FILS Syndrome

1. Disease Information

Overview. FILS syndrome is an ultra-rare autosomal recessive multisystem disorder caused by biallelic hypomorphic mutations in POLE, which encodes the catalytic subunit of DNA polymerase epsilon (Pol ε), the principal leading-strand replicase. It was first delineated in 2012 in a large consanguineous French/Algerian kindred, and as of the most recent literature review (through 2022) only ~16–20 patients from a handful of families worldwide have been reported, making it one of the rarest recognized inborn errors of DNA replication with combined immunologic and growth phenotypes (Frontiers Pediatrics 2022; PMC9441657).

Key identifiers: - OMIM: #615139 (phenotype) — "FACIAL DYSMORPHISM, IMMUNODEFICIENCY, LIVEDO, AND SHORT STATURE; FILS" (OMIM #615139) - Gene OMIM: *174762 POLE (OMIM *174762) - Orphanet: ORPHA:352712 (Orphanet) — prevalence documented as <1/1,000,000, onset in infancy/neonatal period - HGNC: HGNC:9177 (POLE; historically also called POLE1) - MedGen Concept ID: C3554576 - Locus: 12q24.33 (also written 12q24.1 in older nomenclature) - Related/overlapping entries: OMIM #614732 (IMAGe syndrome, CDKN1C-related) and OMIM #618336 (IMAGe syndrome with immunodeficiency, "IMAGEI," the POLE-related digenic-haplotype allelic subtype)

Synonyms: Polymerase epsilon 1 (POLE1) deficiency; Facial dysmorphism-immunodeficiency-livedo-short stature syndrome; occasionally grouped in the literature as part of the broader "POLE-related replisome disorders" spectrum.

Data provenance: All existing knowledge derives from aggregated case reports/case series (individual, deeply phenotyped patients) rather than large cohort/EHR data, given the extreme rarity (literature search to March 2022 found only 3 publications totaling 16 patients) (PMC9441657).

2. Etiology

Disease causal factor — genetic (monogenic, autosomal recessive). FILS is caused by homozygous or compound-heterozygous hypomorphic (partial loss-of-function) variants in POLE leading to reduced but not absent cellular Pol ε levels/activity.

Founding molecular lesion (index French/Algerian family): A homozygous A→G transition in intron 34 causes skipping of exon 34, a frameshift, and premature termination at residue 1561, producing a truncated protein lacking the C-terminus. Patient T cells showed two transcript species — wild-type (~10%) and the exon-34-skipped mutant (~90%) — consistent with a leaky, hypomorphic allele rather than a null allele (a complete null is presumed embryonic lethal, paralleling Pole knockout lethality in mice) (OMIM #615139; JEM 2012, PMID:23230001).

Subsequently reported variants: - A recurrent hypomorphic splice-altering intronic variant, c.1686+32C>G, found on a shared haplotype in combination with different loss-of-function variants in trans across 15 individuals from 12 families with the POLE-linked IMAGe-like phenotype — establishing digenic-like compound heterozygosity (one recurrent hypomorphic allele + one severe LOF allele) as a recurring mechanism (Logan et al. 2018, AJHG, PMID:30503519). - Chinese patient: compound heterozygous c.5811+2T>C (splicing, maternal, causing exon 42 skipping) and c.2006G>A (nonsense, paternal, p.W669X, truncating within the DNA polymerase type-B catalytic domain) (PMC9441657). - A homozygous missense variant c.100C>T (p.Arg34Cys) reported in a child presenting with poikiloderma, expanding the dermatologic spectrum beyond livedo alone.

Risk factors: - Genetic: Biallelic POLE hypomorphic variants are necessary and sufficient; consanguinity substantially raises risk in affected families (the founding family was consanguineous). No modifier genes have yet been identified. - Environmental: None established; this is a purely monogenic disorder of DNA replication machinery, not modulated by known environmental/lifestyle exposures.

Protective factors: None reported. Retention of ~10% residual wild-type transcript/enzyme activity in the founding family is thought to be compatible with survival — i.e., allelic "leakiness" itself is protective against embryonic lethality, but this is an allele property rather than an independent protective factor.

Gene–environment interactions: Not applicable/not reported; no data on environmental modifiers of expressivity.

