Immunoglobulin gene translocations in chronic lymphocytic leukemia: A report of 35 patients and review of the literature

Abstract

Chronic lymphocytic leukemia (CLL) represents the most common hematological malignancy in Western countries, with a highly heterogeneous clinical course and prognosis. Translocations involving the immunoglobulin (IG) genes are regularly identified. From 2000 to 2014, we identified an IG gene translocation in 18 of the 396 patients investigated at diagnosis (4.6%) and in 17 of the 275 analyzed during follow-up (6.2%). A total of 4 patients in whom the IG translocation was identified at follow-up did not carry the translocation at diagnosis. The IG heavy locus (IGH) was involved in 27 translocations (77.1%), the IG κ locus (IGK) in 1 (2.9%) and the IG λ locus (IGL) in 7 (20.0%). The chromosome band partners of the IG translocations were 18q21 in 16 cases (45.7%), 11q13 and 19q13 in 4 cases each (11.4% each), 8q24 in 3 cases (8.6%), 7q21 in 2 cases (5.7%), whereas 6 other bands were involved once (2.9% each). At present, 35 partner chromosomal bands have been described, but the partner gene has solely been identified in 10 translocations. CLL associated with IG gene translocations is characterized by atypical cell morphology, including plasmacytoid characteristics, and the propensity of being enriched in prolymphocytes. The IG heavy chain variable region (IGHV) mutational status varies between translocations, those with unmutated IGHV presumably involving cells at an earlier stage of B-cell lineage. All the partner genes thus far identified are involved in the control of cell proliferation and/or apoptosis. The translocated partner gene becomes transcriptionally deregulated as a consequence of its transposition into the IG locus. With the exception of t(14;18)(q32;q21) and its variants, prognosis appears to be poor for the other translocations. Therefore, searching for translocations involving not only IGH, but also IGL and IGK, by banding and molecular cytogenetics is required. Furthermore, it is important to identify the partner gene to ensure the patients receive the optimal treatment.

1. Introduction

Reciprocal chromosomal translocations are recurrent features of several hematological malignancies (http://cgap.nci.nih.gov/Chromosomes/Mitelman; http://atlasgeneticsoncology.org/). The cloning of the genes located at the breakpoints of chromosomal translocations in leukemia and lymphoma has led to the identification of new genes involved in carcinogenesis. These rearrangements generate new genes, called fusion genes, or lead to the activation of a proto-oncogene by relocation near active regulatory sequences. The second mechanism is the hallmark of lymphoma and B-cell chronic lymphocytic leukemia (CLL).

CLL represents the most common hematological malignancy in Western countries, with a highly heterogeneous clinical course and prognosis, and a time-to-progression varying from months to several years. Some patients live for prolonged periods without therapy, while others rapidly develop progressive disease and require treatment (1,2).

Morphologically, the World Health Organization recognizes 3 subtypes (3): The ‘typical’ CLL is characterized by small lymphocytes with mature chromatin and minimal cytoplasm without nucleoli, and <10% of prolymphocytes (PLs). The prolymphocytic leukemia (PLL) has >55% PLs, which are large cells with a round nucleus, a prominent vesicular nucleolus, condensed nuclear chromatin and abundant cytoplasm (4). Occasionally, a subset of CLL cases acquires an increased number (10–55%) of PLs (CLL/PL) and eventually transforms to a neoplasm composed predominantly of PLs (5). The immunophenotype of CLL is CD5⁺, CD23⁺, FMC7⁻, CD20 dim⁺ and clonal surface immunoglobulin (sIg) dim⁺ (6).

Transformation into a fast-growing diffuse large B-cell lymphoma is encountered in ~5% of CLL patients and it is referred to as Richter's syndrome (7,8). Under rare circumstances, plasmablastic lymphoma or plasmablastic transformation may be observed, representing an unusual example of Richter's syndrome.

Numerous studies have searched for reliable prognostic markers capable of predicting the progression and outcome of this disease. They include two different clinical staging systems, one described by Rai et al (9) and the other by Binet et al (10). Other prognostic markers, such as CD38 and ZAP-70 expression and IG heavy chain variable region (IGHV) mutational status, are being evaluated or used (2,11). More recently, next-generation sequencing has identified mutations in a few genes that may have a prognostic impact (12,13).

At present, interphase fluorescent in situ hybridization (iFISH) targeting four chromosomal abnormalities [deletion of 13q14, trisomy 12, deletion of 11q22 (ATM) and deletion of 17p13 (TP53)] remains the most effective outcome predictor (14). However, iFISH only provides a partial view of the karyotypic complexity of CLL; specific cytogenetic abnormalities, including translocations, which remain undetected by iFISH, are currently identified, due to the use of more effective B-cell mitogens. Among these, translocations involving the immunoglobulin (IG) genes have emerged as one of the cytogenetic abnormalities that have clinical and biological specificities and prognostic implications.

