Diffuse large B-cell lymphoma and red cell autoimmunity - clinical role and pathogenesis

Diffuse large B-cell lymphoma (DLBCL) is the most common form of B-cell non-Hodgkin lymphoma (B-NHL) and results in significant morbidity and mortality despite advancements in treatment. Lymphoma and autoimmune disease both result from breakdowns in normal cell regulatory pathways, and epidemiological studies have confirmed both that B-NHL is more likely to develop in the setting of autoimmune diseases and vice versa. Direct Antiglobulin Test (DAT) positivity is a marker of red cell autoimmunity and occurs at higher rates in patients with DLCBL. The immunoglobulin genotype IGHV4-34 is seen more frequently in DLBCL clones than in normal B cells and has intrinsic autoreactivity to self-antigens on red cells, which is largely mediated by two motifs within the first framework region (FR1); Q6W7 and A24V25Y26. These motifs form a hydrophobic patch which determines red cell antigen binding and are frequently mutated away from self-reactivity in normal B cells. If mutations in the hydrophobic patch region of B cells using IGHV4-34 do not occur this may provide constant B cell receptor signalling which encourages lymphoma development, a theory known as antigen driven lymphomagenesis. This project aimed to explore the relationship between red cell autoimmunity and DLBCL. We found that red cell autoimmunity as identified by DAT positivity was more common in DLBCL as compared to healthy controls (20.4% vs 3.7%, p=0.0005). The prognostic value of DAT positivity was explored by conducting a retrospective cohort study of DLBCL patients diagnosed in our centre; patients included had DAT performed as part of screening at the time of DLBCL diagnosis. Univariate analysis found no significant difference in overall survival (OS) between the DAT positive (DAT+) and DAT negative (DAT-) groups (p=0.19). When controlling for cell of origin at 5 years DAT predicted for poor OS, with DAT- patients 8.927 times more likely to be alive 5 years after diagnosis (B=2.189, p=0.042, OR=8.927, 95%CI 1.084-73.522). We used high-throughput sequencing to look for common mutations that may drive lymphomagenesis and autoimmunity in DLBCL and could be potential therapeutic targets. The most recurrently mutated genes in 15 tested samples were KMT2D (n=13), MYOM2 (n=9), EP300 (n=8), SPEN (n=7), and ADAMTSL3 (n=7), which were mutated in both DAT+ and DAT- groups. BIRC3, a gene in the non-canonical NFKB pathway involved in apoptosis regulation and previously linked to type 1 diabetes mellitus, was mutated in DAT+ patients but not in DAT- patients. This may represent a target for future pathogenesis and novel therapeutics studies. As with previous studies, IGHV4-34 was over-represented (15.6% of fresh samples) in our DLBCL cohort. We hypothesised that hydrophobic patch mutations would be less frequent in our IGHV4-34-expressing DLBCL cohort than reported rates in normal B cells. Of 6 IGHV4-34-expressing DLBCL samples 5 had unmutated hydrophobic patch mutations. Mutation analysis of these 5 samples demonstrated high frequency of mutations in several genes, including CREBBP and NCOR2. Mutations CREBBP and NCOR2 may work in conjunction with the preserved QW and AVY motifs to promote lymphomagenesis in IGHV4-34-expressing B cells and may present targets for future research into treatment protocols