Abstract 317

Evidence for Antigen-driven Development of Molecularly Classified Burkitt Lymphomas

The presence of somatic mutations and intraclonal diversity (ID) in the variable region of rearranged immunoglobulin (Ig) genes is commonly used as a marker of the cellular passage of B cells through the germinal center (GC) reaction in secondary lymphoid tissues following an in vivo antigen encounter. Such features have also been documented in different subgroups of aggressive B-cell lymphomas, suggesting that antigen selection pressure has taken place during lymphoma development. We undertook a detailed analysis of the Ig repertoire and somatic hypermutation (SHM) status of 54 molecularly classified Burkitt lymphomas (BL) (Hummel et al. 2006). The IGHV gene repertoire was remarkably biased, with only three genes accounting for 40.8% of all cases (IGHV4-34, 18.5%; IGHV4-39, 11.1%; IGHV3-30, 11.1%). Of note, this bias was even more evident when cases were grouped according to age < or > 14 years, with the IGVH4-34 and IGVH4-39 rearrangements occurring almost exclusively in pediatric BL (7/11 and 5/6 cases respectively). Only 2/54 cases (3.7%) carried unmutated IGHV genes (100% germline identity, GI); 12/54 cases (22.2%) carried IGHV genes of borderline/minimally mutated status (98–99.9% GI); finally, the remaining 40/54 cases (74.1%) carried mutated IGHV genes (<98% GI). The median GI of the cohort was 95.2% (range: 86.4–100%) with a significantly different mutation load according to age below or above 14 years (GI 95.8 vs. 92.9 respectively, p=0.0168). Shared replacement mutations (“stereotyped” amino acid (AA) changes) at certain codon positions were identified in 12/54 (22.2%) BL cases. Stereotyped AA changes were found only amongst rearrangements utilizing the IGHV4-34 and IGHV4-39 genes, strongly indicating evidence for an antigen driven SHM process in BL. The majority of stereotyped AA changes identified here were distinct from those previously reported in other entities, in particular CLL, and thus can be considered as BL-biased. Also, of note, the IGVH4-34 specific motif responsible for binding the N-acetyl-lactosamine antigenic determinant was found intact in 11 IGVH4-34 BL sequences, implying that these clones retained the ability to bind to and be activated by superantigens. Furthermore, the distribution of mutations was different in IGHV4-34 vs. non-IGHV4-34 cases, as evidenced by low R-to-S ratios in the heavy complementarity-determining regions (especially HCDR1), indicating a different selection pattern of VH-genes according to varying antigens.

To assess the presence of ID through ongoing SHM, we evaluated 411 subcloned sequences from 11 IGVH4-34 and 6 IGVH4-39 cases. Only one case carried identical sets of subcloned sequences; seven cases carried mutations in single subcloned sequences that were considered as unconfirmed; finally, 9 cases carried at least one confirmed mutation among two or more subcloned sequences and, thus, were considered as exhibiting confirmed ID. However, analysis of the distribution of the so called unconfirmed changes provided further evidence for the very precise targeting of mutations introduced as part of the ID process. Thus, certain changes identified in single subcloned sequences from one case were shared by most or all subcloned sequences of another case utilizing the same IGHV gene (“confirmed by another case”). In conclusion, our data demonstrate that BL is characterized by a highly distinctive IG gene repertoire with a precisely targeted SHM process. In addition, they provide strong evidence for a functionally driven SHM, supporting a role for antigen-driven selection of the clonogenic progenitors.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.