2. Brest University Hospital cohort

Patients

Between 2000 and 2014, the Cytogenetics Laboratory of the Brest University Hospital collected blood or bone marrow samples from CLL patients diagnosed and/or followed up in 10 hospitals in Brittany. The clinical diagnosis of CLL in each patient was based on a persistent lymphocytosis of >5×109 cells/l and a typical immunophenotypic picture (CD5⁺, CD19⁺, CD20⁺, CD23⁺ and weak expression of sIg) (15–17). A total of 396 patients were referred at diagnosis and 275 during follow-up.

Methods

Peripheral blood and bone marrow were cultured for 72 h. Cell stimulation was performed with tetradecanoylphorbol acetate from 2000 to 2010, and with the immunostimulatory CpG-oligonucleotide DSP30 and interleukin-2 from 2010 onwards. The chromosomes were R-banded and the karyotype was described according to the International System for Human Cytogenetics Nomenclature (18).

iFISH was performed on fixed cells from the cultures using the Vysis CLL FISH Probe kit (Abbott Molecular, Rungis, France). The CLL panel includes two set of probes: A first set consisting of LSI TP53 SpectrumOrange and ATM SpectrumGreen probes, and a second set consisting of LSI D13S319 SpectrumOrange/13q34 SpectrumAqua and CEP 12 SpectrumGreen probes. A total of 300 interphase nuclei were studied for each set of probes. Based on the recommendations of the CLL Research Consortium FISH standardization project, FISH cut-off was set at 10% for each hybridization (19).

Metaphase and iFISH using the IGH Breakapart Probe (Cytocell, Compiègne, France or Abbott Molecular) was not systematically performed on all samples, but only on those in which an abnormality of chromosome 14 was suspected. Translocations involving the IG κ locus (IGK) or IG λ locus (IGL) genes were assessed with the IGK Breakapart Probe or the IGL Breakapart Probe (Cytocell).

Results

Overall, an IG translocation was identified in 35 patients (Table I). The patients included 22 men and 13 women. The mean age was 72 years (range, 40.8–85 years; standard deviation, 8.5 years) and the median age was 72.3 years. The patient sample included 18 patients among the 396 investigated at diagnosis (4.6%) and 17 of the 275 analyzed during follow-up (6.2%). An IG translocation was present at diagnosis and follow-up in 3 patients. A total of 4 patients for whom the IG translocation was identified at follow-up did not carry the translocation at diagnosis, while no metaphases were observed at diagnosis in another patient.

Table I.

Demographic data and karyotype of 35 patients with an IG translocation (Brest cohort).

