Oligomannose-Type Glycans in the Antigen Binding Site Identify Origin and Prognosis of Diffuse Large B Cell Lymphoma
Bibliographic record
Abstract
The acquisition of N-glycosylation sites (AGS) in the complementarity-determining region (CDR) of the tumor immunoglobulin (Ig) is an early clonal requirement of classic follicular lymphoma (FL). In FL, the AGS are occupied by oligomannose-type glycans, which mediate lymphoma Ig interaction with the microenvironmental lectin dendritic cell-specific intercellular adhesion molecule-3-grabbing nonintegrin (DC-SIGN). This interaction induces a persistent low-level signal, favouring adhesion and ultimately survival and growth of the lymphoma cells in their protected environment . Oligomannose-type glycans can also occupy the CDR-located AGS of a subset of germinal-centre B cell-like (GCB) diffuse large B-cell lymphomas (DLBCL) but the consequences of AGS in DLBCL are less known. We determined the frequency, distribution, and composition of the glycan structures occupying the AGS in DLBCL using a discovery-validation approach with two large independent international public cohorts (BCC, dataset ID: EGAD00001003783; NCI accession phs001444.v1.p1), following our RNAseq Ig assembly pipeline (Ig-Seq-R) and high-resolution mass-spectrometry. Glycan types were compared to our FL cohort (12 samples) and correlated with DLBCL cell of origin (COO), genetic, clinical characteristics and outcome. The full IGHV-IGHD-IGHJ-IGHC transcript sequences were obtained from 251 DLBCL of the BCC, and 339 of the NCI cohort. Sample distribution by COO, clinical or molecular features, LymphGen subtype, progression-free survival (PFS) and overall survival (OS) were not different from the extended public cohorts. AGS were observed in 48-55% GCB-DLBCL, with 84-85% of them located in the CDR. Overall, frequency of CDR-located AGS contrasted dramatically between GCB-DLBCL (41-46%) and activated B-cell like (ABC)-DLBCL (7-10%). Also, within GCB-DLBCL, CDR-located AGS associated mostly with the EZB subtype (68-73%). However, only 66-61% EZB were AGS+ve, all of which were CDR+ve, while the remaining 34-39% had no AGS (AGS-ve EZB). The glycan structures of the synthetic F(ab)s from 35 AGS+ve DLBCL and 12 FL were analyzed by site-specific mass spectrometry. The DLBCL comprised 12 AGS+ve EZB and 23 AGS+ve non-EZB (14 CDR+ve, 9 located in the framework region). Glycan analysis revealed that all AGS+ve EZB and FL were invariably occupied by oligomannose-type glycans (Ig-Mann+ve DLBCL). In contrast, the other AGS+ve DLBCL were not occupied by oligomannose type glycans. They were either occupied by complex glycans (the remaining non-EZB GCB-DLBCL) or not occupied (preferentially the ABC-DLBCL), irrespective of AGS location. These data indicated that CDR+ve EZB Ig were universally and exclusively Ig-Mann+ve DLBCL, share the COO of FL, and have likely been influenced by an environmental driver distinct from the other GCB-DLBCL and ABC-DLBCL. We compared the clinical behavior of Ig-Mann+ve DLBCL versus other DLBCL subtypes including AGS-ve EZB and the other non-EZB GCB-DLBCL and ABC-DLBCL. In both the BCC and NCI cohorts, Ig-Mann+ve DLBCL status identified the subset with the worst PFS and OS of all GCB-DLBCL, not different from ABC-DLBCL (Figure 1A). This contrasted with the other GCB-DLBCL (AGS-ve/EZB and AGS+ve/non-EZB), which all shared an extremely good PFS and OS (Figure 1A). Univariate and multivariate analyses revealed that Ig-Mann+ve status was an independent prognostic factor for PFS and OS (Figure 1B). Gene expression profile analysis revealed that Ig-Mann+ve status associated with an enrichment of MYC, PI3K-Akt-mTORC1, pro-survival, and cell cycle pathways, while proinflammatory and apoptotic pathways were decreased in the Ig-Mann+ve DLBCL compared to AGS-ve/EZB and AGS+ve/non-EZB, independently of MYC translocations and double-hit signature. These data indicate identical COO of a EZB subset with FL cells, and point to a highly selective chronic environmental pressure on Ig-Mann+ve DLBCL negatively affecting patients survival. The combination of Ig gene analysis with the lymphGen classification can predict Ig glycan structure and provide a new fundamental approach to identifying the most aggressive GCB-DLBCL subtype, and the precise environmental tumour interaction (DC-SIGN:Ig-Mann) to intercept therapeutically early in the natural history of Ig-Mann+ve DLBCL.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".