The most frequent t(14;19)(q32;q13)-positive B-cell malignancy corresponds to an aggressive subgroup of atypical chronic lymphocytic leukemia

The translocation t(14;19)(q32;q13) was described as recurrent in B-cell chronic lymphocytic leukemia (CLL) in 1985. The chromosomal rearrangement of the immunoglobulin heavy chain (IGH) gene locus resulted in upregulated expression of a new gene, named BCL3, encoding a protein member of the IκB family.1 It was then reported that t(14;19) was not entirely specific to CLL, because it was also found in other types of chronic B-cell malignancies (for review, see Soma et al.2). The Groupe Francophone de Cytogénétique Hématologique (GFCH) collected 43 chronic lymphoproliferative disorders with t(14;19) or variant BCL3-translocations between 1988 and 2007, to analyze to what extent this uncommon abnormality could define a subgroup among B-cell malignancies, and to document the consequences of the rearrangement.

Clinical and biological patients’ informations were gathered at diagnosis when available. There were 27 men and 16 women, with a median age of 62 years (range, 39–89 years). Thirty-three of 36 patients (92%) had an absolute lymphocytosis, with a median lymphocyte count of 24 × 109/l (range, 5.4–514 × 109/l). Seventeen of 39 patients (44%) had splenomegaly. Of the 25 CLL for whom Binet’ stage was available, there were 16 (64%) A stages, five (20%) B, four (16%) C. B/C stages were treated at diagnosis, 10 (62%) A stages required therapy quickly. The median of the time to treatment was 1, 2 months for all stages, 10 months for A stages (range, 0 month–7 years). Six of 20 were dead 3 years after diagnosis (Supplementary Table 1).

The cytological, histopathological and immunological review were centralized and performed for 30/43 cases by expert hematologists (PF, H M-B) and hematopathologists (FB, F Gaillard) (Table 1 and Supplementary Table 1). There were 20 CLL/small lymphocytic lymphoma (SLL) (WHO classification), including five ‘classical’ CLL, one SLL and 14 ‘atypical’ CLL with heterogeneous mixture of cells (prolymphocytes and/or lymphoplasmacytoid cells). Fifteen of 20 cases displayed features of disease progression with an increased number of larger cells (para-immunoblasts) and mitoses, and 1/20 cases was transformed into a large B-cell lymphoma (total: 16/20, 80%), six of 20 were borderline with the diagnosis of marginal zone lymphoma (MZL). The other diagnoses were MZL (six patients), lymphoplasmacytic lymphoma (LPL) (borderline MZL) (one patient), diffuse large B-cell lymphoma (DLBCL) (two patients) and T-cell rich B-cell lymphoma (one patient). Two cases (one DLBCL, one T-cell rich LBCL) could be transformed MZL. Two of six MZL displayed features of disease progression. The other cases which could not be reviewed were classified as 10 CLL/SLL, including four ‘classical’ CLL and six ‘atypical’ CLL; two DLBCL; one mantle cell lymphoma. In total, there were 30 CLL/SLL out of 43 cases (70%). Of the 36 patients with the CD5 expression available, 31 (86%) were CD5+, including 28/30 (93%) CLL/SLL (six of which being borderline with MZL), 2/5 (40%) MZL and 1 LPL. The Matutes’ score (or modified score by Moreau et al.) could be applied for 23 cases. Regarding CLL patients, 14/18 (78%) had a score 3.

Table 1 Diagnostic, immunophenotypic, cytogenetic and genetic data

Forty-two conventional cytogenetic (CC) analyses were performed, 35 prior treatment, including 27 at diagnosis, and all were reviewed by the GFCH's members (Table 1). Only two cases had an isolated translocation: one CLL and one MZL with variant t(2;19). In three cases (5, 12, 33), the t(14;19) appeared as a secondary change during karyotypic progression. Twenty-two (52%) exhibited complex karyotype (3 abnormalities), fifteen (36%) unbalanced translocation(s) in addition to t(14;19). The most frequent chromosomal abnormalities associated with t(14;19) were: trisomy 12 in 22/42 (52%), 6q deletion in 7/42 (17%), 1q rearrangement in 7/42 (17%), trisomy 3/partial trisomy 3q in 6/42 (14%), monosomy 11/11q deletion in 6/42 (14%), monosomy 17/17p deletion in 5/42 (12%), trisomy 18 in 4/42 (10%), monosomy 7/7q deletion in 4/42 (10%), 13q deletion in 4/42 (10%). When trisomy 12 was present, it was the sole additional abnormality in 13/22 cases (59%). Trisomy 12 was confirmed by FISH in 21 patients and detected in one patient (case 13) who lacked conventional cytogenetic analyses. The following losses were detected with commercial probes by FISH: 6q21 in 8/37 analyzed cases (22%), TP53 in 6/35 (17%), 13q14 in 5/38 (13%), ATM in 3/34 (9%) (Table 2). To confirm the presence of BCL3 involvement, FISH using a BCL3 break-apart probe (RP11-876A24 centromeric, RP11-1147O10 telomeric) was applied. A signal split was observed in all 42 examined cases. Juxtaposition of BCL3 to IGH was proven with the IGH probe in 41/42 cases with t(14;19). BCL3 and IGK probes were juxtaposed in the case with t(2;19).