Patient	Gender	Age (years)	Evolution	Karyotype
1	M	79.1	E	46, XY, del(11)(q21q23)[4]/46,idem, add(17)(p11)[6]/46,idem,t(3;14)(q21;q32)[6]/46,idem, add(13)(q14),add(17)(p11)[2]/46,idem, del(17)(p11)[2]/46,XY[1]
2	M	62.7	D	47,XY,+12,t(14;18)(q32;q21)[22]
3	M	65.4	E	46,XY,t(7;22)(q21;q11),add(13)(p11)[11]/46,XY[8]
4	F	74.4	D	45,XX, add(3)(q27),der(8)t(8;12)(p11;q11),t(11;14)(q13;q32),-12,del(17) (p1?2),-17,+mar1,avec variations[17]/46,XX[4]
5	M	62.9	E	47, XY,+12[16]/47,sl, del(13)(q13qq21)[4]/47,sl2,t(9;14)(q13;q32)[2]/47,sl3,add(11)(q21)[2]
6	M	78.8	D	47,XY,t(10;17)(q21;q22),t(14;19)(q32;q13),del(14)(q23q32),+21[5]/46,XY[18]
6	M	79.0	E	47,XY,t(10;17)(q21;q22),t(14;19)(q32;q13),del(14)(q23q32),+21[7]/46,XY[18]
7	M	72.0	D	48,XY,+4,t(11;14)(q13;q32),+12,[15]/46,XY[6]
8	F	71.8	E	48,XXX,+12,t(18;22)(q21;q11)[6]/46,XX[15]
8	F	81.7	E	48,XX,+X,+12,t(18;22)(q21;q11)[16]/46,XX[6]
9	M	78.3	E	48,XY,+8,dup(12q)×2,add(13)(q34),t(14;18)(q32;q21),+20[16]/50,sl,+X,+7[2]/50,sdl1, (p13)[4]/46,XY[1]
10	M	76.0	D	46,XY,t(14;18)(q32;q21)[5]/47,idem,+12[1]/46,XY[16]
11	M	73.7	D	46,XY,t(14;18)(q32;q21)[6]/47,idem,+12[3]/46,XY[13]
12	M	78.1	E	46,XY,t(14;18)(q32;q21)[23]
13	M	69.7	D	47,XY,+12,t(14;18)(q32;q21)[4]/46,XY[18]
14	F	79.6	D	47,XX,t(1;14)(p11;q32),+12[7]/46,XX[14]
15	F	71.3	E	47,XX,+12,t(18;22)(q21;q11)
15	F	71.8	E	47,XX,+12,t(18;22)(q21;q11)[6]/47,idem,t(1;16)(p10?q10)[2]/46,XX[14]
16	M	81.4	E	48,XY,+3,del(10)(q2?4),del(13q),ins(13;?)(q14;?),+18[4]/48,idem,t(4;14) (q2?1;q32[1]/46,XY[18]
17	M	81.2	E	46, XY, del(13q), add(18p)[1]/46,XY,t(11;14)(q13;q32)[1]/46,XY[25]
18	F	68.5	D	47,XX,+12,t(14;18)(q32;q21)[5]/46,XX[2]
18	F	70.7	E	47,XX,+12,t(14;18)(q32;q21)[17]/46,XX[5]
19	M	72.3	E	46,X,-Y,t(11;14)(q13;q32),del(11)(q22),+mar1[7]/46,XY[17]
20	M	76.6	D	47,XY,+12[16]/47,idem,t(14;19)(q32;q13)[3]/46,XY[1]
21	F	84.1	D	46,XX,t(14;22)(q32;q11)[2]/46,XX[33]
22	M	81.4	D	47,XY,+12,t(14;18)(q32;q21)[12]/47,idem, del(6)(q12q16)[8]/46,XY[1]
23	F	75.7	E	46,XX, add(3)(q22),der(8)t(8;14)(q24;q32),del(10)(q2?5),del(13)(q13q31),add(14)(q32)
24	M	85.0	E	46, XY, del(1)(p34)[13]/45,sl,t(8;14)(q24;q32)[3]/46,sdl1,del(11)(q21)[3]/46,XY[2]
25	F	74.2	D	47,XX,+12,t(14;19)(q32;q13)[18]/47,idem, del(12)(q23)[2]/46,XX[2]
26	M	54.0	D	47,XY,+12[5]/47,idem,t(14;18)(q32;q21)[8]/47,idem, add(17)(q2?5)[8]/46,XY[1]
27	F	65.5	D	46,XX,t(14;18)(q32;q21)[12]/46,XX[7]
27	F	75.4	E	46,XX,t(14;18)(q32;q21)[5]/46,XX[3]
28	M	67.9	E	45,X,-Y[6]/47,XY,+12[2]/47,idem,t(6;14)(p21;q32)[7]/46,XY[6]
29	F	66.5	D	47,XX,+12[2]/47,XX,+12,t(18;22)(q21;q11)[5]/50,XX,+3,+12,+17,+18[2]/46,XX[10]
30	M	61.9	E	46, XY, t(18;22)(q21;q11)[6]/46, XY, del(13)(q13;q21),t(18;22)(q21;q11)[12]/46,XY[2]
30	M	66.5	E	46,XY,t(18;22)(q21;q11)[7]/46,idem, del(13)(q13q21)[13]/46,XY,t(7;13) (p2;q14),t(8;13) (p23;q14),t(18;22)(q21;q11)[2]
31	F	40.8	D	46,XX,t(14;19)(q32;q13)[17]/46,XX[1]
32	M	70.1	E	46,XY,t(8;22)(q24;q11)[22]
33	M	59.9	E	46,XY,t(18;22)(q21;q11)[19]/46,XY[1]
34	F	78.9	E	46,XX,del(5)(q1?4q2?3)[15]/45,idem,-4,del(4)(p1?3),der(11) (11pter->11p12::?::11cen->11q22::?)[4]/46,idem,t(14;18)(q32;q21)[2]
35	F	69.0	D	47,XX,+12[12]/46,XX,t(2;7)(p11;q21)[2]/46,XX[10]

IG, immunoglobulin; M, male; F, female; E, during follow-up; D, at diagnosis.

IG heavy locus (IGH; chromosome band 14q32) was involved in 27 (77.1%), IGK (band 2p11) in 1 (2.9%) and IGL (band 22q11) in 7 (20.0%) translocations. The chromosome band partners of the IG translocations were 18q21 in 16 cases (45.7%), 11q13 and 19q13 in 4 cases each (11.4% each), 8q24 in 3 cases (8.6%) and 7q21 in 2 cases (5.7%), whereas 6 other bands were involved once (2.9% each). However, as rearrangements involving IGH were not systematically searched for by FISH, we cannot exclude that some cryptic translocations may have escaped detection.

iFISH with the CLL FISH Probe kit was performed on 25 patients carrying an IG translocation (Table II). Mono-allelic deletion of 13q14 was identified in 9 patients (36%), including a bi-allelic in 2 patients. Trisomy 12 was found in 12 patients (48%). Deletions of 11q22 and 17p13 were observed in 1 (4%) and 3 (12%) patients, respectively (Table II). Recently, we reported our iFISH results on 638 patients. Del(13q14) was found in 65% and a trisomy 12 in 22.1% of the patients. Deletions of 11q22 and 17p13 were observed in 13.3 and 8.6% of the patients, respectively (20).

