Splenic marginal zone lymphomas and lymphoplasmacytic lymphomas originate from B-cell compartments with two different antigen-exposure histories

Since the beginning of the nineties, lymphoplasmacytic lymphoma (LPL) entity has been repeatedly challenged against marginal zone lymphomas, mainly splenic marginal zone lymphoma (SMZL). This is mainly due to difficulties in differential diagnosis related to overlapping morphological features. Various authors raise the question of the genuine existence of LPL, considering that plasma cell differentiation is a morphological aspect of marginal zone lymphomas. Moreover, some authors consider that secretion of an IgM monoclonal component at high levels, so-called Waldenström macroglobulinaemia (WM), is just a clinical/biological syndrome.1 Sequence analysis of immunoglobulin variable regions could be of interest in understanding the relationships between these two entities. For example, 1/3 of SMZL cases exhibit unmutated immunoglobulin variable heavy-chain (IgVH) genes, with a heterogeneous usage of VH segments.2 In contrast, IgVH somatic mutations are frequent with a predominance of VH3 usage in LPL patients.3 But, it is worthnoting that these results are not fully conclusive because they have been obtained on heterogeneous biological materials and in different separated series.

To clarify the relationship between these two entities, we analysed IgVH gene and somatic mutation pattern on homogeneous material consisting of infiltrated bone marrow trephine biopsies. Patients were retrospectively enrolled according to institutional regulations and after approval from the scientific committees of the tumour banks of the University Hospitals of Bordeaux. SMZL (n=11) and LPL (n=14) cases with bone marrow infiltration were retrospectively retrieved from our files.

Clinical and biological information of each patient are presented in Supplementary Table 1. The median age was 62 years (range, 43–78 years) and 67 years (range, 54–82 years) with a female-to-male ratio of 6/5 and 10/4 for SMZL and LPL patients respectively. Histopathological diagnoses were reviewed by two of us (MP and BP). For eight patients with SMZL, splenectomy allowed to confirm the diagnostic of SMZL. In the three cases remaining without splenectomy, diagnosis was proposed on immunocytological aspect and pattern of medullar involvement. Medullar infiltration of SMZL consisted of intrasinusoidal infiltration, isolated or associated with a nodular pattern. Lymphoma cells were small- to medium-sized, characterized by a pale cytoplasm surrounding a round or oval nucleus with dense chromatin and small nucleolus, without marked features of plasma cell maturation at both morphological and immunological levels. Tumour cells were CD20-positive (highlighting the intrasinusoidal infiltration), DBA44-positive in only four cases (36%) and CD5-, CD23- and CD138-negative. Labelling of CD138 evidenced plasma cell population (<5%) that was polytypic plasmocytosis (reactive or residual?) as shown by κ and λ light-chain labelling. All cases of LPL exhibited an interstitial, nodular or diffuse pattern. Lymphomatous infiltrate was composed of small lymphocytes with a variable degree of plasma cell maturation. Mastocytosis hyperplasia was frequent as assessed by Giemsa stain and CD117 labelling (n=14) (whereas present in only three cases of SMZL). Lymphomatous cells were CD20+, CD5− and CD23−. CD138 positivity was restricted to cells presenting marked plasmacytic differentiation and associated with a monotypic κ or λ light-chain expression. In contrast with SMZL cases, CD23 staining did never reveal follicular dendritic cell network. All LPL cases were DBA44-negative.

We performed the comparative analysis of IgVH gene sequences using Biomed24 primers and IMGT5 and Ig BLAST. The distribution of the VH, DH and JH segments among this series of patients is shown in Table 1, and was compared with published VH frequencies on normal B lymphocytes6, 7 (Supplementary Figure 1). VH usage was heterogeneous in SMZL cases with VH1 in 5/11 cases (45%), VH3 in 3/11 cases (27%) and VH4 in 3/11 cases (27%). JH4, JH6, JH2 and JH1 segments were found in 6/11 (55%), 3/11 (27%), 1/11 (9%) and 1/11 (9%) SMZL cases respectively. Eleven out of 14 cases (79%) of LPL had a rearranged VH gene belonging to the VH3 family, with an overrepresentation of the VH3–23 segment (6/14 cases, 43%). Ten out of 14 (71%) had used the JH4 gene, nine of them (64%) with both VH3 and JH4 rearranged segments. JH4-02 was the most commonly used individual JH4 allele. A slight overrepresentation of the DH gene D3-3 was observed in LPL (4/14 cases, 29%). Thus, VH and JH were different between SMZL and LPL (χ2 test, P=0.02). When compared to VH distribution of normal B cells, SMZL cases appeared to be characterized by an increased usage of VH1 (45% in our study versus 7.1% in normal B cells, P=0.01). As previously described in LPL,3 we found a marked VH3 restriction usage (78.3% of LPL cases in our study versus 33.1% in normal B cells) with a highly significant predominance of VH3-23 subtype (χ2 test, P<10−6).

