I've pulled the two main RELB papers (Le Voyer 2024, Sharfe 2022). I'm now checking the PMID of the original 2015 RELB report before I write the report.

Immunodeficiency 53 (IMD53; RELB deficiency): research report

Sourcing. I could not open OMIM, Orphanet, ClinGen or GenCC directly, so those statements come from search snippets or aggregators. Two papers I read in full text through PMC (Le Voyer 2024 and Sharfe 2022), except the last ~39k characters of the Le Voyer page. Anything not in a paper I read is marked unverified.

• Do not use any PMID marked unverified as evidence until it has been fetched. The repository's just fetch-reference check applies.
• Every CURIE below is a suggestion from memory. None was looked up in OAK or the term caches. Per CLAUDE.md, each one must be checked before it is bound.

1. Disease information

Overview. IMD53 is a rare autosomal recessive combined T- and B-cell immunodeficiency caused by biallelic loss-of-function or hypomorphic variants in RELB. RelB is a subunit of the alternative (non-canonical) NF-κB pathway. The syndrome combines recurrent viral, bacterial and fungal infections with immune decline over time. Autoimmunity and neutralizing anti-type I interferon autoantibodies are also seen.

Identifiers.

In this repository the gene would be written hgnc:9956 (lowercase prefix).

Synonyms. RELB deficiency, RelB deficiency, and combined immunodeficiency due to RELB deficiency.

Data origin. The evidence comes from individual-patient case reports and small series (3 + 3 + 2 patients), not from registries or EHR data.

2. Etiology

• Cause. Biallelic RELB variants, either homozygous or compound heterozygous, inherited from carrier parents. No environmental cause is described.
• Genetic risk and modifiers. None are described. gnomAD homozygous missense variants D417N and D487Y behaved like wild type in the assay (Le Voyer 2024, PMID 39231201).
• Protective factors. None are known.
• Gene–environment interaction. Not described. The patients' infections are of ordinary community pathogens, plus JC polyomavirus and Cryptococcus, as a consequence of the immunodeficiency.
• Strength of gene–disease evidence. This is graded unevenly.
• ClinGen's SCID-CID group rates RELB "limited", per a search snippet.
• PanelApp lists it as high evidence, per a search snippet.
• OMIM marks the phenotype with "?".
• Curators should record the validity classification only as classified_by an external source, never assign it themselves.

3. Phenotypes

The frequencies below are qualitative or n/N from at most 8 patients. HPO suggestions are unverified.

Onset and course. - Presentation is in infancy to early childhood, per MalaCards (secondary source). - The 2015 index family presented between 7 days and 4 months of age, per search-result summaries (unverified). - Immune function declines over time. The siblings showed age-related loss of naïve T cells and B cells. - Some adults escape early diagnosis. P1 and P2 were diagnosed as adults.

Quality of life. No formal measures are reported. P2 died of progressive multifocal leukoencephalopathy at age 34.

4. Genetic and molecular information

Variants.

• The Le Voyer variants are absent from gnomAD and other databases.
• The same P364L allele appears in two independent reports, once homozygous and once in trans with E145K.
• All variants are germline.
• There is no known modifier gene and no epigenetic or chromosomal data.

Mechanism type. Loss of function (complete) or hypomorphism (partial). Use functional_impact_category: LOSS_OF_FUNCTION or PARTIAL_LOSS_OF_FUNCTION per variant, and cite the functional assay.

Primary literature for the original report. - Merico et al. 2015, LymphoSign Journal, DOI 10.14785/lpsn-2015-0005. It is not in PubMed, so it cannot go through fetch-reference as a PMID. A DOI fetch may work. - Sharfe N et al. 2015, "The effects of RelB deficiency on lymphocyte development and function", PMID 26385063. This PMID comes from a GenCC-derived search summary and is unverified. Fetch it before use.

