Diffuse large B-cell lymphoma transformed from a follicular lymphoma with BCL2 and CCND1 rearrangements
Patient follow up: The patient was unwilling to undergo further aggressive chemotherapy and was transitioned to hospice and died of the disease 2 months following the diagnosis of transformation to large B-cell lymphoma.
The IGH/CCND1, t(11;14)(q13;q32) is considered as a molecular hallmark for mantle cell lymphomas (MCL) and is observed in >95% of MCL cases (1) .The IGH/ CCND1 rearrangement results in deregulation and overexpression of the G1-phase cell cycle gene, CCND1 that encodes cyclin D1, mediated by enhancer elements found throughout the IGH locus(2). Nuclear cyclin D1 expression is also seen in >95% of mantle cell lymphomas and only very rare cases of mantle cell lymphomas with CCND1 rearrangement lack cyclin D1 expression and conversely only very rare cases of MCL with cyclin D1 expression do not show a demonstrable CCND1 rearrangement (1). However, cyclin D1 expression is not specific for MCL, and has been seen in ~15% of plasma cell neoplasms usually associated with CCND1 rearrangements (1, 3). Other B-cell lymphomas that may be cyclin D1 positive without CCND1 rearrangements include hairy cell leukemia (50-70% of cases), ~1-4% of large B-cell lymphomas, rare cases of chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) and rare cases of follicular lymphoma (FL) (3, 4). The proliferation centers in CLL/SLL may be more commonly cyclin D1 positive. Cyclin D1 expression is also seen in normal histiocytes and endothelial cells which may be admixed with lymphomas, but normal lymphoid cells are cyclin D1 negative.
The frequency of cyclin D1 positivity in large B-cell lymphomas is low and has been reported in different studies to range from 1.5% to 4.3%. Ok et al reported a frequency of 2.1% in a large cohort of 1435 DLBCL patients (5). The cyclin D1 expression in these cases was attributed to the altered copy numbers, gains, or amplification of CCND1 in few cases; however the mechanism of cyclin D1 expression in majority of these cases was unclear with lack of CCND1 rearrangements. On comparison with the cyclin D1 negative DLBCL, the cyclin D1 positive DLBCL was seen at a younger age and more common in men, but no other significant differences in clinical presentation, pathologic features, overall survival, or progression-free survival was identified between these two subgroups of patients (5). In this study all the cyclin D1 positive cases were CD5 negative. These cases more commonly show centroblastic morphology, are usually CD10 negative, BCL6 and MUM1 positive and are negative for SOX11. This immunophenotype is similar to the case presented here and is consistent with a non-germinal center phenotype as per the Hans algorithm; however, in the study by Ok et al 56.7% of their cases were classified as GCB type by gene expression profiling (5).
CCND1 rearrangements are also not restricted to MCL and have been rarely reported at relapse in some B-cell lymphomas including diffuse large B-cell lymphoma (DLBCL), CLL/SLL and FL. In these lymphomas the data suggests that the CCND1 rearrangement may present a secondary event during lymphoma evolution, unlike in MCL where it is considered as a primary genetic event (3).
Transformation or progression of FL, usually to DLBCL, occurs in 25-35% of patients with FL. Transformation of FL is associated with acquisition of additional genetic alterations that lead to deregulation of the cell-cycle progression and DNA damage responses and most commonly include loss of CDKN2A/CDKN2B, MYC alterations (most commonly rearrangements), and TP53 aberrations leading to loss of function, as well as aberrant somatic hypermutation (6). In particular, acquisition of MYC rearrangement; leading to a combination of BCL2 and MYC rearrangements, diagnosed as high-grade B-cell lymphoma with MYC and BCL2 rearrangements (so-called " double hit" lymphomas) is associated with a particularly aggressive course(1). Recently cases of high grade B-cell lymphomas with MYC and BCL2 and/or BCL6 gene rearrangements along with CCND1 rearrangements, so-called " quadruple hit" lymphomas have been reported, where CCND1 rearrangements are considered as a secondary transformational event (3).
B-cell lymphomas with translocations involving with 11q13 (CCND1), 18q21 (BCL2) and/or 3q27 (BCL6) but lacking the MYC (8q24) translocations were studied by Kodru et al (7). 15 out of 119 cases with at least 2 of the 3 translocations previously mentioned, showed both 11q13 and 18q21 rearrangements and included eight FL, four DLBCL, two CLL/SLL, and one MCL. Cyclin D1 positive large B-cell lymphomas with CCND1 and BCL6 rearrangements have been reported (8, 9), with one of the cases reported by Al-Kawaaz et al showing the CCND1 rearrangement only at relapse of DLBCL (8). In these cases the presence of the BCL6 rearrangement, along with the absence of SOX11 expression was considered by the authors to support the diagnosis of DLBCL over MCL.
There are limited data regarding the prognosis and survival of these rare primary or secondary lymphomas with CCND1 and BCL2 and/or BCL6 gene rearrangements; however in the limited case reports (8, 10), these appear to have a better prognosis than the double or triple hit lymphomas with MYC rearrangements.
BCL6 (3q27) rearrangements are uncommon in MCL and are associated with BCL6 expression(1). In the present case BCL6 IHC was positive and the FISH studies showed loss of 5' signal of BCL6 in 42% of the cells, which may be seen in BCL6 rearrangements. A similar pattern in BCL6 FISH also reported in a case of CCND1+ DLBCL by Gao et al with 39-52% of interphase nuclei examined showing an atypical pattern for the BCL6 locus consistent with either atypical rearrangement or partial deletion of the locus (9).
These rare cases of cyclin D1 positive large B-cell lymphoma with CCND1 rearrangements raise a differential diagnosis of DLBCL vs pleomorphic MCL which can morphologically overlap and present a diagnostic challenge with variable immunophenotypes described in both. The distinction between the two is important as pleomorphic MCL has a worse prognosis and requires more aggressive therapy than RCHOP, which is routinely used for DLBCL. SOX11 is positive in MCL and negative in cyclin D1-positive DLBCL and is a useful marker in differentiating the two (11). CyclinD1+ DLBCLs usually show weaker cyclin D1 staining than MCLs and most are CD5 negative. In our case, in view of the patient's history of follicular lymphoma, lack of SOX11 expression together with absence of CD5, this is considered as an unusual DLBCL transformation from a FL with CCND1, BCL2 and possibly also BCL6 rearrangements.
In conclusion, this case of an unusual FL transformation to DLBCL with CCND1 rearrangement highlights the diagnostic challenges with classification of these lymphomas and the need for a comprehensive work up of such cases including IHC for CD5, CD10, BCL6, MUM-1 and SOX11, and FISH for CCND1, BCL2, BCL6, and MYC for complete characterization and to differentiate between pleomorphic variants of MCL and DLBCL or high-grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangements.
Contributed by Shweta Bhavsar, MBBS, MD and Nathanael G. Bailey, MD