Colocalization of BCL2-positive and -negative follicular lymphoma

Abstract

Composite lymphomas are regularly reported and several combinations of lymphoma cell proliferations can be observed. Follicular lymphoma in situ (FLIS) was first recognized by Elaine Jaffe and her group in 2002.1 In their report, they identified 23 lymph node biopsies with focal germinal centres containing centrocytes staining strongly for BCL2, whereas most of the remaining lymph nodes showed BCL2-negative follicular hyperplasia.1 Of interest, in the series, two cases of FLIS were associated with other low-grade B-cell lymphomas.1 In a series of paired FLIS and manifest follicular lymphoma (FL), Schmidt et al.2 demonstrated that FLIS and manifest FL cases were clonally related on the basis of the immunoglobulin heavy chain (IGH) rearrangement patterns and BCL2 breakpoint sequences. Two years earlier, in a very elegant study, Bonzheim et al.3 reported a case of FLIS with BCL2 expression synchronously presenting with a BCL2-negative manifest FL, both lesions being clonally related and carrying the same t(14;18)(q32;q21) translocation. However, the two lesions were observed in two distinct lymph nodes. Other situations may exist as illustrated in this report of an unusual case of BCL2-negative FL containing foci of BCL2-positive tumour cells with mainly FLIS growth pattern altogether in the same lymph node. The case occurred in a 54-year-old woman with stage II disease. The biopsy corresponded to a supraclavicular lymph node involved with grade 1–2 FL (CD20+, CD10+, BCL6+ and BCL2− for the majority of the follicles) (Figure 1a). A careful examination of the haematoxylin and eosin staining revealed two distinct cell populations (Figures 1d and e). The major part of the tumour was composed of large neoplastic follicles made of medium-to-large-sized centrocytes admixed with centroblasts (Figures 1a and d). The latter compartment was classified as grade 2 according to the World Health Organization classification.4 At the periphery of the lymph node, one could notice areas made of small follicles composed of small centrocytes with very rare centroblasts (Figures 1a and e). The latter follicles were strongly stained with anti-BCL2 antibodies (Figures 1b and c). Of note, some circulating BCL2-positive cells were seen within the BCL2-negative follicles and between BCL2-positive and -negative GCs (Figure 1c) suggesting that this lymphoma was not strictly in situ. In addition, one observed a higher expression of CD10 in BCL2-positive follicles (Figure 1f) compared with BCL2-negative ones. This observation has been reported1, 4 but our observation allows a direct comparison of CD10 staining intensity, on the same slide, between BCL2-positive and -negative FL subpopulations. Fluorescence in situ hybridization analysis using a BCL2 break-apart probe (Dako, Glostrup, Denmark) and BCL2-IGH dual probe (LSI IgH Spectrum Green/LSI BCL2 Spectrum Orange, Vysis, Downers Grove, IL, USA) confirmed a BCL2 breakpoint in BCL2-positive and -negative areas confirming the presence of the t(14;18)(q32;q21) translocation (Figures 1d' and e'). The latter finding argues in favour of a clonal derivation of the two lymphoma populations as already demonstrated.2, 3 Because of the low amount of material from the BCL2-positive compartment (a very thin rim of BCL2-positive follicles was present) IGH PCR clonality analysis (Biomed 2 protocols) performed in microdissected tissue remained unsuccessful. BCL2-negative FL have no particular characteristics but recent investigation of molecular signatures revealed that although overall survival did not differ between FL with and without t(14;18), distinct microRNA profiles were observed.4, 5 It is admitted that in around 15% of grade 1–2 FL, immunohistochemical staining for BCL2 remains negative.2, 3, 6, 7 It is also worth noting that in subsets of BCL2-negative FL, significant number of cases carry the IGH-BCL2 translocation in the absence of BCL2 expression.5, 6 Alternatively, the epitope recognized by immunohistochemistry can be mutated and BCL2 detection can give rise to false-negative results.3, 5 However, in our case the two cell populations were presumably clonally related but the most aggressive component did not express BCL2 (we used two different anti-BCL2 clones as recently suggested (Clone 124 Dako and clone SP66 Cell Marque)).6, 7 Our case is the first report of colocalization of BCL2-positive and -negative FL cells in the same lymph node. This report also confirms, on the same tissue section, that CD10 expression is enhanced in BCL2-expressing FL cells. Patient’s consent for this study has been obtained according to the policy of our Institutional Review Board (Hopitaux de Toulouse).