3. Phenotypes

FILS syndrome's four defining phenotype domains, each present with high frequency but variable severity across reported patients:

Severity/progression: Highly variable between patients — even within the same allelic class. Growth impairment is progressive from early childhood; immunodeficiency and infection susceptibility can attenuate with age in milder cases (the Chinese patient's infections resolved by age 4) but can be fatal in infancy in more severely affected siblings (a reported elder brother died at 50 days of age, likely from pneumonia, with more severe IUGR) (PMC9441657).

Quality of life impact: Not systematically studied (no EQ-5D/SF-36 data identified); qualitatively, recurrent infection burden and short stature affect early childhood morbidity, but at least one reported 8-year-old had age-appropriate academic performance with only mild motor-milestone delay.

4. Genetic/Molecular Information

• Causal gene: POLE (DNA polymerase epsilon, catalytic subunit A; HGNC:9177; historically POLE1), OMIM *174762, chromosome 12q24.33.
• Protein: Catalytic (largest) subunit of the four-subunit Pol ε holoenzyme (POLE/POLE2/POLE3/POLE4); contains an N-terminal polymerase domain and a 3′→5′ proofreading exonuclease domain (residues ~223–517); primary leading-strand replicase of the eukaryotic replisome.
• Variant classes in FILS: Hypomorphic — intronic splice-altering variants causing partial exon skipping (e.g., intron 34 A>G; c.5811+2T>C; recurrent c.1686+32C>G), and nonsense/truncating variants in trans (e.g., c.2006G>A/p.W669X). None are complete nulls; complete loss is presumed lethal.
• Zygosity: Homozygous in the consanguineous founding family; compound heterozygous in most subsequently reported unrelated patients.
• Functional consequence: Partial loss-of-function/reduced Pol ε dosage — a quantitative, not purely qualitative, defect. Retained wild-type transcript fraction (~10% in the index family) appears essential for viability.
• Modifier genes: None established.
• Allelic spectrum (differential diagnosis within POLE):
• POLE-linked IMAGe syndrome with immunodeficiency (IMAGEI, OMIM #618336): biallelic — one recurrent hypomorphic splice variant (c.1686+32C>G) in trans with a distinct LOF variant; clinically overlaps with classic CDKN1C-related IMAGe (IUGR, metaphyseal dysplasia, adrenal hypoplasia congenita, genital anomalies) plus variable immunodeficiency (Logan et al. 2018, PMID:30503519).
• Polymerase proofreading-associated polyposis (PPAP): heterozygous germline exonuclease-domain missense variants (e.g., p.Leu424Val, p.Pro286Arg) causing autosomal dominant colorectal adenomatous polyposis/cancer predisposition via a hypermutator mechanism — mechanistically and inheritance-wise distinct from FILS.
• Constitutional POLE variants causing a CMMRD-like phenotype: heterozygous, generally de novo, stronger "mutator" exonuclease variants causing early-onset multi-cancer/café-au-lait/pilomatricoma phenotype resembling constitutional mismatch repair deficiency (Sehested et al. 2022, Human Mutation, PMID:34816535; PMC5243902).
• Population frequency: Individual causal alleles are absent or present only as rare heterozygotes in gnomAD; no founder-effect population enrichment has been reported outside the shared IMAGEI haplotype among unrelated IMAGe-immunodeficiency families.
• Epigenetics: No disease-specific DNA methylation/chromatin studies identified for FILS specifically.
• Chromosomal abnormalities: None — this is a single-gene, sequence-level disorder, not a copy-number/structural disorder.

5. Environmental Information

No environmental, lifestyle, dietary, or infectious agents have been implicated as causal or risk-modifying factors for FILS syndrome. Infections (respiratory pathogens, meningitis-causing organisms) are a consequence of the immunodeficiency rather than a cause of the disease.

6. Mechanism / Pathophysiology

Molecular pathway. POLE encodes the catalytic subunit of the Pol ε holoenzyme, which — together with the CMG helicase (CDC45–MCM2-7–GINS) and Pol δ — forms the core eukaryotic replisome, with Pol ε as the dedicated leading-strand polymerase and its exonuclease domain performing proofreading during S-phase DNA synthesis.

Cellular process disrupted. Reduced cellular Pol ε abundance (from hypomorphic biallelic variants) causes delayed/impaired G1-to-S phase transition and cell-cycle progression, rather than a mutator phenotype per se. This was directly demonstrated in patient-derived T lymphocytes, B lymphocytes, chondrocytes, and osteoblasts, all of which showed impaired proliferation and delayed S-phase entry (JEM 2012, PMID:23230001; Logan et al. 2018, PMID:30503519).

