IGG-SWITCHED CLL HAS A DISTINCT IMMUNOGENETIC SIGNATURE FROM THE COMMON MD VARIANT: ONTOGENETIC IMPLICATIONS

Background:

IgG-switched CLL (G-CLL) is a relatively rare variant of CLL whose origin and ontogenetic relationship to the common IgM/IgD (MD-CLL) variant remains undefined. Previous studies have reported that most G-CLL cases carry somatically mutated B-cell receptors (BcR) pointing to possible derivation from post-germinal center B-cells. However, the relevant studies concerned small series, precluding definitive conclusions.

Aims:

We sought for evidence about the origin of G-CLL versus MD-CLL by comprehensively profiling the respective immunoglobulin (IG) gene repertoires in the largest series thus far studied for this purpose.

Methods:

Detailed immunogenetic analyses were conducted in a series of 1256 cases selected based on expression of MD (n=1087) or G isotype (n=169) as determined by peripheral blood flow cytometry (n=1019) or expression of tumor-specific Cμ/Cδ and/or Cγ transcripts as determined by RT-PCR (n=186).

Results:

The IGHV gene repertoire of G-CLL differed significantly from that of MD-CLL, with over-representation of the IGHV4-34 (P<0.0001) and IGHV4-39 (P=0.003) genes, and, in contrast, under-representation of the IGHV1-69 (P=0.003), IGHV3-21 (P=0.017), IGHV1-2 (P=0.04) and IGHV3-48 (P=0.05) genes. Regarding somatic hypermutation (SHM), G-CLL included significantly more cases with mutated IGHV genes (following the 98% cut-off value for identity to the germline) compared to MD-CLL (84% vs 46%, respectively; P<0.0001). At the other extreme of the SHM spectrum, significantly more MDCLL cases carried BcR IGs with no SHM (100% identity to the germline) compared to G-CLL (32% versus 11%, P<0.0001). The search for CLL-specific stereotyped BcR IGs revealed that the extreme skewing of the G-CLL repertoire was due to the fact that almost one-third of all cases was represented by only three subsets utilizing either the IGHV4-34 gene, namely mutated subsets #4 (relative frequency 18.3% of all G-CLL) and #16 (4%), or the IGHV4-39 gene, namely unmutated subset #8 (7.3%). These subsets (especially, subsets #4 and #8) represent polar opposites in terms of prognosis and outcome, with subset #4 cases experiencing indolent disease unlike subset #8 cases who tend to follow an aggressive disease course often complicated by the development of Richter’s syndrome. In keeping with our previous reports, subset #4 and #16 rearrangements identified in the present study exhibited recurrent SHM at certain positions throughout the VH domain. However, despite an overall high SHM load, almost all IGHV4-34 rearrangements assigned to these two G-CLL subsets carried intact FR1 motifs for superantigenic-like interactions with N-acetyllactosamine epitopes present in self and exogenous antigens. In sharp contrast to subsets utilizing the IGHV4-34 and IGHV4-39 genes, all major sterotyped subsets utilizing the IGHV1-69, IGHV3-21, IGHV1-2 and IGHV3-48 genes (defined as reported in Blood 2012;119:4467) concerned exclusively MD-CLL.

Summary and Conclusions:

G-CLL exhibits an overall distinct immunogenetic signature from MD-CLL, even when restricting the comparison to cases with mutated BcR IGs, prompting speculations about distinct ontogenetic derivation and/or distinct immune triggering. Nonetheless, questions abound regarding e.g. the presence of class switch recombination in the absence of SHM in subset #8 or the over-representation of the IGHV4-34 gene in G-CLL, especially in light of studies showing that healthy IGHV4-34 cells are physiologically excluded from germinal center reactions in order to maintain peripheral B-cell tolerance.

Abstrakta v časopise Haematologica 2013, Suppl1

Online Program