Table 2 Partial cytogenetic data

IGHV mutation analysis was performed in 24 cases including 20 CLL, two MZL, one LPL, one LDBCL (Table 1). Based on a conventional 98% identity cutoff, in 19 cases (79%) including 18 CLL, IGHV sequences showed 99–100% sequence identity to germline VH-genes and were considered as unmutated. In the remaining five cases (21%) including two MZL, one LDBCL and two CLL, the sequence identity ranged from 87.8 to 97.1% indicating somatic hypermutations had occurred. Four cases (17%) expressed the IGHV4-34 gene, all of them having 100% identity with the germline sequence. Six cases (25%) expressed the IGHV4-39 gene, five of which (21%) with the same IGHV4-39/IGHD6-13/IGHJ5 rearrangements exhibiting a highly similar eight aminoacid long HCDR3 junction. These five cases were all CD5+, nonmutated, and the frequency of this subset among CLL cases was 25% (5/20).

To study the consequences of the t(14;19) or t(2;19) on transcription, we measured the expression of BCL3 and five other genes located close to BCL3 locus on 19q13 (Supplementary Figure 1). Quantitative RT-PCR revealed a threefold to 18-fold increase of BCL3 expression in t(14;19)-CLL cases compared to the median of CLL controls without t(14;19). The normalized expression level of BCL3 in the two groups was significantly different (median: 11.5 vs 2.9, P=0.002) (Figure 1). The variations of the expression of the other genes were not statistically different with respect to the controls (Supplementary Figure 2 and data not shown). Interestingly, the two lymphomas (cases 34, 38) expressed higher levels of PVRL2 and lower levels of BCL3, with respect to the other t(14;19) patients. Furthermore, RELB was concomitantly highly expressed in one of these two latter patients (case 38). These variations could merely reflect different transcriptional control in the region.

Figure 1
figure1

Normalized expression of BCL3 in t(14;19)-positive CLL (11 cases), t(14;19)/t(2;19)-positive lymphomas (two cases) and t(14;19)-negative controls (13 cases). mRNA expression was measured by real-time quantitative reverse transcription, using taqman gene expression assay (Applied Biosystems). Each sample was tested in duplicate, and normalized with respect to ABL expression. Expression analysis was performed in CLL (cases 8, 12, 6, 26, 7, 21, 20, 23, 13, 30, 3), 1 t(2;19)-MZL (case 34), 1 DLBCL (case 38).

Although they share features with common CLL, such as age, sex distribution and CD5 expression, particular characteristics of t(14;19)-CLL argue in favor of a distinct subgroup among CLL. The first of them are morphologic features (cell polymorphism (70% of cases), signs of disease progression (80% of cases)) and immunophenotype with a Matutes score 3 (78% of cases). The current data obtained on a large series extend a previous report of seven t(14;19)-positive cases, which were all CD5+ with Matutes score 3.3 Similarly, cytogenetic and FISH analysis revealed significantly different distributions of abnormalities associated with t(14;19). According to conventional CLL series and our own data (not shown), trisomy 12q and 6q21 deletion were more frequent than in common CLL (67 vs 13.6–16%, P<10−11, and 19 vs 4.6–6%, P<0.02 respectively), while 13q14 deletion was less frequent (14 vs 55–57.4%, P<10−5).4 Trisomy 12 also was frequent in t(14;19)-CLL in other series, but to our knowledge, the high frequency of del(6q21) and the low frequency of del(13q14) in t(14;19)-CLL are reported for the first time.2, 5 The vast majority (90%) of the t(14;19)-CLL of the present series expressed unmutated IGHV genes, which is remarkably higher than the 46% reported in the literature for common CLL (P<0.001).6 Strikingly, two genes (IGHV4-39 and IGHV4-34) accounted for almost half (45%) of the IGH repertoire. Five cases (all CLL) expressed a stereotyped IGHV4-39/IGHD6-13/IGHJ5 receptor, which has been previously described in CLL. However, its frequency in t(14;19)-CLL was much higher than in conventional CLL series (25 vs 0.9–1.2%, P<0.001). Martin-Subero et al.5 also recently reported in an independent series of ours the same rearrangement in three t(14;19)-CLLs. These findings strongly suggest that these IGHV4-39-expressing CLL cases may recognize similar antigenic structures. In common CLLs, this subset has been found to be associated with specific features such as its occurrence in female patients, an IgGκ isotype expression. In our series they differed by a male predominance (4/5) and their IgGλ isotype (3/4 with available information). The IGHV4-34 gene is frequently used in CLL (8.1%), mostly in a mutated form (79%). In contrast four (20%) of our t(14;19)-CLL expressed this gene, all being completely unmutated, whereas these represent only 15% of IGHV4-34 expressing common CLL.6 Finally, as expected, t(14;19) is associated with BCL3 high expression in CLLs. BCL3 high expression is, however, not limited to t(14;19)-CLL nor to B-cell malignancy. Abnormal BCL3 expression was also reported in Hodgkin's disease and peripheral T-cell lymphoma. With respect to t(14;19)-CLL cases, our results confirm previous reports5 and support an individualization of a CLL subgroup.3 Patients with t(14;19)-CLL usually have an aggressive clinical course. (see review Soma et al.2).3 In the current study, the median time from diagnosis to initial therapy was 10 months for A stages, 1.3 months for all CLL stages, 2.3 months regarding t(14;19)-CLL cases with trisomy 12, which was very short compared to the median treatment-free intervals for common CLL groups with 17p deletion (9 months), 11q deletion (13 months), 12q trisomy (33 months) and 13q deletion as the sole abnormality (92 months) reported in the literature.4 Moreover the high percentage of unmutated CLL (90%) and complex karyotype (13/29, 45%), the presence of translocation t(14;19) itself, which are poor prognosis factors, would support the same conclusion.7, 8