Causal chain (proposed): 1. Biallelic hypomorphic POLE variants → reduced Pol ε protein/holoenzyme dosage or reduced full-length transcript. 2. Slowed leading-strand replication and replication stress → delayed G1–S progression in proliferating cell compartments. 3. In lymphocytes (T and B cells): impaired antigen-driven clonal expansion → reduced naive/functional lymphocyte pools → variable hypogammaglobulinemia and susceptibility to recurrent bacterial respiratory infection and meningitis (immunodeficiency arm of the phenotype). 4. In chondrocytes/osteoblasts: impaired proliferation of growth-plate and bone-forming cells → intrauterine and postnatal growth restriction, short stature, and abnormal cortical bone architecture (short-stature/skeletal arm). 5. In dermal/vascular tissue: mechanism of livedo/poikiloderma is less well characterized mechanistically but is presumed to reflect replication-dependent effects on cutaneous microvasculature and/or keratinocyte turnover; not fully elucidated at the cellular level in the literature reviewed. 6. Craniofacial dysmorphism (malar hypoplasia, forehead prominence) likely reflects impaired proliferation of neural-crest-derived facial skeletal precursors during a critical embryonic window, analogous to other "ribosomopathy"/replisome-disorder craniofacial phenotypes, though this has not been directly mechanistically tested in FILS.

Upstream vs. downstream: The Pol ε dosage deficit is the singular upstream molecular lesion; all four clinical domains (facial dysmorphism, immunodeficiency, livedo, short stature) are proposed to be parallel downstream consequences of tissue-specific sensitivity to reduced replicative capacity in rapidly dividing cell populations (lymphocytes, chondrocytes, osteoblasts, craniofacial mesenchyme) during development and ongoing immune responses.

Suggested GO terms: GO:0006261 (DNA-templated DNA replication), GO:0000082 (G1/S transition of mitotic cell cycle), GO:0006974 (DNA damage response), GO:0045005 (DNA-templated DNA replication maintenance of fidelity). Suggested CL terms: CL:0000084 (T cell), CL:0000236 (B cell), CL:0000138 (chondrocyte), CL:0000062 (osteoblast).

Note on allelic mechanism divergence: In contrast to FILS (quantitative Pol ε insufficiency, cell-cycle delay), the heterozygous exonuclease-domain PPAP/CMMRD-like variants act through a qualitative gain of a hypermutator function (loss of proofreading fidelity → genome-wide hypermutation → cancer), a mechanistically distinct process from the FILS growth/immune phenotype despite being in the same gene.

7. Anatomical Structures Affected

• Organ/system level: Craniofacial skeleton (malar/zygomatic hypoplasia, frontal bone), skin/cutaneous vasculature (livedo, poikiloderma), immune system (lymphoid compartments — respiratory tract as secondary infection site, meninges), skeletal system (long bones, growth plates), and in the overlapping IMAGe-immunodeficiency subtype, the adrenal cortex and genitourinary system.
• Tissue/cell level: Lymphocytes (T cells, B cells), chondrocytes (growth plate cartilage), osteoblasts (cortical/trabecular bone formation), dermal microvasculature/keratinocytes (livedo/poikiloderma), craniofacial mesenchyme/neural crest derivatives.
• Subcellular level: Nucleus — specifically the replisome/replication fork (Pol ε acts at the leading-strand replication fork during S-phase); relevant GO Cellular Component: GO:0043625 (delta DNA polymerase complex, analogous), GO:0045142 (triplex DNA binding — not directly relevant), most precisely GO:0008622 (epsilon DNA polymerase complex).
• Anatomical localization (UBERON suggestions): UBERON:0001707 (nasal cartilage/malar region — approximate), UBERON:0002385 (muscle/facial structures), UBERON:0002316 (bone marrow — lymphoid), UBERON:0000178 (blood), UBERON:0002370 (thymus, T-cell development), UBERON:0001007 (digestive/skin — for livedo distribution on cheeks/forearms/legs).
• Laterality: Bilateral/symmetric for facial dysmorphism and livedo distribution as reported.

