Non-Hodgkin Lymphoma: GWAS to Drug Target Druggability Analysis

COMPREHENSIVE GWAS-TO-DRUG-TARGET DRUGGABILITY ANALYSIS

Non-Hodgkin Lymphoma (NHL)

Section 1: Disease Identifiers

Cross-reference richness: MeSH D008228 has 3,482 cross-references (largest); EFO:0005952 links to 92 GWAS associations and 272 ChEMBL molecules; MONDO:0018908 links to 847 clinical trials.

Section 2: Gwas Landscape

Summary:

• Total associations: 92 (via EFO:0005952)
• Unique GWAS studies: 28
• Key publications: Rashkin SR (Nat Commun, 2020), Berndt SI (Leukemia, 2022), Sato G (Nat Commun, 2023), Verma A (Science, 2024), Thorball CW (Haematologica, 2020), Liu TY (Sci Adv, 2025)

TOP 50 GWAS Associations (sorted by p-value)

Key observations:

• HLA region dominance: 6p21.32-33 harbors the strongest signals (p=3e-19), consistent with NHL being an immune-mediated malignancy
• Replicated loci: IRF4 (6p25.3), HLA-DQB1, PVT1 (8q24.21), BCL2 (18q21.33) replicated across multiple studies
• Strongest effect size: IRF4 region (OR=1.53), HLA-DRB1/DQA1 (OR=1.43)

Section 3: Variant Details (Dbsnp)

Variant Classification by Functional Consequence

Genetic Evidence Tier Classification

Section 4: Mendelian Disease Overlap

Mendelian Genes Associated with NHL

Genes with BOTH GWAS + Mendelian Evidence

Critical finding: No single gene has both genome-wide significant GWAS associations AND Mendelian NHL mutations. However, several Mendelian genes (BRAF, TP53, BCL2) are in pathways directly connected to GWAS genes (BCL2L11 → BCL2, BRAF → MAPK). BCL2 is the closest overlap: GWAS association at 18q21.33 (p=2e-12) and BCL2 is a core pathogenic gene in lymphomagenesis, though the Mendelian ClinVar entries are linked to the broader MONDO concept rather than BCL2 mutations specifically causing familial NHL.

Section 5: Gwas Genes To Proteins

Summary: ~35 unique protein-coding genes identified from GWAS; ~25 map to UniProt protein products (remainder are lncRNAs like PVT1, SNHG10, or pseudogenes).

TOP Protein-Coding GWAS Genes

Section 6: Protein Family Classification

Summary

Section 7: Expression Context

Disease-relevant tissues/cell types for NHL: B lymphocytes, germinal center B cells, T cells, lymph nodes, spleen, bone marrow, thymus.

TOP 30 GWAS Gene Expression Analysis

Key finding: The majority of top GWAS genes (IRF4, BCL2, CD40, CXCR5, MEF2B, TCF3, CD86) show strong lymphoid/B-cell-specific expression, directly relevant to NHL pathobiology. NTM and GIPC2 are NOT expressed in disease-relevant tissue (lower confidence targets).

Section 8: Protein Interactions

Hub Genes (by BioGRID interaction count)

GWAS Gene-Gene Interactions (Pathway Clustering)

Key interactions among GWAS genes:

• BCL2 ↔ BCL2L11: Direct binding (BH3-domain interaction) — central to apoptosis regulation in lymphoma
• CD86 ↔ CD40: Both in B-cell activation/costimulation pathways
• IRF4 ↔ IRF8: Both interferon regulatory factors; co-regulate immune genes
• BRAF → BCL2L11: BRAF/MAPK pathway regulates BIM (BCL2L11) phosphorylation and stability
• CD40 → BCL2: CD40 signaling upregulates BCL2 expression in B cells

Undrugged GWAS Genes with Drugged Interactors

Section 9: Structural Data

Structure Availability Summary

Structure Details for Key Targets

Undrugged Targets Structure Status

Section 10: Drug Target Analysis

Summary

Genes with APPROVED Drugs

Key Approved Drugs for NHL (from MeSH → ChEMBL, 272 molecules)

Phase 4 (Approved):

• Rituximab, Obinutuzumab, Ofatumumab (anti-CD20)
• Venetoclax (BCL2 inhibitor)
• Ibrutinib, Acalabrutinib, Zanubrutinib (BTK inhibitors)
• Idelalisib, Copanlisib, Duvelisib, Umbralisib (PI3K inhibitors)
• Bendamustine, Cyclophosphamide, Doxorubicin (chemotherapy)
• Brentuximab vedotin, Polatuzumab vedotin, Loncastuximab tesirine (ADCs)
• Tisagenlecleucel, Axicabtagene ciloleucel (CAR-T)
• Pembrolizumab, Nivolumab, Atezolizumab (checkpoint inhibitors)
• Mosunetuzumab (bispecific)
• Tafasitamab (anti-CD19)
• Romidepsin, Belinostat (HDAC inhibitors)
• Temsirolimus, Everolimus (mTOR inhibitors)
• Ruxolitinib (JAK inhibitor)

OPPORTUNITY GAP: 52% of GWAS genes have NO drug development at all — these represent potential novel targets.