t(14;19) is not restricted to CLL and is associated with heterogeneous B-cell malignancies in previous series.2, 5 In our series, it is noteworthy that diagnosis was difficult for seven cases between CLL, MZL and LPL, the most frequent diagnosis after CLL being MZL. Common features with t(14;19)-CLL are, however, frequently observed among MZL cases, such as CD5 expression, trisomy 12, leukemic presentation and morphological characteristics of disease progression. In addition, MZL-associated chromosomal abnormalities such as 7q deletion, trisomy 3 and 18 were rare in our series (Table 2).

In conclusion, the two most frequent t(14;19)-associated disorders are CLL (70%) and MZL (14%), and both have atypical features. t(14;19)-CLL deserves to acquire the status of particular aggressive subgroup defined by cell polymorphism and signs of disease progression, CD5+ immunophenotype and Matutes score 3, trisomy 12 and 6q deletion associated with t(14;19), unmutated IGHV status with an over-representation of the V4-39/D6-13/J5 repertoire, and high BCL3 expression. These observations indicate the deregulation of the NF-kB pathway in mature B-cell malignancies. Considering the already known function of BCL3 and its relation with NF-kB, it could represent a future therapeutic target.

References

  1. 1

    McKeithan TW, Rowley JD, Shows TB, Diaz MO . Cloning of the chromosome translocation breakpoint junction of the t(14;19) in chronic lymphocytic leukemia. Proc Natl Acad Sci USA 1987; 84: 9257–9260.

  2. 2

    Soma LA, Gollin SM, Remstein ED, Ketterling RP, Flynn HC, Rajasenan KK et al. Splenic small B-cell lymphoma with IGH/BCL3 translocation. Hum Pathol 2006; 37: 218–230.

  3. 3

    Huh YO, Abruzzo LV, Rassidakis GZ, Parry-Jones N, Schlette E, Brito-Bapabulle V et al. The t(14;19)(q32;q13)-positive small B-cell leukaemia: a clinicopathologic and cytogenetic study of seven cases. Br J Haematol 2007; 136: 220–228.

  4. 4

    Dohner H, Stilgenbauer S, Benner A, Leupolt E, Krober A, Bullinger L et al. Genomic aberrations and survival in chronic lymphocytic leukemia. N Engl J Med 2000; 343: 1910–1916.

  5. 5

    Martin-Subero JI, Ibbotson R, Klapper W, Michaux L, Callet-Bauchu E, Berger F et al. A comprehensive genetic and histopathologic analysis identifies two subgroups of B-cell malignancies carrying a t(14;19)(q32;q13) or variant BCL3-translocation. Leukemia 2007; 21: 1532–1544.

  6. 6

    Murray F, Darzentas N, Hadzidimitriou A, Tobin G, Boudjogra M, Scielzo C et al. Stereotyped patterns of somatic hypermutation in subsets of patients with chronic lymphocytic leukemia: implications for the role of antigen selection in leukemogenesis. Blood 2008; 111: 1524–1533.

  7. 7

    Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK . Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood 1999; 94: 1848–1854.

  8. 8

    Mayr C, Speicher MR, Kofler DM, Buhmann R, Strehl J, Busch R et al. Chromosomal translocations are associated with poor prognosis in chronic lymphocytic leukemia. Blood 2006; 107: 742–751.

Download references

Acknowledgements

We gratefully acknowledge M Busson, J Ong and L Merlin for excellent technical assistance, F Gaillard for histopathological review, C Lesty for statistical analysis, and the other participants of each institute: K Maloum, S Choquet, S Daliphard, B Chetaille, E Bergoin, M Lessard, F Dupré, J-L Lai, D Mühlematter, P-Y Lovey, F Picard.

Author information

Correspondence to F Nguyen-Khac.

Additional information

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Supplementary information

Supplementary Figure 1 (JPG 19 kb)

Supplementary Figure 2 (JPG 33 kb)

Supplementary Table 1 (DOC 44 kb)

Supplementary Figure and Table Legends (DOC 19 kb)

Rights and permissions

Reprints and Permissions

About this article

Further reading