Table II.

iFISH results of 35 patients with an IG translocation (Brest cohort).

	iFISH
Patients	13q14	bi13q14	+12	ATM	p53
1	+/−		N	+/−	+/−
2
3	+/+	+/+	N	+/+	+/−
4	+/+	+/+	+	+/+	+/−
5
6
6
7
8
8	+/+	+/+	+	+/+	+/+
9
10	+/+	+/+	+	+/+	+/+
11	+/+	+/+	+	+/+	+/+
12	+/−	+/+	N	+/+	+/+
13	+/+	+/+	+	+/+	+/+
14	+/+	+/+	+	+/+	+/+
15	+/+	+/+	+	+/+	+/+
15	+/+	+/+	+	+/+	+/+
16
17	+/−	+/+	N	+/+	+/+
18	+/+	+/+	+	+/+	+/+
18	+/+	+/+	+	+/+	+/+
19
20	+/+	+/+	+	+/+	+/+
21	+/+	+/+	N	+/+	+/+
22
23	+/−	+/+	N	+/+	+/+
24	+/+	+/+	N	+/+	+/+
25	+/+	+/+	+	+/+	+/+
26	+/+	+/+	+	+/+	+/+
27
27	+/−	+/+	N	+/+	+/+
28	+/−	−/−	+	+/+	+/+
29
30	+/−	−/−	N	+/+	+/+
30	+/−	−/−	N	+/+	+/+
31	+/+	+/+	N	+/+	+/+
32	+/+	+/+	N	+/+	+/+
33	+/−	+/+	N	+/+	+/+
34	+/−	+/+	N	+/+	+/+
35

iFISH, interphase fluorescent in situ hybridization; IG, immunoglobulin; +/+, normal; +/−, monoallelic deletion; -/-, biallelic deletion; N, normal; 13q14, del(13q14); ATM, del(11q22); p53, del(17p13). Certain patients were studied at diagnosis and during follow-up; thus, they may appear twice.

3. Frequency and distribution of IG translocations

For years, banding karyotyping has been hampered by the low mitotic index of CLL cells, and the majority of the studies currently rely on iFISH with an IGH probe to determine the frequency of IGH translocations. A total of 18 studies were retrieved from the literature (Table III). The overall frequency was 8.3% (327/3,922) but the range varied considerably, from 1.9% in a study from the USA, to 26.1% in a Canadian study. Although a geographic/ethnic uneven distribution cannot be ruled out, it is likely that differences in diagnosis (PLL included or not), referral, access to care and healthcare systems may have affected the frequency. Even for those investigated at diagnosis, we cannot exclude that socioeconomic status or access to healthcare may have delayed the time at which diagnosis was made, with the consequence that patients in some series were at a more advanced stage (20,21).

Table III.

Frequency of translocations involving IGH detected by iFISH.

First Author/(Refs.)	Country	No. of patients	No. of patients with IGHt	Frequency of IGHt (%)
Gerrie (133)	Canada	290	27	9.3
Gerrie (132)	Canada	142	37	26.1
Lu (134)	USA	149	23	15.4
Shanafelt (135)	USA	159	3	1.9
Aoun (136)	USA	58	10	17.2
Nelson (137)	USA	109	10	9.2
Flanagan (138)	USA	42	8	19.0
Jenderny (139)	Germany	129	4	3.1
Haferlach (140)	Germany	500	30	6.0
Jeromin (13)	Germany	1,158	57a	4.9
Alhourani (141)	Germany	85	3	3.5
Döhner (142)	Germany	325	12	3.7
Berkova (143)	Czech Republic	146	11	7.5
Amare (144)	India	116	7	6.0
Yoon (145)	South Korea	48	6	12.5
Xu (146)	China	83	8	9.6
Qiu (147)	China	143	28	19.6
Wang (148)	China	240	43	17.9

^aIncluding IGH deletions. IGH, immunoglobulin H; iFISH, interphase fluorescent in situ hybridization; IGHt, translocations involving IGH.

A limited number of studies attempted to identify the partner genes. Even in those cases, they used commercially available Dual Color, Dual Fusion Translocation Probes to detect IGH/B-cell lymphoma 2 (BCL2), IGH/cyclin D1 (CCND1), sometimes completed by v-myc avian myelocytomatosis viral oncogene homolog (MYC) and BCL3 Dual Color Breakapart rearrangement probes. Furthermore, none of these studies used probes targeting the IGK and IGL genes and, therefore, underestimated the true frequency of IG translocations.

We conducted a thorough search in the literature looking for IG translocations, using the Mitelman Database of Chromosome Aberrations and Gene Fusions in Cancer (http://cgap.nci.nih.gov/Chromosomes/Mitelman) as the starting point.

Some translocations have been found to be frequent, while others have only been reported in a few or single cases. At present, 31 partner chromosome bands have been described, but the partner gene has only been identified in 10 translocations. We identified 4 chromosomal bands that had never been shown to be involved in IG translocations among the 35 patients in the Brest cohort, bringing the total to 35 (Fig. 1). The number of bands involved in IG translocations is significantly lower compared with that involved in Ets variant 6 or Runt-related transcription factor 1 gene translocations (48 and 55 bands, respectively) (22,23). The difference is likely the result of the B-cell lineage specificity of the partner genes deregulated by the IG translocations.