Section 11: Bioactivity & Enzyme Data

TOP Most-Studied Proteins (ChEMBL target records)

Enzyme Data for QPCT (BRENDA)

QPCT (glutaminyl-peptide cyclotransferase) is a metalloenzyme (zinc-dependent) with:

• 40+ crystal structures with inhibitor co-crystals
• Multiple imidazole-based and benzimidazole inhibitors (PBD-150, SEN177, NHV-1009)
• Active site is druggable; zinc-chelating pharmacophore
• Druggability: HIGH — well-characterized enzyme with known inhibitors for Alzheimer’s disease (CCL2 modification); could be repurposed

For UNDRUGGED genes:

• IRF4, IRF8, EOMES, MEF2B, TCF3: No ChEMBL targets — transcription factors, traditionally undruggable, though degrader approaches (PROTACs, molecular glues) are emerging
• GRAMD1B, CCHCR1, NTM, AIF1, GIPC2: No bioactivity data — largely uncharacterized as drug targets

Section 12: Pharmacogenomics

PharmGKB Gene Status

All 10 queried genes are PharmGKB VIP (Very Important Pharmacogenes), indicating established drug-gene interactions.

Key pharmacogenomic implications:

• BCL2 expression/variants predict venetoclax response — directly relevant for NHL
• BRAF V600E guides kinase inhibitor selection (hairy cell leukemia)
• IRF4 expression is a biomarker for lenalidomide/IMiD response — critical for DLBCL (ABC subtype)
• TP53 mutation status dictates chemotherapy response/resistance

Section 13: Clinical Trials

Total NHL clinical trials: 847+ (from MONDO:0018908)

Breakdown by Phase

TOP 30 Drugs in NHL Clinical Trials

*Approved for other indications

GWAS Gene Targeting in Clinical Trials

Section 14: Pathway Analysis

TOP Pathways Enriched in GWAS Genes (Reactome)

Pathway druggability insight: Even when a GWAS gene itself is undruggable (e.g., IRF4 as a TF), its pathway contains druggable nodes: IRF4 is in interferon signaling (JAK/STAT inhibitors) and is degraded by lenalidomide/IMiDs. Similarly, EOMES and MEF2B feed into pathways with HDAC inhibitor targets.

Section 15: Drug Repurposing Opportunities

TOP 30 Repurposing Candidates

Priority scoring based on: (1) GWAS p-value strength, (2) Mendelian evidence, (3) Druggable protein family, (4) Expression in B cells/lymphoid tissue, (5) Known safety profile, (6) Existing clinical use in hematology.

Section 16: Druggability Pyramid

Section 17: Undrugged Target Profiles

TOP High-Value Undrugged Targets

1. IRF4 (Interferon Regulatory Factor 4)

• GWAS p-value: 2e-10 (replicated in 4+ studies)
• Variant type: Intergenic regulatory (6p25.3)
• Protein function: Master TF controlling B-cell differentiation, plasma cell development; essential for germinal center exit
• Family: Transcription factor (DIFFICULT)
• Structure: 18 PDB structures; DBD solved; AlphaFold pLDDT 72.48
• Expression: VERY HIGH in germinal center B cells — directly relevant
• Interactions: Interacts with BCL2 pathway, PU.1, BATF
• Why undrugged? Transcription factor — no enzymatic pocket; however, lenalidomide degrades IRF4 via cereblon — validated indirect approach
• Druggability potential: HIGH (via degraders/IMiDs; cereblon E3 ligase modulators)

2. EOMES (Eomesodermin)

• GWAS p-value: 5e-13 (very strong)
• Variant type: Regulatory (3p24.1)
• Protein function: T-box TF controlling T-cell and NK cell effector function; tumor immune surveillance
• Family: Transcription factor (DIFFICULT)
• Structure: AlphaFold only (pLDDT 56.74 — low quality)
• Expression: T/NK cell-specific — relevant for immune microenvironment
• Interactions: Controls granzyme B, perforin expression
• Why undrugged? Novel TF target; no small molecule approaches
• Druggability potential: LOW (no structural pocket; immunotherapy approaches may modulate function)

