it has been detected in hairy cell leukemia cases, nuclear staining for cyclin D1 is highly characteristic of MCL in the context of an appropriate B-cell phenotype.[35] Some groups have obtained specific nuclear staining in nearly all MCL cases that correlated with the direct detection of the t(11;14) by cytogenetic techniques.[24,28,29] Other groups have reported a lack of sensitivity for cyclin D1 immunodetection, with 21% to 31% negative MCL cases, and the absence of cyclin D1 immunostaining may be problematic when morphologic features and immunophenotype both are suggestive of MCL.[30,34] In our experience, the protein is subject to rapid degradation since delay in fixation or long storage of frozen tissues or slides may lead to a loss of positivity (our personal results). In skin specimens, staining of epidermal basal cells and hair follicle cells of the sweat glands or ducts serves as an internal positive control. Its absence made it possible to exclude skin sections of case 2 for cyclin D1 interpretation.

Difficulties in cyclin D1 immunodetection have led to the development of molecular techniques for the amplification of cyclin D1 messenger RNA or the t(11;14) genomic breakpoint. Semiquantitative or competitive RT-PCR assays make it possible to distinguish between the overexpression of cyclin D1 in MCL and the expression of cyclin D1 transcripts that may be expressed by non-MCL lymphomas or atypical lymphoid hyperplasia.[27,30,34,36] Epithelial samples that contain cells with constitutive expression of cyclin D1 have been discarded from competitive RT-PCR assays.[27] In our study, cyclin D1 overexpression was observed in lymph node and blood specimens of our case 1, but the presence of variable levels of cyclin D1, D2, or D3 transcripts in 6 nonspecific dermatoses clearly demonstrates that such an assay is not applicable to skin specimens. To overcome this problem, a chromogenic in situ hybridization procedure would help to localize cyclin D1 transcripts precisely, with a great sensitivity and specificity, in paraffin-embedded tissues.[30] However, such a technique is not a quantitative assay for the determination of overexpression.

Genomic PCR also may allow the amplification of t(11;14) breakpoints, but its sensitivity is hampered by their scattering on 11q13.[25,26] Between 35% and 60% of t(11;14) breakpoints were amplified in these series,[25,26] as in 1 of our 3 cases. Such a technique remains of interest for monitoring minimal residual disease in positive patients.[37] To detect the t(11;14) directly, we applied interphase FISH analysis to skin sections or touch preparations in parallel with the other available samples for our cases. Detection of the t(11;14) by interphase FISH previously has been validated as a major criterion for the diagnosis of MCL on cytologic preparations from blood, bone marrow, and lymph node samples.[23,24,38,39] It also has been applied successfully after extraction of nuclei from paraffin-embedded material but was not found suitable for paraffin sections owing to the high density of cells or loss of signals within truncated nuclei.[23,24] By performing 2- or 3-µm-thin sections and using a dual color-dual fusion FISH assay, the control of the number of signals per nucleus made it possible to rule out these artifacts. However, touch preparations of frozen material provided better individualization of nuclei, and the interpretation of sections with adequately separated cells was always possible in areas of interest targeted by H&E staining of an adjacent section. On skin sections processed for FISH, the dermal lymphoid infiltrate was directly visible and analyzed, while cells could not reliably be recognized on nuclei preparations.

Skin manifestation of MCL should be diagnosed on the basis of frequent concomitant lymph node, blood, and bone marrow involvement. The specificity of the skin lesions can be assessed on cytologic and phenotypic features. While cyclin D1 immunodetection is a simple and cost-effective technique, interphase FISH proved reliable for detecting the t(11;14). Since it is applicable on sections and on touch preparations, interphase FISH may be of interest for studying other recurrent translocations on skin lymphoid infiltrates.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....