5. Environmental information

None are known. - Infectious agents as consequences, not causes: Candida, Cryptococcus neoformans, varicella-zoster virus, JC polyomavirus and human papillomaviruses (epidermodysplasia verruciformis-type warts). These are opportunistic outcomes, so they belong under phenotypes or complications, not infectious_agent. - If a curator wants NCBITaxon terms, look each one up individually.

6. Mechanism and pathophysiology

Causal chain.

1. Biallelic RELB variants lead to absent or hypomorphic RelB protein. This is demonstrated by loss of expression for Q72fs and by reduced stability for P364L.
2. Loss of RelB impairs lymphotoxin- and TWEAK-induced alternative NF-κB signaling, shown as reduced NFKB2 and VCAM1 induction in patient fibroblasts. Canonical NF-κB responses to TNF and IL-1β stay intact (Le Voyer 2024).
3. In the thymus, loss of RelB in stromal cells is thought to give hypoplastic thymus and few AIRE-expressing medullary thymic epithelial cells. The thymic hypoplasia in P1 is observed. The AIRE-deficiency link is an inference from the 2022 Le Voyer study in the search results, which I did not read (unverified).
4. Defective stromal-driven central tolerance leads to anti-type I IFN autoantibodies. This is observed in both adults, and the antibodies persist after HSCT, which supports a stromal (not hematopoietic) origin. That reading is the authors' inference.
5. Naïve T-cell depletion, low Th17 cells and reduced IL-17A/F/IL-22 production, together with poor memory and class-switched B-cell output, lead to combined immunodeficiency. For B cells, P1's naïve cells could not make IgG or IgA in response to CD40L plus IL-21 in vitro.
6. Combined immunodeficiency leads to chronic mucocutaneous candidiasis, bacterial respiratory infections, cryptococcal meningitis, viral skin disease and PML.
7. Branch for partial deficiency (P364L). In Sharfe 2022, fibroblasts showed elevated pro-inflammatory NF-κB target genes after poly(I:C) or LIGHT stimulation. The authors interpret this as loss of RelB's inhibitory effect on RelA, and they describe their evidence as indirect. This is inferred, not demonstrated. Reduced RelB then leads to autoimmune enteritis, hepatitis-like liver disease, bronchiolitis and skin inflammation, with a decline in immune cells in the second decade.

Suggested GO and CL terms (unverified). - GO: non-canonical NF-κB signal transduction (GO:0038061), positive regulation of non-canonical NF-κB signal transduction (GO:1901224), T cell differentiation in thymus (GO:0033077), B cell differentiation (GO:0030183), central tolerance induction (GO:0002507), type 17 immune response (GO:0072538). - CL: naïve thymus-derived CD4-positive alpha-beta T cell (CL:0000895), memory B cell (CL:0000787), medullary thymic epithelial cell (CL:0002293), fibroblast (CL:0000057).

Not reported. No transcriptomic, proteomic, metabolomic, single-cell or screen data were found.

7. Anatomical structures affected

• Primary: thymus (hypoplastic in P1; UBERON:0002370), lymphoid tissue and blood.
• Secondary: upper and lower airways, gut, liver and biliary tree, skin, CNS (cryptococcal meningitis, PML) and thyroid.
• Cells: naïve and effector memory T cells, Th17 cells, B cells, medullary thymic epithelial cells and fibroblasts.
• Subcellular: cytoplasm to nucleus translocation of the RelB/p52 dimer.
• Lateralization: not applicable; the disease is systemic.

8. Temporal development

• Onset: infancy to childhood in the complete-deficiency families. Partial deficiency can present later, and the two adults in Le Voyer 2024 were diagnosed in adulthood.
• Progression: progressive immune decline and escalating organ autoimmunity in the second decade (P364L).
• Duration: lifelong without definitive therapy.
• Remission: no spontaneous remission is described.
• Critical period: early HSCT in infancy is associated with survival in the complete-deficiency children (search summary; unverified).

