Purpose: Elevated serum M component of IgG isotype is typically associated with multiple myeloma (MM). However, our group has previously reported cases with an elevated serum monoclonal IgG and a leukaemic B-cell lymphoplasmocytoid lymphoma (LPL) similar to Waldenstrom’s macroglobulinaemia (WM). The aim of this study was to extend analysis of IgH locus events in a larger series of IgG-secreting LPL.
Patients and Methods: We investigated 20 patients with an elevated serum monoclonal IgG (>20g/l) and LPL (IgG-LPL). Morphological classification and immunophenotyping analysis were performed at diagnosis (serum IgG>4 g/l, CD19+ cells >30%, presence of lymphoplasmocytoid cells in blood and/or bone marrow). Histological classification and FISH analysis were performed when possible to further characterize those cases. Analysis of VH genes was carried out from RNA with VHLeader and CH primers. 14 patients were examined for both IgG and IgM transcripts; VH-Cμ and VH-Cg transcripts could be compared in 9 patients.
Results: Of 25 Ig VH rearrangement sequences, 23 were functional and expressed in each case. VH3 family members appeared to be over-represented (19/21 patients (90.5%) as compared to 40/71 (56.3%) in normal B-cell repertoire (1). VH3-23 was the most frequently used segment (10/21 patients) and is frequently utilized in normal B-cells. IgG-LPL JH family use resembled the normal B-cell repertoire (predominance of JH4 and JH6 segments). The median CDR3 length was 10 amino acids [5–19]. However, and in contrast with features seen in other leukaemia, there was no evidence of homologous CDR3 motifs. All VH genes revealed highly somatically mutated sequences, with a median mutation rate 8.8% [0.7 – 11.1%] (IMGT database(2)). We compared pre (VH-Cμ) and post-switch (VH-Cg) transcripts, and 4/9 patients had identical clonally-derived sequences, and two 2/9 had divergent sequences.
Interpretation and conclusion: This intended study of IgG-LPL reveals consistent features that argue for common origins of IgG-LPL with Waldenstrom’s macroglobulinemia. One feature is extensive somatic mutations in VH genes, suggesting origins from a cell that may have undergone successive rounds of mutation. Patterns of mutations in pre-and post-switched clonally derived sequences suggest that the final neoplastic event has occurred in a IgM+ memory cell undergoing isotype switch. However, no aberrant chromosomal translocation accrue at the IgH locus, as apparent from FISH data. This indicates that, unlike in typical MM, switch activity does not generate 14q32 abnormalities nor that such lesions play a role in the pathogenesis of LPL. Extensive mutations are also seen in VH genes in WM, and there is evidence that WM cells can undergo class switch events in vivo. In this tumor, switching occurs at a low subclonal level in some cells, which in rare cases can lead to the emergence of a dual population of clonally identical IgM and IgG expressing WM tumor cells at a later stage of disease (7). In typical WM, 14q32 abnormalities are also generally not seen. These data suggest that IgG-LPL and WM could be two variants of the same entity, with IgG-LPL exposed to persistent switch stimuli following transformation.
Disclosures: No relevant conflicts of interest to declare.