The BCL6 gene encodes a POZ/Zink finger sequence-specific transcription inhibitor important for T- and B-cell maturation, cell-cycle control, apoptosis, and inflammation. Recent studies indicate that high BCL6 expression correlates with good prognosis for diffuse large B-cell lymphoma (DLBCL) patients.1,2  Moreover, mutations and polymorphisms, such as the BCL6397G/C polymorphism, in the 5′ regulatory region of the BCL6 gene have been suggested to increase BCL6 expression and to be involved in lymphoma progression and transformation from follicular lymphoma (FL) to diffuse large B-cell lymphoma,3-5  whereas no correlation was evident for de novo DLBCL.6  DLBCL can now be classified into 2 major subgroups: one derived from germinal-center B cells, the GC subtype, often with a high expression of BCL6, and one derived from B cells, often with a lower expression of BCL6, the non-GC subtype.1,2 

To investigate the impact of the BCL6397G/C polymorphism, we screened 120 de novo DLBCLs from Uppsala and Umeå University Hospitals regarding their BCL6397G/C genotype and compared the allele and genotype frequencies with samples from 230 healthy donors. Subclassification into GC and non-GC DLBCL according to Hans et al1  was available for all 120 tumors.

In the present material, the frequency of the BCL6397C allele did not differ significantly in our DLBCL patients compared with the control group, which is in accordance with previous data.6  Furthermore, we did not find any difference in overall survival between patients with the BCL6397GG genotype and BCL6397GC/CC genotypes. No significant difference was indicated between the 2 groups concerning sex, clinical stage, or age.

Interestingly, the allele frequencies of BCL6397G/C polymorphism were significantly different between GC DLBCL and non-GC DLBCL (P < .001; Table 1) as well as between GC DLBCL and healthy controls (P = .005). The genotype frequencies were also shown to differ between GC DLBCL and non-GC DLBCL (BCL6397GG vs BCL6397GC/CC; P = .001), with a preference for the C allele in the GC group and for the G allele in the non-GC group. We also found a significant difference between GC DLBCL and controls concerning the genotype frequencies (P = .02; Table 1). A tendency to a difference was indicated between non-GC DLBCL and controls with a higher frequency for the G allele in the DLBCL patients (P = .07). Furthermore, 95 patients diagnosed with FL were screened for the BCL6397G/C polymorphism. No significant difference was found concerning allele or genotype frequencies between FL and controls, which is in contrast to the result by Lossos et al.5  The previously reported higher risk of transformation for FL patients with the C allele could not be verified (P = .8).7 

In conclusion, we could not verify earlier findings by Lossos et al5  or Jardin et al7  concerning the associations between the BCL6397G/C polymorphism and FL and transformed DLBCL. However, a highly significant difference in allele frequency (P < .001) of the BCL6397G/C polymorphism was evident between GC-DLBCL and non-GC DLBCL, with a preference for the C allele in the GC group and a preference for the G allele in the non-GC group. Jardin et al7  have suggested that the BCL6397G/C polymorphism may influence the binding of an unknown transcription factor. A possible indirect effect by linkage disequilibrium with another gene was also suggested. Our finding adds to the concept that the GC and non-GC groups of DLBCL are molecularly distinct and probably 2 different types of lymphoma.

The authors declare no competing financial interests.

The authors are grateful to Drs Lars Klareskog and Leonid Padyukov and to the Epidemiologisk Identifiering ar Orsaker Till Reumatoid Artrit (EIRA) group for help with the collection of the healthy control material and to Dr Thomas Axelsson for technical assistance. The single nucleotide polymorphism (SNP) genotyping was performed by the Wallenberg Consortium North (WCN) SNP platform, Department of Medical Sciences, Uppsala University.

M. Berglund designed and performed research, analyzed data, and wrote the paper; R.-M.A. analyzed data and was involved in the histopathologic review; M. Book performed research; R.R. and G.R. collected and analyzed data; and U.T. designed and performed research, analyzed data, and wrote the paper.

1
Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray.
Blood
.
2003
;
103
:
275
-282.
2
Berglund M, Thunberg U, Amini RM, et al. Evaluation of immunophenotype in diffuse large B-cell lymphoma and its impact on prognosis.
Mod Pathol
.
2005
;
18
:
1113
-1120.
3
Lossos IS, Warnke R, Levy R. BCL-6 mRNA expression in higher grade transformation of follicle center lymphoma: correlation with somatic mutations in the 5' regulatory region of the BCL-6 gene.
Leukemia
.
2002
;
16
:
1857
-1862.
4
Lossos IS, Levy R. Mutation analysis of the 5' noncoding regulatory region of the BCL-6 gene in non-Hodgkin lymphoma: evidence for recurrent mutations and intraclonal heterogeneity.
Blood
.
2000
;
95
:
1400
-1405.
5
Lossos IS, Jones CD, Zehnder JL, Levy R. A polymorphism in the BCL-6 gene is associated with follicle center lymphoma.
Leuk Lymphoma
.
2001
;
42
:
1343
-1350.
6
Susova S, Trney M, Soucek P. Single nucleotide polymorphism in 5'-flanking region of BCL6 is not associated with increased risk of non-Hodgkin's lymphoma.
Cancer Lett
.
2006
;
238
:
142
-145.
7
Jardin F, Ruminy P, Parmentier F, et al. Clinical and biological relevance of single-nucleotide polymorphisms and acquired somatic mutations of the BCL6 first intron in follicular lymphoma.
Leukemia
.
2005
;
19
:
1824
-1830.
Sign in via your Institution