9. Inheritance and population

• Inheritance: autosomal recessive (HP:0000007). Consanguinity is documented in all three kindreds with homozygous variants.
• Prevalence and incidence: not documented (prevalence_class: NOT_YET_DOCUMENTED). At most 8 patients have been published.
• Populations: Turkish, Iranian Jewish, and (per the 2015 report) other consanguineous families.
• Penetrance, anticipation, mosaicism, carrier frequency: no data. Heterozygous parents were healthy in the families reported.
• Founder effect: P364L has been seen in two unrelated reports. This is a lead, not evidence of a founder effect.
• Sex ratio: both sexes are affected.

10. Diagnostics

• Immunophenotyping:
• Low naïve CD4⁺ and CD8⁺ T cells.
• Raised CD4/CD8 ratio, per MalaCards and the aggregator snippets.
• Reduced TREC levels, per MalaCards (secondary).
• Low memory B cells, hypogammaglobulinemia and poor vaccine responses.
• Reduced PHA and anti-CD3 proliferation (Sharfe 2022).
• Functional assays:
• Fibroblast stimulation with lymphotoxin or TWEAK, reading out NFKB2 and VCAM1.
• Western blot for RelB protein.
• Autoantibodies: neutralizing anti-type I IFN antibody testing. These antibodies appeared about 5 years after HSCT in P1.
• Genetic testing:
• WES or WGS, or an inborn-errors-of-immunity panel that includes RELB.
• Sanger confirmation.
• Segregation analysis in consanguineous families.
• Differential diagnosis: other combined immunodeficiencies (NIK/MAP3K14 deficiency, which shares the alternative NF-κB pathway; NFKB2 haploinsufficiency; APS-1/AIRE deficiency; CD40 ligand deficiency).
• Screening: reduced TRECs might be detectable on newborn screening. No program is established.

11. Outcome and prognosis

• Survival data are limited to individual cases.
• P2 died of progressive multifocal leukoencephalopathy at age 34.
• P1 was alive 3 years post-HSCT with normal immunoglobulins.
• Complications are CNS infection, lymphoma (P2), liver disease and lung damage requiring lobectomy.
• Prognostic factors are unknown. Complete deficiency presents earlier and more severely.
• No quality-of-life instruments have been applied.

12. Treatment

• Definitive: allogeneic HSCT.
• P1 had HSCT at age 14 from a fully matched sibling, with successful engraftment (Le Voyer 2024).
• At 6 months, most T-cell cytokine defects and naïve B-cell defects had corrected. IL-17A/F production stayed low.
• At 3 years, B-cell proportions and immunoglobulin levels were normal, IVIg had been stopped, and no fungal infection occurred.
• Anti-IFN autoantibodies persisted.
• NCIT suggestion: hematopoietic cell transplantation (NCIT:C15431), therapeutic_modality: CELL_THERAPY.
• Supportive: immunoglobulin replacement (used in Sharfe 2022 siblings and in P1 before HSCT), antimicrobial prophylaxis or treatment, and rituximab (given to P2).
• NCIT suggestion: Pharmacotherapy (NCIT:C15986). Bind specific agents as therapeutic_agent only after a CHEBI or NCIT lookup.
• Surgical: lobectomy for chronic bronchiolitis (Sharfe 2022).
• Not found: gene therapy, targeted therapy and clinical trials. I did not search ClinicalTrials.gov, so absence is unconfirmed.
• Whether HSCT stops autoimmunity of stromal origin is unknown, because the autoantibodies persisted.

13. Prevention

• Genetic counseling for families, with carrier testing of relatives and prenatal or preimplantation diagnosis in known families.
• Early diagnosis permits prophylaxis and early HSCT.
• No vaccination guidance specific to RELB deficiency was found. Live vaccines are usually avoided in combined immunodeficiency, but I did not find a source stating this for IMD53.

14. Other species and natural disease

No naturally occurring animal disease is documented (OMIA not checked). Relb is conserved across mammals. The mouse ortholog is Relb (MGI).