Table 1 Identification of IgVH, D and J segment, % of homology of VH genes when compared to germline sequences, analysis of IgVH somatic mutations with the multinomial (Lossos) method and CDR3 sequences

Percentages of homology of VH segments with germline sequences are given in Table 1, and the percentage of somatic mutations of this series was given by 100 minus the percentage of homology (Figure 1a). Mean of somatic mutations was 5.7% (ranging 0–7%) in SMZL and 9.3% (ranging 5–16%) in LPL (Student's t-test, P=10−3). Five out of 11 patients (45%) with SMZL had a less than 5% mutation rate of IgVH gene, four of them with less than 2%. By contrast, all LPL cases had an IgVH mutation rate higher than 5% (Figure 1a). It is worthnoting that case M24 with an IgVH mutation rate of 11% was diagnosed as SMZL with an IgG secretory component at 18 g l−1.

Figure 1

Distribution of IgVH mutation rate (a) and CDR3 length (b) among SMZLs and LPLs. LPL, lymphoplasmacytic lymphoma; SMZL, splenic marginal zone lymphoma.

The distribution of replacement (R) and silent (S) mutations in complementary determining (CDR) and framework regions (FR) could be statistically analysed according to the multinomial model proposed by Lossos et al.8 in 8/11 SMZL (those with at least three mutations in the CDR and/or FR sequences) and in all cases of LPL (Table 1). A significant increase of R mutations in CDR with a P-value <5% was observed in 0/8 SMZL and 3/14 (21%) LPL, and a significant decrease of R mutations in FR with a P-value <5% was seen in 3/8 (27%) SMZL and 10/14 (71%) LPL. This strongly suggests that somatic mutations of IgVH in LPL were submitted to a selection process eliminating those that could alter the structure of the immunoglobulin (decrease of R mutations in FR) and increasing the diversity of antigenic-binding sites (increase of R mutations in CDR), a result indicative of antigenic selection.9 Thus, LPL B cells do show the IgVH signature of germinal centre-experienced memory B cells with a high rate of IgVH mutations and a disequilibrium of the distribution of R mutations between FR and CDR.

Analysis of the diversity of VDJ junction evidenced a marked difference in the length of added nucleotides between SMZL and LPL. Translation of IgVH sequences clearly showed that CDR3 length was significantly higher in SMZL than in LPL (Figure 1b, non-parametric Mann and Whitney test, P=10−4). Regarding overlapping cases, cases L5 and L1 with intermediate CDR3 length were typical LPL cases with an IgM peak, and without spleen nor lymph node enlargement, and case M18 with short CDR3 was one of the three cases diagnosed as SMZL without splenectomy. Table 1 shows clearly that increase of the CDR3 length in SMZL corresponds to GC-rich N insertions, classically generated by Terminal deoxyribonucleotide Transferase.10 The enhanced length of CDR3 segment, the low mutation rate of IgVH genes and the homogeneous distribution of R mutations between FR and CDR suggest that tumour B cells of SMZL cases were selected by autoantigens outside germinal centres.11, 12, 13 It is worthnoting that digestive mucosa-associated lymphoid tissue lymphoma, believed to correspond to expansion of autoreactive B lymphocytes, also exhibits long CDR3 segment of IgVH genes.12

For the first time, we have extensively analysed IgVH gene sequences of bone marrow infiltration of both SMZL and LPL, with the major advantage of a homogeneous and comparable material between these two categories of tumour. We showed that low IgVH somatic mutation rate and long CDR3 segments but with no significant IgVH restriction usage are characteristics of SMZLs whereas IgVH of LPL cases frequently involved VH3 and JH4 and had a high mutation rate with an R mutation profile indicative of an antigen selection process. These results definitely demonstrate that the history to antigen exposure of these tumours diagnosed on the sole basis of clinical, morphological and immunohistochemical criteria, was completely different, arguing very strongly in favour of two different unrelated entities derived from B cells with two different fates and likely from two different compartments.


  1. 1

    Berger F, Traverse-Glehen A, Felman P, Callet-Bauchu E, Baseggio L, Gazzo S et al. Clinicopathologic features of Waldenstrom's macroglobulinemia and marginal zone lymphoma: are they distinct or the same entity? Clin Lymphoma 2005; 5: 220–224.