8. Temporal Development

• Onset: Congenital/prenatal for facial dysmorphism, livedo, and IUGR; infancy/early childhood for recurrent infections and progressive short stature. Onset pattern is generally insidious/chronic rather than acute, punctuated by episodic infections.
• Progression: Growth impairment is progressive through early childhood, plateauing into variable short stature by adulthood. Immunodeficiency-related infection frequency in reported patients tends to be worst in infancy/early childhood and can improve with age in milder cases; severity is markedly variable between patients and even between siblings in the same family (one sibling pair showed markedly discordant severity, with the more severely affected sibling dying in early infancy).
• Disease course pattern: Chronic, non-remitting for the structural/growth phenotype; episodic/recurrent for the infectious complications.
• Critical periods: Prenatal and early postnatal periods appear to be the highest-risk window for mortality (recurrent pneumonia, severe IUGR); this may represent a critical period for replicative-capacity-limited tissues (immune reconstitution, skeletal growth) when demand for cell proliferation is greatest.

9. Inheritance and Population

• Inheritance pattern: Autosomal recessive (biallelic POLE hypomorphic variants).
• Epidemiology: Prevalence <1 per 1,000,000 (Orphanet); fewer than 20 molecularly confirmed patients reported in the literature as of the most recent 2022 case report/review (Orphanet ORPHA:352712; PMC9441657).
• Penetrance: Appears fully penetrant for biallelic causal genotypes, but expressivity (severity across the four core domains) is highly variable.
• Consanguinity: A significant risk factor — the index/founding family was consanguineous; however, subsequent reported cases (e.g., the Chinese patient) arose in non-consanguineous parents each carrying a distinct heterozygous variant, indicating the disease is not restricted to consanguineous pedigrees.
• Founder effects: The recurrent c.1686+32C>G hypomorphic splice variant, shared on a common haplotype across 12 unrelated IMAGe-immunodeficiency families, suggests a founder allele contributing to a meaningful fraction of the allelic (IMAGEI) subtype (Logan et al. 2018, PMID:30503519).
• Population demographics: Reported patients span European (French/Algerian), and East Asian (Chinese) ancestries; no clear geographic/ethnic restriction beyond the shared founder haplotype in the IMAGEI subgroup. No sex predilection reported for classic FILS (genital anomalies are specific to males in the overlapping IMAGe-related subtype, an expected consequence of genitourinary embryology rather than a sex-linked inheritance pattern).

10. Diagnostics

• Laboratory/immunologic workup: Lymphocyte subset immunophenotyping (T, B, NK cell counts/proportions — often normal to mildly reduced), quantitative immunoglobulins (IgG/IgG subclasses, IgA, IgM, IgE — variable, ranging from isolated IgG4 deficiency to panhypogammaglobulinemia), vaccine-response titers.
• Imaging: Skeletal radiographs showing thin long bones/thickened cortex with narrow medullary cavity; growth curve tracking (height SDS).
• Genetic testing (primary diagnostic modality): Whole-exome or whole-genome sequencing to identify biallelic POLE variants is the standard approach given genetic/allelic heterogeneity and phenotypic overlap with other syndromes; targeted POLE Sanger sequencing can confirm/segregate variants once identified. No commercial single-gene panel is highlighted as standard-of-care in the literature reviewed, reflecting the disease's extreme rarity.
• Differential diagnosis to exclude via genetic testing:
• IMAGe syndrome (CDKN1C, maternally-inherited gain-of-function) — distinguish from POLE-linked IMAGEI.
• Other combined immunodeficiency/growth-restriction "replisome disorders" (e.g., Meier-Gorlin syndrome — ORC1/ORC4/ORC6/CDT1/CDC6; Seckel syndrome).
• Other causes of congenital livedo reticularis (e.g., Adams-Oliver syndrome, STING-associated vasculopathy).
• PPAP/CMMRD-like POLE cancer syndromes (distinguished by heterozygous exonuclease-domain variants and dominant/de novo inheritance rather than biallelic hypomorphic variants).
• Screening: No population or newborn screening program exists given extreme rarity; diagnosis is case-by-case via clinical suspicion (tetrad of facial dysmorphism + immunodeficiency + livedo + short stature) followed by molecular confirmation.