15. Model organisms

• Relb knockout mice die at 2–6 weeks from severe, T-cell-dependent, multi-organ inflammation (Le Voyer 2024 discussion).
• Human patients show a narrower picture: mainly infection, IFN autoantibodies and, in partial deficiency, organ autoimmunity.
• Fidelity is therefore partial (PARTIALLY_RECAPITULATES, fidelity: MODERATE).
• Limitation: species difference in the severity and timing of autoimmunity.
• Conditional mouse models: a Relb knockout restricted to medullary thymic epithelial cells gives T-cell infiltration of liver, lung and skin. This resembles the organ pattern of the partial-deficiency patients and supports an epithelial-intrinsic role in central tolerance (Le Voyer 2024 discussion). Use evidence_source: MODEL_ORGANISM.
• In vitro: patient fibroblasts, T cells and B cells (IN_VITRO).
• No zebrafish, Drosophila or iPSC models were found. I searched only briefly.

Curation cautions

• PMIDs read in full text: 39231201 (Le Voyer 2024, PNAS) and 36402602 (Sharfe 2022, J Autoimmun). Page 1 of the Le Voyer text covers only the first 100,000 of 139,108 characters.
• Snippets are not verified. WebFetch returned summaries because of its 125-character quote limit. Do not paste any wording from this report as a snippet. Run just fetch-reference PMID:39231201 and PMID:36402602, then copy the exact sentences from the cache files.
• Fetch before use: PMID 26385063 and the LymphoSign DOI.
• Not read: the 2022 Le Voyer study (autoantibodies in NIK and RELB deficiency, AIRE-expressing thymic cells) was cited only from search-result snippets.
• Retrieval issue: the PubMed fetch for 36402602 failed with a proxy error. I used the PMC copy instead.
• Run just preflight-dr on this report before using its content, per CLAUDE.md.
• Record RELB gene validity only by copying an external classification (ClinGen via CGGV: or GenCC), never by assigning one.

Sources - Le Voyer et al., PNAS 2024 (PMC11406260) - Sharfe et al., J Autoimmun 2022 (PMC10861023) - MalaCards: Immunodeficiency 53 - GenCC submission for RELB / MONDO:0054696 - Disease Ontology DOID:0111992 - DrugMap disease record - MedGen C4539811 mirror - GenCC SGC-106323.2

Reference Validation

Checked with linkml-reference-validator 0.3.0.

All extracted references resolved successfully.

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

• MONDO:0054696 (2 mentions) - the report calls it "MONDO"; MONDO calls it immunodeficiency 53
• DOID:0111992 (5 mentions) - the report calls it "DOID"; DOID calls it immunodeficiency 53
• HP:0100550 (1 mention) - the report calls it "P1 and P2"; HP calls it Tendon rupture
• HP:0002014 (1 mention) - the report calls it "Siblings"; HP calls it Diarrhea
• HP:0004322 (1 mention) - the report calls it "Oldest sibling (Sharfe 2022)"; HP calls it Short stature

Unresolved terms

These identifiers do not exist in an ontology that resolved other terms from the same prefix, so they were most likely invented:

• HP:0003290 (1 mention), reported as "Sharfe 2022 siblings" - HP does not contain this term

Obsolete terms

These terms are real but deprecated. Citing one is not a fabrication; it does mean the report is naming something the ontology has retired:

• HP:0005407 (obsolete Decreased proportion of CD4-positive helper T cells) (1 mention) - replaced by HP:0032218

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

• GO:0038061 (1 mention) - the report calls it "GO: non-canonical NF-κB signal transduction"; GO calls it non-canonical NF-kappaB signal transduction
• CL:0000895 (1 mention) - the report calls it "CL: naïve thymus-derived CD4-positive alpha-beta T cell"; CL calls it naive thymus-derived CD4-positive, alpha-beta T cell
• NCIT:C15431 (1 mention) - the report calls it "NCIT suggestion: hematopoietic cell transplantation"; NCIT calls it Hematopoietic Cell Transplantation
• NCIT:C15986 (1 mention) - the report calls it "NCIT suggestion: Pharmacotherapy"; NCIT calls it Pharmacotherapy