    Article  PubMed  Google Scholar 

  2. 2

    Papadaki T, Stamatopoulos K, Belessi C, Pouliou E, Parasi A, Douka V et al. Splenic marginal-zone lymphoma: one or more entities? A histologic, immunohistochemical, and molecular study of 42 cases. Am J Surg Pathol 2007; 31: 438–446.

    Article  PubMed  Google Scholar 

  3. 3

    Martin-Jimenez P, Garcia-Sanz R, Balanzategui A, Alcoceba M, Ocio E, Sanchez ML et al. Molecular characterization of heavy chain immunoglobulin gene rearrangements in Waldenstrom's macroglobulinemia and IgM monoclonal gammopathy of undetermined significance. Haematologica 2007; 92: 635–642.

    CAS  Article  PubMed  Google Scholar 

  4. 4

    van Dongen JJ, Langerak AW, Bruggemann M, Evans PA, Hummel M, Lavender FL et al. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 Concerted Action BMH4-CT98-3936. Leukemia 2003; 17: 2257–2317.

    CAS  Article  PubMed  Google Scholar 

  5. 5

    Giudicelli V, Chaume D, Lefranc MP . IMGT/V-QUEST, an integrated software program for immunoglobulin and T cell receptor V-J and V-D-J rearrangement analysis. Nucleic Acids Res 2004; 32: W435–W440.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  6. 6

    Brezinschek HP, Foster SJ, Brezinschek RI, Dorner T, Domiati-Saad R, Lipsky PE . Analysis of the human VH gene repertoire. Differential effects of selection and somatic hypermutation on human peripheral CD5(+)/IgM+ and CD5(−)/IgM+ B cells. J Clin Invest 1997; 99: 2488–2501.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7

    Stamatopoulos K, Belessi C, Moreno C, Boudjograh M, Guida G, Smilevska T et al. Over 20% of patients with chronic lymphocytic leukemia carry stereotyped receptors: Pathogenetic implications and clinical correlations. Blood 2007; 109: 259–270.

    CAS  Article  Google Scholar 

  8. 8

    Lossos IS, Alizadeh AA, Eisen MB, Chan WC, Brown PO, Botstein D et al. Ongoing immunoglobulin somatic mutation in germinal center B cell-like but not in activated B cell-like diffuse large cell lymphomas. Proc Natl Acad Sci USA 2000; 97: 10209–10213.

    CAS  Article  PubMed  Google Scholar 

  9. 9

    Chang B, Casali P . The CDR1 sequences of a major proportion of human germline Ig VH genes are inherently susceptible to amino acid replacement. Immunol Today 1994; 15: 367–373.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10

    Alt FW, Baltimore D . Joining of immunoglobulin heavy chain gene segments: implications from a chromosome with evidence of three D-JH fusions. Proc Natl Acad Sci USA 1982; 79: 4118–4122.

    CAS  Article  PubMed  Google Scholar 

  11. 11

    Shlomchik MJ, Aucoin AH, Pisetsky DS, Weigert MG . Structure and function of anti-DNA autoantibodies derived from a single autoimmune mouse. Proc Natl Acad Sci USA 1987; 84: 9150–9154.

    CAS  Article  PubMed  Google Scholar 

  12. 12

    Yumoto N, Kurosu K, Furukawa M, Mikata A . CDR3 sequences of MALT lymphoma show homology with those of autoreactive B-cell lines. Jpn J Cancer Res 1999; 90: 849–857.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  13. 13

    Wardemann H, Yurasov S, Schaefer A, Young JW, Meffre E, Nussenzweig MC . Predominant autoantibody production by early human B cell precursors. Science 2003; 301: 1374–1377.

    CAS  Article  Google Scholar 

Download references


This work was supported by the Cancéropôle Grand-Sud-Ouest, the Institut National contre le Cancer (INCa) contrat ACI-2004–2006 and ACI 2007–2008, Ligue contre le Cancer Comité du Limousin and Ligue Nationale contre le Cancer, Conseil Régional du Limousin. We thank the tumour bank of Hôpital Haut Leveque, Pessac, for providing biological samples.

Author information



Corresponding author

Correspondence to J Feuillard.

Additional information

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

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Parrens, M., Gachard, N., Petit, B. et al. Splenic marginal zone lymphomas and lymphoplasmacytic lymphomas originate from B-cell compartments with two different antigen-exposure histories. Leukemia 22, 1621–1624 (2008). https://doi.org/10.1038/leu.2008.24

Download citation

Further reading


Quick links