Figure 1.

Distribution of the chromosomal band partners of immunoglobulin gene translocations. Squares, translocations with known partner gene; circles, translocations with unknown partner gene; triangles, translocations with unknown partner gene identified in Brest.

4. Recurrent IG translocations with known partner genes

As previously mentioned, IG translocations relocate genes near active regulatory sequences, leading to their overexpression. However, two exceptions are known. In 2004, Schmidt et al (24) reported a patient with B-cell CLL carrying a t(12;14)(q23;q32), in whom the carbohydrate (chondroitin 4) sulfotransferase 11 gene (CHST11) was fused to IGH to create a chimeric gene. Fusion RNAs may be translated into truncated proteins and lead to deregulation of the CHST11 protein trafficking across intracellular membranes, resulting in loss of function, as observed with other fusions (23).

A t(14;22)(q32;q11) involving the IGH and IGL genes was retrieved from the literature (25) and another was identified in our series (patient 21). Although the mechanism of neoplastic transformation remains unknown, one cannot exclude that a nearby gene may be involved. Indeed, the BCL11B (zinc finger protein) gene, closely related to BCL11A involved in t(2;14)(p16;q32), may be deregulated, although it is located 6.3 Mb from the IGH locus. It was suggested that BCL11B acted as a transcriptional repressor as well as an activator in a context-dependent manner and may function on the P53 signaling pathway (26,27).

t(14;18)(q32;q21) and variants (IGH/BCL2)

All 3 translocations, t(14;18) (q32;q21), t(2;18)(p11;q21.3) and t(18;22)(q21.3;q11), and their molecular consequences, IGH/BCL2, IGK/BCL2 and IGL/BCL2, have been reported (28,29). These translocations are present at diagnosis or arise during evolution and are usually associated with additional karyotypic changes, more particularly with trisomy 12 (30–32).

A total of 144 cases reported in the literature were reviewed. A t(14;18) was identified in 111 cases (77.1%), a t(2;18) in 8 cases (5.5%) and a t(18;22) in 25 cases (17.4%). Trisomy 12 was found in 71 cases (49.3%). The IG translocation was the sole abnormality in 39 cases (27.1%) and part of a complex karyotype (defined as composed of ≥3 abnormalities) in 37 cases (25.7%).

These are found in typical CLL (sometimes during Richter's transformation) and in CLL/PL (29,30,32). Atypical cytological characteristics (increased number of lymphoid cells with irregular nuclear contour, plasmacytoid features or PLs) and/or immunophenotypic profile (lack of CD23 and intermediate/strong CD20 expression) is reported and hypothesized to be linked to trisomy 12 (29,33,34). IG/BCL2 fusion is significantly associated with mutated IGHV status (30–32), which led Baseggio et al (30) to conclude at a post-germinal center cellular origin.

There is persisting controversy regarding the prognostic impact of IG/BCL2, being considered as poor by certain authors (31), while others found that the clinical course was not affected (30,32).

The t(14;18)(q32;q21) and its variants lead to the overexpression of the BCL2 gene (35,36). This gene encodes an integral outer mitochondrial membrane protein that blocks the apoptotic death of some cells, such as lymphocytes (37,38).

t(14;19)(q32;q13) (IGH/BCL3)

This translocation appears at diagnosis in the primary clone in the majority of the cases, and as a secondary change during karyotypic progression in a limited number of cases (39). This is rarely the sole cytogenetic aberration, with trisomy 12 being the most frequent associated abnormality (40–45).

A total of 123 cases reported in the literature were reviewed. No variant translocation was found. Trisomy 12 was identified in 72 cases (58.5%). The translocation was the sole abnormality in 13 cases (10.6%) and part of a complex karyotype in 52 cases (42.3%).

A high proportion of patients with CLL and t(14;19) are aged <40 years (40–44) and the median age is significantly lower compared with that observed in non-carriers (20).

CLL with t(14;19) is associated with atypical morphological (small cells, often with nuclear indentations) and immunophenotypical characteristics (40,41,46). The translocation is found in all three subtypes and during Richter's transformation (42,43,45,47). The vast majority of t(14;19) express unmutated IGHV genes (39–41,45,46), which is significantly higher compared with the 46% reported in the literature for typical CLL (48). Patients have an aggressive clinical course and the overall prognosis is poor (42,43,45,46).

