Search tips
Search criteria 


Logo of jclinpathJournal of Clinical PathologyVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Clin Pathol. 2007 September; 60(9): 1061–1064.
PMCID: PMC1972422

Follicular lymphoma with trisomy 18 exhibiting loss of BCL‐2 expression on transformation to a large cell lymphoma

Over‐expression of BCL‐2 in follicular lymphoma is commonly caused by a t(14;18) chromosomal translocation. Follicular lymphomas that do not carry the t(14;18) translocation but are positive for BCL‐2 protein may have other cytogenetic abnormalities, including BCL‐2 amplification, reciprocal translocations involving the IGK/IGL loci1 and trisomy 18.2,3

Take‐home message

Expression of BCL‐2 protein in a follicular lymphoma carrying three copies of the BCL‐2 gene (due to trisomy 18) was lost following transformation to a high‐grade lymphoma despite persistence of the genetic abnormality.

Transformation of follicular lymphoma to diffuse large B cell lymphoma is generally associated with a poorer prognosis.4,5 However, there have been no extensive studies in the literature of the BCL‐2 gene and protein status in transformed diffuse large B cell lymphoma arising from BCL‐2‐positive follicular lymphoma.

We describe a case of follicular lymphoma carrying trisomy 18 (without evidence of translocations involving the BCL‐2 locus), associated with strong BCL‐2 protein expression. However, the same biopsy sample contained an area of diffuse large B cell lymphoma in which BCL‐2 protein was absent despite retaining the same genetic abnormality. As far as we know, there have been no other reports of BCL‐2 protein that is expressed in follicular lymphoma but that ceases to be detectable after transformation.

Case report

A female patient presented with a localised enlargement of a submental lymph node. No significant abnormalities were detected in her blood count and in the staging investigations.

Excision biopsy revealed an enlarged and effaced lymph node. Three distinct areas, lying adjacent to each other (fig 1A,B1A,B),), could be identified from their morphology and phenotype. One area comprised a classical follicular lymphoma, where the neoplastic follicles contained an admixture of centrocytes with smaller numbers of centroblasts, consistent with a grade 2 follicular lymphoma. The second area showed larger follicles that were less clearly demarcated, although their cellular composition resembled that seen in the first area. A third smaller area of the tumour consisted of diffuse large B cell lymphoma comprising largely centroblastic cells. These three areas are marked 1, 2 and 3 in fig 1A1A.

figure cp43034.f1
Figure 1 (A) Left: low power view of the lymph node immunostained for BCL‐2 showing three distinct areas: (1) small follicles, (2) large follicles and (3) diffuse large cell area. BCL‐2 is positive in area 1 (small follicles) but ...

All components were positive for B cell markers (CD20, CD79a), for BCL‐6, BLIPM1, FOXP1, p53 and, focally, for CD10. MUM1 showed different levels of labelling in the three areas: negative in area 1, approximately 20% of cells positive in area 2 and >50% in area 3. CD21 revealed expanded follicular dendritic cell meshworks typical of follicular lymphoma in area 1 but not in areas 2 and 3 (fig 1B1B).). Immunostaining for PU.1 was negative in all components.

The pattern of BCL‐2 expression was unusual since it was positive in area 1 (small follicles) but was negative in areas 2 (large follicles) and 3 (diffuse large cell) (fig 1A,B1A,B).). In order to confirm that BCL‐2 negativity was not an artefact of immunostaining due to an acquired mutation within the BCL‐2 gene (as described by Schraders et al6), its expression was analysed using two monoclonal antibodies (reagents 124 and C2 from Dako, Denmark, and Santa Cruz Biotechnology Inc, USA, respectively) reactive with different regions of the protein. Both antibodies showed similar staining patterns; thus the presence of an acquired mutation was deemed unlikely.

Fluorescence in situ hybridisation (FISH) was performed on paraffin sections as previously described7 to ascertain the status of the BCL‐2 gene and of other loci commonly involved in lymphoma pathogenesis (Table 11).). The dual‐fusion assay for the t(14;18) translocation showed no evidence of the reciprocal translocation but at least one extra BCL‐2 signal was present in all areas of the biopsy, in addition to the normal pattern of signals (fig 1A1A).). Absence of a translocation involving the BCL‐2 locus was confirmed by the normal pattern (that is, non‐separated signals) observed using the BCL‐2 break‐apart probe. This probe set also showed an extra BCL‐2 signal (fig 1A1A).). The MALT1 break‐apart probe also showed the normal pattern accompanied by an extra signal in all parts of the tumour. Variant translocations t(2;18) and t(18;22) were excluded using the IGK and IGL break‐apart probes, respectively, which showed normal signal patterns. A centromeric probe for chromosome 18 showed a gain of the centromeric region of chromosome 18 in all three areas of the tumour (fig 1A1A).). A search for other chromosomal abnormalities was negative in all three areas. Table 11 summarises the results of the FISH analysis, indicating that the tumour harbours the same chromosomal abnormalities in all three areas.