11. Outcome/Prognosis

• Survival/mortality: Variable and genotype/severity-dependent. Most reported patients have survived into childhood with supportive management, but at least one reported sibling died in early infancy (~50 days) from probable pneumonia in the context of more severe IUGR, illustrating that severe presentations can be fatal in infancy.
• Morbidity: Recurrent respiratory infections and meningitis in infancy/early childhood are the principal source of acute morbidity; short stature and skeletal changes are the chronic structural morbidity.
• Developmental outcomes: At least one well-documented mildly affected patient (the Chinese case) had age-appropriate academic performance with only mild motor-milestone delay by age 8, suggesting neurocognitive development can be preserved in milder cases.
• Prognostic factors: Genotype severity (degree of residual Pol ε function), severity of early infectious complications, and possibly gestational/birth growth parameters (more severe IUGR correlated with the more severely affected/deceased sibling in one family).
• Theoretical cancer risk: Because Pol ε maintains replication fidelity, there is biological plausibility for elevated cancer risk in FILS patients, and case reports recommend long-term surveillance (skin cancer/lymphoma screening, tumor marker monitoring) as a precaution; however, a defined increased malignancy rate specific to biallelic hypomorphic FILS-type variants (as opposed to the well-established PPAP/CMMRD-like heterozygous exonuclease-domain variants) has not yet been systematically documented in the literature reviewed.

12. Treatment

No disease-specific or FDA-approved targeted therapy exists for FILS syndrome; management is entirely supportive, individualized to the immunologic and growth phenotype:

• Immunodeficiency management: Antimicrobial prophylaxis and, where indicated by significant hypogammaglobulinemia, immunoglobulin replacement therapy (a general principle for humoral immunodeficiencies; not FILS-specific data, but the standard extrapolated approach). Suggested NCIT term: NCIT:C15747 (Supportive Care); pharmacotherapy under NCIT:C15986.
• Infection surveillance/prophylaxis: Standard management of recurrent infections, prompt treatment of respiratory infections and suspected meningitis.
• Hematopoietic stem cell transplantation: Not specifically documented as performed or established for FILS syndrome in the literature identified (in contrast to other combined immunodeficiency syndromes such as ICF syndrome, where HSCT is curative); this remains an unaddressed/theoretical option rather than a reported management strategy for FILS specifically.
• Growth management: Growth hormone axis is typically reported as biochemically normal, but clinical response to growth hormone therapy appears poor/limited based on case reports — reflecting that the growth defect is a primary cell-proliferation/replicative defect rather than a GH-axis deficiency, so exogenous GH is not an established effective therapy.
• Oncologic surveillance: Case reports recommend periodic skin cancer and lymphoma screening plus tumor marker monitoring (e.g., CA-199) as a precautionary measure given the gene's role in replication fidelity, though this is expert/case-based recommendation rather than an evidence-based protocol. Suggested term: NCIT:C15343 (Cancer Screening, approximate).
• Skeletal monitoring: Serial radiographic examination of long bones to monitor for bone lesions/fragility.
• Genetic counseling: Recommended for families given autosomal recessive inheritance (25% recurrence risk for future pregnancies of carrier parents); NCIT:C15240 (Genetic Counseling).
• Experimental treatments: No clinical trials (NCT-registered) specific to FILS syndrome were identified.

13. Prevention

• Primary prevention: None beyond genetic counseling and carrier awareness in consanguineous or previously affected families; prenatal diagnosis (via targeted variant testing once a family's causal variants are known) and preimplantation genetic diagnosis are theoretically applicable AR-disorder options, though not specifically documented as used for FILS in the literature reviewed.
• Secondary prevention: Early recognition of the FILS tetrad and infection-prophylaxis measures to reduce morbidity/mortality from recurrent respiratory infections and meningitis in infancy — the period of highest risk.
• Screening: No population-level screening program exists; family-based cascade testing is the applicable model once a proband's variants are identified.
• Immunization: No FILS-specific vaccination guidance identified; general principles for immunodeficient patients (avoiding live vaccines if cellular immunity is significantly compromised, ensuring close contacts are vaccinated) would apply by extrapolation from general immunodeficiency management, not FILS-specific literature.

14. Other Species / Natural Disease

• No naturally occurring FILS-like disease has been reported in non-human species (e.g., no OMIA entry identified for a POLE hypomorphic disorder in companion animals or livestock).
• Orthologous gene: Pole (mouse ortholog; MGI:1196391), broadly conserved across eukaryotes given the essential, conserved role of Pol ε in DNA replication.
• Comparative biology: Complete germline Pole loss is embryonic lethal in mice, mirroring the inference that a complete human null allele would likely be non-viable — consistent with all reported human FILS alleles being hypomorphic/leaky rather than complete loss-of-function.