The t(14;19)(q32;q13) and its variants lead to the overexpression of the BCL3 gene (49). This gene encodes a protein that is a member of the IκB family and is present predominantly in the nucleus. It contributes to the regulation of transcriptional activation of nuclear factor-κB target genes and to the regulation of cell proliferation. It binds to the promoter of CCND1, thus stimulating gene transcription, and inhibits P53-induced apoptosis (50–55).

t(11;14)(q13;q32) t(11;22)(q13;q11) (IGH/CCND1)

Although the (11;14)(q13;q32) translocation was previously considered to be the hallmark of mantle cell lymphoma (MCL), it is currently identified in 10–20% of PLL and 2–5% of CLL cases (56). CLL cases in which this translocation has been found are usually atypical in terms of morphology (majority of small lymphocytes with PLs and/or large lymphocytes) and immunophenotype (CD5⁺, CD19⁺, sIg⁺, FMC7⁺ and CD10⁻) (57–60). Recognition of this cytogenetic subset of atypical CLL is crucial, as, given its poor prognosis, it may require early treatment (61).

We identified 106 CLL or PLL cases associated with t(11;14) in the literature. Only one variant translocation, t(11;22)(q13;q11), was found (62). Trisomy 12 was identified in 5 cases (4.7%). The translocation was the sole abnormality in 21 cases (19.6%) and part of a complex karyotype in 64 cases (59.8%).

Although t(11;14) is usually present at diagnosis, Nishida et al (63) reported a patient with CLL associated with trisomy 12 and t(14;19)(q32;q13) at diagnosis, who developed a t(11;14)(q13;q32) at relapse.

Distinction of t(11;14) translocation-associated CLL and MCL in the leukemic phase is not unequivocal (60,64). It is hypothesized that MCL and atypical CLL with the t(11;14) represent the extremes of a spectrum of disorders of follicle mantle lineage (60).

The t(11;14)(q13;q32) leads to the overexpression of the CCND1 gene (65). The protein encoded by this gene belongs to the highly conserved cyclin family. It interacts with the cyclin-D-dependent kinases (CDK) 4 and 6 that phosphorylate the retinoblastoma 1 (RB1) tumor suppressor, thus regulating a process that promotes the G1/S cell cycle transition (66,67).

t(8;14)(q24;q32) and variants (IGH/MYC)

This abnormality is rare in typical CLL and is associated with increased PLs (CLL with occasional PLs, CLL/PL and PLL) (68–71). Male and elderly patients are predominantly affected, as in CLL/PL patients (69–71).

This translocation may be acquired in the chronic phase, but is associated with an advanced clinical stage at presentation (68,71). It is usually included in a complex karyotype. It may represent a secondary abnormality contributing to disease progression and carries a poor prognosis (68–70).

We reviewed 38 cases from the literature. A t(8;14) was identified in 22 cases (57.9%), a t(2;8) in 5 cases (13.2%) and a t(8;22) in 11 cases (28.9%). Trisomy 12 was rarely associated, being identified in only 3 cases (7.9%). The IG translocation was the sole abnormality in 8 cases (21.1%) and part of a complex karyotype in 18 cases (47.4%).

Put et al (70) reported a prevalence of del(11)(q22) and del(17)(p13) in a cohort of 25 patients with a t(8;14) or variants, which was higher compared with that reported in CLL [6/25 (24%); and 7/25 (28%), respectively] (70).

The t(8;14)(q24;q32) and its variants lead to the overexpression of the MYC gene (72). MYC is a transcription factor that controls functions associated with cell cycle progression, growth, differentiation, apoptosis, survival and metabolism. It functions as a transcription factor that regulates transcription of >15% of all genes. MYC positively affects cell cycle regulation, apoptosis and metabolism, but negatively affects cellular differentiation and cell adhesion. Therefore, aberrant MYC expression deregulates the balance between survival and apoptosis signals at several different stages (73–76).

t(2;14)(p16;q32) IGH/BCL11A

This rare but recurrent translocation affects young adults, even children (77,78). Morphologically, the disease is characterized by a mixture of small and larger lymphocytes with indented and irregular nuclear contours, also including plasmacytoid lymphocytes and PLs. The immunophenotype is almost always typical of CLL (78).

The (2;14) translocation appears to be an early event, as it has been found to be the sole karyotypic abnormality at diagnosis. It is also present in the main clone, with subclones containing additional abnormalities in other patients (77,78). Only 6 cases were retrieved from the literature, 2 of which were associated with trisomy 12, and 3 of which were present in subclones with a complex karyotype.

All patients thus far analyzed express ZAP-70 and all but one also carried unmutated IGVH genes (78,79). Therefore, it is expected that t(2;14) is associated with an aggressive disease and a poor prognosis, which appears to be the case, although data is sparse (77,78).

The t(2;14)(p16;q32) leads to the overexpression of the BCL11A gene (77). BCL11A encodes a zinc finger protein that interacts directly with BCL6; it is essential for the early lineage commitment steps in lymphopoiesis and functions in the development of T as well as B cells (77,80). The BCL11A protein is expressed in the germinal centers and mantle zones of the lymph nodes (80,81).

t(10;14)(p12;q32) t(10;22)(p12;q11) [IGH/BMI1 proto-oncogene, polycomb ring finger (BMI1)]

The t(10;14)(p11-p13;q32) and t(10;22)(p12;q11) were identified in 5 and 1 CLL cases, respectively (82). The translocation was identified during clinical progression or Richter's transformation and appears to carry a poor prognosis (82).