3. MEF2B (Myocyte Enhancer Factor 2B)

• GWAS p-value: 3e-08
• Variant type: Intronic (19p13.11)
• Protein function: MADS-box TF; controls germinal center gene programs; recurrently mutated in DLBCL and FL
• Family: Transcription factor (DIFFICULT)
• Structure: 7 PDB structures; MADS domain solved; AlphaFold pLDDT 62.25
• Expression: Germinal center B-cell-specific — DIRECTLY relevant
• Interactions: Recruits HDACs (drugged: romidepsin, belinostat)
• Why undrugged? TF; but HDAC inhibitors modulate MEF2B-controlled gene programs
• Druggability potential: MEDIUM (indirect via HDAC inhibitors)

4. SP140 (Nuclear Body Protein SP140)

• GWAS p-value: 1e-08
• Variant type: Intronic (2q37.1)
• Protein function: Bromodomain + PHD finger protein; epigenetic reader in immune cells
• Family: Bromodomain (MODERATE — targetable)
• Structure: 5 PDB (bromodomain solved); AlphaFold pLDDT 58.02
• Expression: Immune-specific (B cells, macrophages)
• Interactions: Chromatin regulation; near BET protein family
• Why undrugged? Emerging target; bromodomain is druggable class
• Druggability potential: HIGH (bromodomain inhibitors are feasible; tool compounds exist)

5. CXCR5 (C-X-C Chemokine Receptor Type 5)

• GWAS p-value: 2e-11
• Variant type: Regulatory (11q23.3)
• Protein function: GPCR; B-cell homing to follicles; B-cell lymphoma pathogenesis
• Family: GPCR (HIGHLY DRUGGABLE)
• Structure: No PDB; AlphaFold available; homology to solved GPCRs
• Expression: B cells, T follicular helper cells — VERY relevant
• Interactions: CXCL13 ligand; G alpha(i) signaling
• Why undrugged? Lack of therapeutic focus; emerging interest
• Druggability potential: VERY HIGH (GPCR — highly druggable family; known to regulate Burkitt lymphoma)

6. IRF8 (Interferon Regulatory Factor 8)

• GWAS p-value: 1e-08
• Variant type: Regulatory (16q24.1)
• Protein function: TF controlling dendritic cell and B-cell development
• Family: Transcription factor (DIFFICULT)
• Structure: AlphaFold only (pLDDT 75.54)
• Expression: Dendritic cells, B cells — relevant
• Interactions: IRF4 family member; JAK/STAT pathway downstream
• Why undrugged? TF; but upstream JAK/STAT pathway is druggable
• Druggability potential: LOW-MEDIUM (indirect via JAK inhibitors)

7. CD69 (Early Activation Antigen CD69)

• GWAS p-value: 1e-06 (suggestive)
• Variant type: Intergenic
• Protein function: C-type lectin receptor; early lymphocyte activation; modulates S1PR1
• Family: C-type lectin (MODERATE)
• Structure: 6 PDB structures (1.37Å resolution available)
• Expression: Activated lymphocytes — relevant
• Interactions: Binds S1PR1 (drugable by fingolimod)
• Why undrugged? Novel; biology emerging; regulates tissue residency
• Druggability potential: MEDIUM (surface protein; antibody approaches feasible)

8. QPCT (Glutaminyl-Peptide Cyclotransferase)

• GWAS p-value: 4e-08
• Variant type: Intronic (2p22.2)
• Protein function: Metalloenzyme; post-translational cyclization of glutamine; modulates CCL2 (monocyte chemoattractant)
• Family: Enzyme/Transferase (HIGHLY DRUGGABLE)
• Structure: 40+ PDB structures with inhibitor co-crystals; AlphaFold pLDDT 92.92
• Expression: Moderate in immune cells
• Interactions: Modifies CCL2 chemokine; immune cell recruitment
• Why undrugged for NHL? Focus has been on Alzheimer’s disease; inhibitors exist (PBD-150, SEN177)
• Druggability potential: VERY HIGH (established enzyme target with clinical-stage inhibitors for other indications)

9. PHLPP1 (PH Domain Leucine-Rich Repeat Phosphatase 1)

• GWAS p-value: 2e-12 (at BCL2 locus, PHLPP1 nearby)
• Protein function: Protein phosphatase; dephosphorylates AKT; tumor suppressor
• Family: Phosphatase (DRUGGABLE)
• Why undrugged? Tumor suppressor — activation rather than inhibition needed
• Druggability potential: LOW (activation of phosphatases is harder than inhibition)