Table thumbnail
Table 1 Results of immunostaining for CD20, BCL‐2, Ki67, CD10, BCL‐6, BLIMP1, FOXP1, PU.1 and MUM1 together with FISH analysis for chromosomal aberrations in the three different areas of the tumour

Interestingly, the proliferation fraction showed an inverse relationship to BCL‐2 in areas 1 and 3 but not in area 2 (Table 11 and fig 1B1B).). Thus, in area 1, BCL‐2 was positive and Ki67 was low (approximately 10% of neoplastic cells), but in area 3 the opposite was observed—that is, BCL‐2 was negative and Ki67 was high (approximately 80%). However, this inverse relationship was not observed in the second area since Ki 67 reactivity was low.

The patient was treated with a combination of chemotherapy and local radiotherapy. She remains healthy 6 months after completing the treatment.


This case presented with several interesting features. Histologically, there were three stages of tumour within the lymph node—small follicles, poorly formed large follicles and diffuse large cell lymphoma—and these were interpreted as representing progression of an initially low‐grade tumour. Although the clonal relation between the three lymphoma components was not proven, it is very unlikely that these components arose independently since they were all carrying the same genetic abnormalities.

Trisomy 18 is usually present as a component of complex cytogenetic changes8 and is rarely the primary karyotypic change.9 In this report we were not able to demonstrate any additional cytogenetic abnormalities, including c‐MYC rearrangement (which is frequently associated with transformation). It is possible that the gain of chromosome 18 was responsible for the BCL‐2 expression2,3,9 in the follicular component (area 1), but as the cells transformed, the association between trisomy 18 and BCL‐2 was for some reason lost. However, there has not been any convincing evidence that trisomy 18 drives the BCL‐2 expression in follicular lymphoma. Other mechanisms, including activation of NF‐κB,10 has been implicated in regulation of BCL‐2 expression in B cell lymphomas. Our own (unpublished) observations, using a double immunofluorescence technique, have shown that in follicular lymphoma, proliferating cells do not express BCL‐2. However, a similar explanation cannot be offered for the loss of BCL‐2 expression in the large follicles (area 2) since the neoplastic cells exhibited only a low proliferation fraction.

In summary, this case indicates the possibility of BCL‐2 down‐regulation when follicular lymphoma transforms. It would be of interest to determine, by examining sequential biopsies, whether this phenomenon can occur in transformed cases of t(14;18)‐positive follicular lymphomas. The clinical implication of transformation‐associated loss of BCL‐2 protein expression is that the larger cell component should be more susceptible to chemotherapy.


Competing interests: None.


1. Bentley G, Palutke M, Mohamed A N. Variant t(14;18) in malignant lymphoma: a report of seven cases. Cancer Genet Cytogenet 2005. 15712–17.17 [PubMed]
2. Pezzella F, Gatter K C, Mason D Y. et al Bcl‐2 protein expression in follicular lymphomas in absence of 14;18 translocation. Lancet 1990. 3361510–1511.1511 [PubMed]
3. Horsman D E, Okamoto I, Ludkovski O. et al Follicular lymphoma lacking the t(14;18)(q32;q21): identification of two disease subtypes. Br J Haematol 2003. 120424–433.433 [PubMed]
4. Bastion Y, Sebban C, Berger F. et al Incidence, predictive factors, and outcome of lymphoma transformation in follicular lymphoma patients. J Clin Oncol 1997. 151587–1594.1594 [PubMed]
5. Ersboll J, Schultz H B, Pedersen‐Bjergaard J. et al Follicular low‐grade non‐Hodgkin's lymphoma: long‐term outcome with or without tumor progression. Eur J Haematol 1989. 42155–163.163 [PubMed]
6. Schraders M, de Jong D, Kluin P. et al Lack of Bcl‐2 expression in follicular lymphoma may be caused by mutations in the BCL2 gene or by absence of the t(14;18) translocation. J Pathol 2005. 205329–335.335 [PubMed]
7. Ventura R A, Martin‐Subero J I, Jones M. et al FISH analysis for the detection of lymphoma‐associated chromosomal abnormalities in routine paraffin‐embedded tissue. J Mol Diagn 2006. 8141–151.151 [PubMed]
8. Mohamed A N, Palutke M, Eisenberg L. et al Chromosomal analyses of 52 cases of follicular lymphoma with t(14;18), including blastic/blastoid variant. Cancer Genet Cytogenet 2001. 12645–51.51 [PubMed]
9. Wong K F, Chan J K. Follicular lymphoma with trisomy 18 and over‐expression of BCL2 in the absence of t(14;18)(q32;q21). Cancer Genet Cytogenet 2000. 12352–54.54 [PubMed]
10. Tracey L, Perez‐Rosado A, Artiga M J. et al Expression of the NF‐kappaB targets BCL2 and BIRC5/Survivin characterizes small B‐cell and aggressive B‐cell lymphomas, respectively. J Pathol 2005. 206123–134.134 [PubMed]

Articles from Journal of Clinical Pathology are provided here courtesy of BMJ Publishing Group