These translocations are not associated with a particular subtype of CLL, or with IGVH mutation status; they do not appear to be driver abnormalities in CLL genesis, but are rather a marker of disease progression (82).

The t(10;14)(p12;q32) and variants lead to the overexpression of the BMI1 proto-oncogene (82). This gene encodes a protein containing a conserved RING finger and a helix-turn-helix motif (83). The BMI1 protein is a core subunit of the PRC1 complex, which plays important roles in the regulation of HOX gene expression (84,85). BMI1 regulates the proliferation activity of normal cells, stem cells and progenitor cells, and plays a role in cell cycle checkpoints and progression, DNA damage repair, cell fate and differentiation, apoptosis and senescence (86,87). BMI acts as transcriptional repressor of the cyclin-dependent kinase inhibitor 2A gene (CDKN2A), also referred to as INK4A or ARF gene (88,89).

t(4;14) (p16;q32) t(4;22)(p16;q11) [IGH/fibroblast growth factor receptor 3 (FGFR3)-multiple myeloma SET domain (MMSET)]

Although the t(4;14) has long been known to be a recurrent abnormality in multiple myeloma, it was first described by Bacher et al (90) in 2 patients exhibiting the immunophenotype of CLL and CLL/PL. Cerny et al (91) reported a CLL patient with the t(4;22)(p16;q11) and a typical immunophenotype, in whom the lymphoid proliferation was composed of small lymphocytes (with round nuclei, condensed chromatin, indistinct nucleoli and scant cytoplasm) and PLs (larger nucleolated lymphocytes). A fourth case was reported by Geller et al (92) in a patient with CLL/PL and a typical immunophenotype. A total of 2 cases had a trisomy 12.

These 4 patients exhibited no consistent characteristics, other than the presence of PLs. The translocation was identified at diagnosis or during evolution. The cells were negative or positive for CD38 and ZAP-70, and had a mutated or non-mutated IGVH status. Although the number of cases reported is small, this translocation should be considered as indicative of adverse prognosis.

The t(4;14)(p16;q32) is a unique example of IG translocation, as it simultaneously deregulates 2 genes with oncogenic potential. Indeed, MMSET domain (also referred to as Wolf-Hirschhorn syndrome candidate 1) and FGFR3 reside on either side of the 4p16 breakpoint. After the translocation, MMSET remains on the derivative chromosome 4, while FGFR3 moves to the derivative chromosome 14 (93–95). The MMSET protein has histone methyltransferase activity and may act as a transcriptional regulator controlling cell cycle and apoptosis (96,97). The FGFR3 protein is a tyrosine protein kinase that acts as a cell surface receptor for fibroblast growth factors and triggers downstream mitogen-activated protein kinase and phosphatidylinositol 3-kinase signaling. It plays an essential role in the regulation of cell growth and differentiation (98,99).

t(7;14)(q21;q32) and variants (IGH/CDK6)

Although these translocations, particularly t(2;7)(p11;q21), have been reported in several cases of splenic marginal zone lymphoma (SMZL), they were unequivocally identified in 12 CLL patients (100–104). Unfortunately, no data on cell morphology, immunophenotype, or prognosis, are available.

Of note, in SMZL and CLL, t(2;7) is more frequent compared with the two other translocations, t(7;14) and t(7;22). No explanation has been provided as to why IGK is more likely than IGH to juxtapose to CDK6, contrary to the other translocations in which IGH is much more frequently rearranged (104). Of the 11 cases retrieved from the literature, 7 had a (2;7)(p11;q21) translocation (63.6%), 3 had a t(7;14) (27.3%) and 1 had a t(7;22)(q21;q11) (9.1%). Our series included 1 case with t(2;7) and 1 case with t(7;22). The translocation was the sole abnormality or included in a complex karyotype in 4 cases each.

The t(7;14)(q21;q32) and its variants lead to overexpression of the CDK6 gene (103,105,106). CDK6 and CDK4 are serine/threonine kinases, members of the cyclin-dependent protein kinase family, involved in the control of the cell cycle and differentiation. CDK6 is important for G1 phase cell cycle progression and G1/S transition by phosphorylating and inactivating the RB1 protein (107–110).

t(9;14)(p13;q32) [IGH/paired box gene 5 (PAX5)]

Although the t(9;14)(p13;q32) and its variants have been mostly identified in diffuse large B-cell lymphoma (109), the t(9;14) has also been found in 7 CLL cases (111–114). Few data are available on the cell morphology, but plasmacytoid differentiation is a common characteristic.

Although no trisomy 12 was found, a duplication of part of its long arm was identified in 2 cases. The t(9;14) was included in a complex karyotype in 4 patients, but present as the sole abnormality in the remaining 3 patients.