15. Model Organisms

• No dedicated mouse model of the FILS hypomorphic/reduced-dosage phenotype has been reported in the literature surveyed. The mouse Pole models that do exist target a mechanistically distinct axis:
• Proofreading-exonuclease-dead knock-in mice (e.g., D272A/E274A "Pol ε^exo-" allele, and the cancer-associated Pole^P286R knock-in) selectively abolish the 3′→5′ exonuclease proofreading activity while preserving polymerase activity, producing a hypermutator/cancer-predisposition phenotype (accelerated spontaneous tumorigenesis, elevated base-substitution mutation rates) — this models the human PPAP/CMMRD-like heterozygous exonuclease-domain disease, not the FILS growth/immunodeficiency phenotype (PNAS 2009; JCI 2018).
• Heterozygous Pole^P286R mouse fibroblasts show earlier replicative senescence without elevated DNA-damage markers — again reflecting a mutator/senescence mechanism rather than the reduced-dosage/proliferation-delay mechanism implicated in FILS.
• Cellular models used to date for FILS mechanism: Primary patient-derived cells (T lymphocytes, B lymphocytes, chondrocytes, osteoblasts) directly assayed for proliferation and cell-cycle (G1–S transition) kinetics — this is the principal "model system" evidence base for FILS pathophysiology, rather than an engineered animal or iPSC model (JEM 2012, PMID:23230001).
• Gap: No FILS-specific hypomorphic knock-in mouse or iPSC-derived model reproducing the reduced-dosage/G1–S-delay mechanism (as opposed to the loss-of-proofreading/hypermutator mechanism) was identified in this search — an important human-model-mismatch caveat for any dismech knowledge base entry (HUMAN_MODEL_MISMATCH classification would apply to any attempt to use the existing exonuclease-dead mouse models as recapitulating FILS, since they model a different, allele-specific mechanism within the same gene).

Summary of Key Evidence Sources (with exact-quote–ready findings)

1. Pachlopnik Schmid J, et al. "Polymerase ε1 mutation in a human syndrome with facial dysmorphism, immunodeficiency, livedo, and short stature ('FILS syndrome')." J Exp Med. 2012;209(13):2323-2330. PMID: 23230001. — Original description; homozygous intron 34 splice mutation; T/B cell, chondrocyte, osteoblast G1–S proliferation defect.
2. Logan CV, et al. "DNA Polymerase Epsilon Deficiency Causes IMAGe Syndrome with Variable Immunodeficiency." Am J Hum Genet. 2018. PMID: 30503519. — Recurrent c.1686+32C>G haplotype in 15 individuals/12 families; establishes the IMAGe-overlapping allelic subtype (OMIM #618336).
3. "Filling in the gaps on FILS syndrome: A case report and literature review." 2020. PMID: 32705701.
4. Case report: A Chinese boy with facial dysmorphism, immunodeficiency, livedo, and short stature syndrome. Front Pediatr. 2022;10:933108. PMC: 9441657. — Detailed phenotype/genotype/lab data for a mildly affected patient; notes only 3 prior studies/16 patients as of March 2022.
5. OMIM #615139 (link) and OMIM *174762 (POLE) (link).
6. Orphanet ORPHA:352712 (link) — prevalence <1/1,000,000.
7. Sehested A, et al. "Constitutional POLE variants causing a phenotype reminiscent of constitutional mismatch repair deficiency." Hum Mutat. 2022. PMID: 34816535. — Differential-diagnosis/allelic-spectrum context.
8. Mouse proofreading-deficient Pole models: PNAS 2009; JCI 2018 — mechanistically distinct hypermutator models, relevant only as a human-model-mismatch caveat.

Note on evidentiary limits: Given that fewer than 20 patients have ever been reported, virtually all phenotype-frequency statements in the literature are qualitative ("most," "all but 1/2 patients") rather than statistically robust percentages; a dismech entry should represent frequencies as FrequencyEnum qualitative bands rather than fabricated precise percentages, and flag the extremely small n explicitly in notes.

Reference Validation

Checked with linkml-reference-validator 0.2.1.

All extracted references resolved successfully.