Comprehensive Analysis of Malignant B-Cell Receptors Provides Insights into B-Cell Lymphoma Pathogenesis

Introduction: B-cell survival and proliferation rely on antigen-specific B-cell receptor (BCR) signaling. Disrupted BCR activation is a key driver of B-cell lymphoma (BCL) pathogenesis, yet the relationship between BCR structure and specificity and the malignant transformation of B-cell lineages remains poorly understood. This study aimed to characterize these complex relationships and the structural variations in BCRs across BCL subtypes.

Methods: Comprehensive transcriptomic analysis of malignant BCRs was performed on blood and tumor samples from diverse BCL subtypes: chronic lymphocytic leukemia (CLL) with unmutated (UM-CLL, n = 304) and mutated (M-CLL, n = 361) IGHV, activated B-cell (ABC, n = 643) and germinal center B-cell (GCB, n = 472) diffuse large BCL (DLBCL), follicular lymphoma (FL, n = 129), mantle cell lymphoma (MCL, n = 139), and multiple myeloma (MM, n = 709). A Bayesian classifier was developed to distinguish BCL from healthy blood samples based on BCR repertoire clonality. The majority of observed patient repertoires (98.9%) were correctly classified as BCLs, resulting in a dataset of heavy and light malignant BCR chains. Specific VDJ combinations, complementary-determining region 3 (CDR3) sequences, somatic hypermutations (SHM), isotypes, and other structural features of the BCR chains were evaluated.

Results: Comparative analysis revealed marked variations in malignant BCR isotype distribution and IGHV SHM rate across BCL subtypes, reflecting their origins from distinct stages of B-cell differentiation. UM-CLL and MCL, often originating from naive B cells, predominantly produced BCRs with the IgM/D isotype (UM-CLL: 94.7% and MCL: 100%) and very low IGHV SHM rates (median rates: UM-CLL: 0 substitutions per base pair [subst/bp]; MCL: 0.01 subst/bp). M-CLL, which often exhibits characteristics of memory B cells, generally possessed BCRs with an unswitched isotype (92.1%) but also featured an elevated IGHV SHM rate (0.04 subst/bp). Germinal center-associated BCLs were more likely to have BCRs with switched isotypes and high SHM rates (DLBCL ABC: 85.5% IgM/D and ~0.06 subst/bp; DLBCL GCB: 42.4% and 0.08; FL: 51.7% and 0.08). MM BCRs, representing the terminal stage of plasma cells, had switched IgG or IgA isotypes and elevated SHM rates (MM: 0% IgM/D and 0.06 subst/bp).

Despite dramatic variations in the number of accumulated IGHV SHMs, most BCLs featured consistent SHM nucleotide context frequencies and correlations between SHM rates of heavy and light malignant BCR chains, suggesting BCL subtypes do not affect the SHM accumulation process. The nonsynonymous to synonymous substitution ratio (dN/dS) was stable across BCL subtypes (median dN/dS = 0.45), indicating negative selection maintains BCR protein structure and specificity.

Certain BCLs had distinct BCR features compared to healthy donors. For example, UM-CLL had a significantly longer heavy chain CDR3 (22 vs 17 amino acids, p < 10-4). Differences in light chains were also noted. Compared to normal clonotypes, M-CLL (51.2%, p = 0.03) was less likely and DLBCL ABC (68.7%, p = 0.04) was more likely to exhibit BCRs containing IGK light chains. Stereotyped variable (V) genes were observed in BCRs of UM-CLL (IGHV1-69: 24.5%), MCL (IGHV3-23: 11.7%), FL (IGHV3-33: 12.4%), and DLBCL ABC (IGHV4-34: 30.5%), indicating low diversity of V genes in BCLs compared to normal clonotypes and, in turn, that B cells with certain V genes are prone to malignant transformation.

Conclusion: In the present work, we identified several lymphoma-specific BCR features that distinguish malignant BCRs from normal B-cell clonotypes. Our findings support the hypothesis that the structure and specificity of BCRs play a major role in BCL pathogenesis and may serve as biomarkers for clinical decision-making.