This translocation leads to the overexpression of the PAX5 gene (115,116). The nuclear transcription factor encoded by this gene is a B-cell lineage-specific activator protein that is expressed during the early, but not the late stages of B-cell differentiation. It restricts the developmental potential of progenitor cells to the B lymphoid pathway by suppressing alternative cell fates and is also involved in the regulation of the CD19 gene, a B lymphoid-specific target gene (117–119).

t(6;14)(p21;q32) (IGH/CCND3)

Although the t(6;14)(p21;q32) has been mostly identified in multiple myeloma (118), it has also been found in diffuse large B-cell lymphoma, splenic lymphoma with villous lymphocytes, SMZL and MCL (120,121). The t(6;14) was identified as the sole chromosomal abnormality in a male patient with CLL associated with atypical cell morphology and involvement of CCND3 (122).

This translocation leads to the overexpression of the CCND3 gene (120,123). This gene, which is closely related in sequence to the CCND1 gene, codes a protein belonging to the highly conserved cyclin family. Similar to CCND1, this protein regulates a process that promotes the G1/S cell cycle transition (124–126).

5. IG translocation-associated CLL is heterogeneous

The main characteristics of the translocations for which the partner gene was identified are summarized in Table IV. IG translocations result in the deregulated expression of genes involved in several pathways. All the partner genes thus far identified are involved in the control of cell proliferation and/or apoptosis (127). In the majority of the cases, the translocated partner gene becomes transcriptionally deregulated as a consequence of its transposition into the IG locus (128).

Table IV.

Summary of the main characteristics of the IG translocations with known partner genes.

IG translocation	Morphology	IGHV status	Trisomy 12	Prognosis	Partner gene	Functions
t(14;18)(q32;q21)	CLL	M	Yes	Controversial	BCL2	Inhibition of apoptosis
	CLL/PL
	Plasmacytoid features
t(14;19)(q32;q13)	CLL	U	Yes	Poor	BCL3	Cell growth
	CLL/PL					Inhibition of apoptosis
	PLL
	Atypical morphology
t(11;14)(q13;q32)	Atypical CLL	ND	No	Poor	CCND1	Cell growth
	PLL					G1/S cell cycle transition
t(8;14)(q24;q32)	CLL with occasional PL	ND	No	Poor	MYC	Cell cycle progression
	CLL/PL					Cell growth
	PLL					Cell differentiation
						Apoptosis
t(2;14)(p16;q32)	Atypical morphology	U	IN	Poor	BCL11A	Early B-lineage commitment
	Plasmacytoid features
	CLL/PL
t(10;14)(p12;q32)	IN	U/M	IN	Poor	BMI1	Regulation of HOX gene expression
						Cell cycle progression
						Cell differentiation
						Apoptosis
t(4;14)(p16;q32)	CLL with occasional PL	U/M	IN	Poor?	MMSET	Cell cycle control
	CLL/PL				FGFR3	Apoptosis
						Cell growth
						Cell differentiation
t(7;14)(q21;q32)	ND	ND	IN	ND	CDK6	G1 cell cycle progression
						G1/S cell cycle transition
						Cell differentiation
t(9;14)(p13;q32)	Plasmacytoid features	ND	IN	IN	PAX5	Early B-cell differentiation
t(6;14)(p21;q32)	Atypical morphology	IN	IN	IN	CCND3	Cell growth
						G1/S cell cycle transition

IG, immunoglobulin; IGHV, IG heavy chain variable region; CLL, chronic lymphocytic leukemia; PL, prolymphocytes; PLL, prolymphocytic leukemia; M, mutated; U, unmutated; ND, no data; IN, insufficient number of cases; BCL, B-cell lymphoma; CCND, cyclin D; MYC, v-myc avian myelocytomatosis viral oncogene homolog; BMI1, BMI1 proto-oncogene, polycomb ring finger; MMSET, multiple myeloma SET domain; FGFR, fibroblast growth factor receptor; CDK, cyclin-dependent kinase; PAX5, paired box gene 5.

Although the number of cases with a given translocation is sometimes low, it appears that CLL associated with IG translocations is characterized by atypical cell morphology, including plasmacytoid characteristics, and the propensity of being enriched in PLs. The IGHV mutational status varies between translocations, those with unmutated IGHV presumably involving cells at an earlier stage of the B-cell lineage.

With the exception of t(14;18), prognosis appears to be poor for the translocations for which sufficient data is available (123,129,130). Davids et al (131) demonstrated that the time to first treatment was significantly shorter among CLL patients harboring 14q32 translocations without t(14;18), compared with those with t(14;18). Furthermore, the presence of an IGH translocation associated with a del(13q) was shown to confer a poorer prognosis compared with del(13q) alone (132).

Although most centers currently use FISH to identify trisomy 12 and deletions of 13q14, 11q22 and 17p13, which are known to have prognostic significance, it is evident that searching for translocations involving not only IGH, but also IGL and IGK, by banding and molecular cytogenetics, will add new information. Furthermore, as the prognosis depends on the partner gene involved in the translocation, it is important to identify this partner gene, at least in recurrent translocations, to ensure the patients receive optimal treatment.