1009 - IMMUNOGENETICS AND ANTIGEN REACTIVITY PROFILING CONTRIBUTE TO UNRAVELLING THE ONTOGENY OF CLL STEREOTYPED SUBSET #4

Introduction:
CLL subset #4 (IGHV4-34/IGKV3-20) is the largest stereotyped subset with somatically hypermutated clonotypic B cell receptor immunoglobulin (BcR IG) and a prototype for indolent disease. The variable heavy complementarity determining region 3 (VH CDR3) of the clonotypic BcR IG is long and enriched in positively charged residues, reminiscent of pathogenic anti-DNA antibodies and contains a (K/R)RYY motif that has been demonstrated to be ‘CLL-biased’ and exclusive to subset #4. Additionally, subset #4 BcR IG are capable of specific homotypic interactions which induce cellular activation. Critical structural determinants mediating such interactions are provided by the gamma constant domain, offering a plausible functional explanation for the ubiquitous expression of IgG-switched BcR IG in subset #4. Subset #4 BcR IG have also been reported to display intense intraclonal diversification in the context of ongoing SHM, suggestive of active, ongoing interaction with antigen(s).
Thus far, the immunogenetic characterization of subset #4 has relied on low-throughput sequencing approaches that are inherently limited, precluding comprehensive assessment of antigenic impact on (sub)clonal composition. In this study, we sought to overcome this limitation by performing next-generation sequencing (NGS) of heavy (HC) and light (LC) chains of 6 subset #4 patients and explore whether/how (auto)antigens could drive the clonal evolution by comparing the reactivity profile of the clonotypic BcR IG with those detected at subclonal level.

Methods:
We studied longitudinal samples from 6 subset #4 and 6 non-subset #4 IGHV4-34-expressing M-CLL cases. Clonotypic HC and LC gene rearrangements were analyzed by NGS. The clonotype accounting for the majority of NGS reads in a given sample was defined as dominant; related clonotypes expressing the same IGHV/IGKV gene and VH CDR3 amino acid (aa) sequence yet differing in the aa sequence of the VH and/or VK domains were defined as subclonotypes. The dominant clonotypes of all cases as well as 32 subclonotypes of subset #4 cases were expressed as recombinant monoclonal antibodies (rmAbs) and screened for antigen reactivity using ELISA and flow cytometry.

Results:
NGS revealed two distinct patterns of subclonal architecture in the IGH repertoire of subset #4. The first was characterized by the presence of a single dominant clonotype (4 cases) whereas the second by the co-existence of 2 significantly expanded clonotypes with comparable frequencies (2 cases). Turning to the IGK repertoire, 5/6 cases carried a single dominant clonotype, while 1 case carried more than 1 expanded clonotypes. In contrast, a single dominant clonotype was detected in all non-subset #4 cases for both HC and LC. Longitudinal analysis revealed significantly (p < 0.05) more pronounced subclonal drift (changing relative frequencies) of the expanded clonotypes in subset #4 vs non-subset #4 cases, particularly for the HC. Surprisingly, in all subset #4 cases we identified clonotypic IgG transcripts (same IGH gene and VH CDR3 length, high VH CDR3 aa sequence identity) with 100% germline IGHV identity (‘truly unmutated’), present at low frequency (median: 0.003%, range: 0.0003-0.03%). Prompted by this, we produced rmAbs for both the dominant, somatically hypermutated BcR IG clonotype of all 6 subset #4 cases, as well as 32 truly unmutated subclonotypes (from all studied subset #4 cases) and compared their antigen reactivity profiles. We report consistent, significantly (p < 0.05) stronger reactivity of the truly unmutated, subclonotypic BcR IG compared to the mutated BcR IG expressed by the dominant CLL #4 clone. With few exceptions, all subclonotypic truly unmutated BcR IG displayed more intense recognition of dsDNA and lipopolysaccharides, antigenic agents from CMV, Influenza A and Mycoplasma pneumoniae as well as viable HEK293 and Jurkat T cells.

Conclusions:
Our findings support that SHM in CLL subset #4 is functionally driven by persistent selection by (auto)antigens, ultimately leading to a significant redemption from autoreactivity. The identification of truly unmutated clonotypic IgG gene transcripts likely reflects a complex trajectory of clonal evolution, offering hints about the precise timing of SHM in relation to class switch recombination in the natural history of CLL subset #4.