10. GRAMD1B (GRAM Domain Containing 1B)

• GWAS p-value: 8e-13 (strong)
• Variant type: Intronic (11q24.1)
• Protein function: Cholesterol transfer protein; lipid metabolism at ER-plasma membrane junctions
• Family: Unknown drug target class
• Structure: AlphaFold only
• Expression: Lymphocytes, broad
• Why undrugged? Function not well understood in lymphoma
• Druggability potential: LOW (novel biology; no pocket identified)

Ranked Undrugged Opportunities

Section 18: Summary

GWAS LANDSCAPE

• Total associations: 92 across 28 studies
• Unique protein-coding genes: ~25 from GWAS + 8 from Mendelian = 31 total
• Coding vs non-coding variants: 0% coding / 100% non-coding (all regulatory/intronic/intergenic)

GENETIC EVIDENCE

• Mendelian overlap genes: 8 (PRF1, CASP10, BRAF, TP53, B2M, RAD54L, RAD54B, BCL10)
• No gene has BOTH genome-wide significant GWAS + Mendelian evidence directly, but BCL2 (GWAS p=2e-12) is functionally linked to the BCL2L11/PRF1 apoptosis axis

DRUGGABILITY

• Overall druggable rate: 32% (10/31 genes in druggable protein families)
• Approved drugs: 23% (7/31)
• In clinical trials: 10% (3/31)
• Opportunity gap: 52% (16/31 with NO drug development)

PYRAMID SUMMARY

CLINICAL TRIAL ALIGNMENT

• ~23% of top NHL trial drugs target GWAS genes — moderate alignment
• Notable disconnect: Most NHL drugs target CD20, BTK, PI3K — genes NOT in GWAS

TOP 10 REPURPOSING CANDIDATES

TOP 10 UNDRUGGED OPPORTUNITIES

TOP 10 INDIRECT OPPORTUNITIES

KEY INSIGHTS

1. HLA dominance in GWAS: The strongest NHL GWAS signals cluster at 6p21 (HLA region), confirming NHL as fundamentally an immune-mediated malignancy. This has treatment implications for immune checkpoint and T-cell-based therapies.

2. BCL2 as validated genetic target: BCL2 is the only gene with BOTH strong GWAS evidence (p=2e-12) AND an approved drug (venetoclax). It serves as proof-of-concept that GWAS-to-drug translation works in NHL.

3. IRF4/cereblon axis is genetically validated: IRF4 has robust GWAS signal (p=2e-10, replicated 4+ times) and is the key effector of lenalidomide/IMiD action. This genetically validates the entire IMiD drug class for NHL, particularly ABC-DLBCL.

4. CXCR5 is the top undrugged opportunity: A GPCR (most druggable protein family) with strong GWAS evidence (p=2e-11), B-cell-specific expression, and known role in lymphoma pathogenesis. No approved drugs or advanced clinical candidates exist.

5. QPCT is a “hidden” druggable target: An enzyme with excellent structural data (40+ PDB co-crystals with inhibitors) and GWAS evidence for NHL. Inhibitors developed for Alzheimer’s disease could be repurposed.

6. SP140 bromodomain is an emerging epigenetic target: Unlike other TF GWAS genes, SP140 contains a druggable bromodomain. This positions it uniquely among the difficult-to-drug transcription factor class.

7. Most NHL drugs DON’T target GWAS genes: Only ~23% of trial drugs target GWAS-implicated genes. CD20, BTK, and PI3K — the main NHL drug targets — are absent from GWAS. This disconnect suggests current therapies may not address the root genetic biology.

8. Mendelian genes provide orthogonal validation: BRAF, TP53, PRF1, and CASP10 implicate the MAPK, p53, and immune cytotoxicity pathways — all with existing drug modalities.

9. Compared to other hematologic malignancies: NHL has a higher proportion of immune regulation genes (HLA, CD86, CD40, CXCR5) vs. CML/AML (dominated by kinases), reflecting its B-cell origin and microenvironment dependence.

10. 48% of GWAS genes are “hard targets” (Level 6) — mostly transcription factors. Emerging technologies (PROTACs, molecular glues, degraders) may unlock these in the future.

Analysis generated using biobtree MCP tools querying GWAS Catalog, ClinVar, GenCC, OMIM, Orphanet, MeSH, UniProt, InterPro, PDB, AlphaFold, ChEMBL, STRING, BioGRID, Reactome, PharmGKB, and Clinical Trials data. All database identifiers verified through cross-referencing.