A telomere biology disorder in which dyskeratosis congenita is produced not by a defect in telomere machinery but by a defect in nucleotide supply. Patients have the classic mucocutaneous triad and progressive bone marrow failure, and the cause is severe germline thymidylate synthase deficiency. Its genetic architecture is the reason this entry exists as its own record rather than as a row under the parent dyskeratosis congenita entry. It is **digenic**, and in an unusual configuration. An affected child inherits a loss-of-function coding variant in TYMS from one parent, and from the other a specific haplotype carrying rare variants in ENOSF1 - the naturally occurring antisense regulator transcribed from the opposite strand of the same locus. Elevated ENOSF1 silences the remaining wild-type TYMS allele post-transcriptionally. So the disease presents as recessive, and both parents are unaffected, yet one of them has an entirely wild-type TYMS coding sequence. Standard recessive segregation is not observed and a single-gene panel will not find it. Two things follow that shaped how this was curated. The mechanism runs through nucleotide metabolism rather than through the telomere. TYMS catalyses the only de novo route to dTMP; losing it depletes the thymidine nucleotide pool, distorts the dNTP balance, and produces genotoxic stress and defective transcription, from which abnormal telomere maintenance follows. The entry orders the chain that way, because the order is what makes this disease mechanistically different from the shelterin and telomerase forms of dyskeratosis congenita that it is clinically indistinguishable from. And there is a directly actionable pharmacogenomic consequence. Thymidylate synthase is the target of fluoropyrimidines, and patient cells are hypersensitive to 5-fluorouracil. Given that dyskeratosis congenita carries a raised risk of squamous cell carcinoma, a patient with this genotype may well be offered exactly the drug class their cells cannot tolerate. **A note on this entry's identifier.** MONDO asserts a causal-gene relationship from MONDO:0859319 to DCLRE1B (`MONDO:0859319 RO:0004003 HGNC:17641`). That assertion is wrong. OMIM 620133, to which MONDO:0859319 is mapped `equivalentTo`, is the TYMS-ENOSF1 entity defined by Tummala et al. DCLRE1B/Apollo is a real telomere gene with a real human phenotype, but it is a different one. This entry curates the disease OMIM 620133 describes; see the discussion below.
Ask a research question about Dyskeratosis Congenita Autosomal Recessive 8. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).
Do not include personal health information in your question. Questions and results are cached in your browser's local storage.
Conditions with similar clinical presentations that must be differentiated from Dyskeratosis Congenita Autosomal Recessive 8:
name: Dyskeratosis Congenita Autosomal Recessive 8
creation_date: "2026-09-11T21:00:00Z"
category: Mendelian
disease_term:
preferred_term: dyskeratosis congenita, autosomal recessive 8
term:
id: MONDO:0859319
label: dyskeratosis congenita, autosomal recessive 8
synonyms:
- DKCB8
- TYMS-ENOSF1 dyskeratosis congenita
- digenic dyskeratosis congenita
- germline thymidylate synthase deficiency
description: >-
A telomere biology disorder in which dyskeratosis congenita is produced not by a defect
in telomere machinery but by a defect in nucleotide supply. Patients have the classic
mucocutaneous triad and progressive bone marrow failure, and the cause is severe
germline thymidylate synthase deficiency.
Its genetic architecture is the reason this entry exists as its own record rather than
as a row under the parent dyskeratosis congenita entry. It is **digenic**, and in an
unusual configuration. An affected child inherits a loss-of-function coding variant in
TYMS from one parent, and from the other a specific haplotype carrying rare variants in
ENOSF1 - the naturally occurring antisense regulator transcribed from the opposite
strand of the same locus. Elevated ENOSF1 silences the remaining wild-type TYMS allele
post-transcriptionally. So the disease presents as recessive, and both parents are
unaffected, yet one of them has an entirely wild-type TYMS coding sequence. Standard
recessive segregation is not observed and a single-gene panel will not find it.
Two things follow that shaped how this was curated.
The mechanism runs through nucleotide metabolism rather than through the telomere. TYMS
catalyses the only de novo route to dTMP; losing it depletes the thymidine nucleotide
pool, distorts the dNTP balance, and produces genotoxic stress and defective
transcription, from which abnormal telomere maintenance follows. The entry orders the
chain that way, because the order is what makes this disease mechanistically different
from the shelterin and telomerase forms of dyskeratosis congenita that it is clinically
indistinguishable from.
And there is a directly actionable pharmacogenomic consequence. Thymidylate synthase is
the target of fluoropyrimidines, and patient cells are hypersensitive to
5-fluorouracil. Given that dyskeratosis congenita carries a raised risk of squamous cell
carcinoma, a patient with this genotype may well be offered exactly the drug class their
cells cannot tolerate.
**A note on this entry's identifier.** MONDO asserts a causal-gene relationship from
MONDO:0859319 to DCLRE1B (`MONDO:0859319 RO:0004003 HGNC:17641`). That assertion is
wrong. OMIM 620133, to which MONDO:0859319 is mapped `equivalentTo`, is the
TYMS-ENOSF1 entity defined by Tummala et al. DCLRE1B/Apollo is a real telomere gene with
a real human phenotype, but it is a different one. This entry curates the disease OMIM
620133 describes; see the discussion below.
parents:
- Dyskeratosis Congenita
prevalence:
- population: Worldwide
measure_type: CASES_IN_LITERATURE
prevalence_class: NOT_YET_DOCUMENTED
notes: >-
Eight independent families in the defining cohort, plus at least one subsequent
unrelated case. No prevalence estimate exists. Eight families is a substantial base
for an ultra-rare entity and is what makes the digenic mechanism credible rather than
anecdotal.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
In a cohort of eight independent DC-affected families, we have identified a
remarkable series of heterozygous germline variants in the gene encoding thymidylate
synthase (TYMS).
explanation: >-
Gives the size and structure of the defining cohort.
pathophysiology:
- name: Digenic TYMS-ENOSF1 Genotype
role: trigger
biological_scale: MOLECULAR
description: >-
The initiating lesion is a genotype rather than a variant: a TYMS loss-of-function
coding allele from one parent, and in trans a specific ENOSF1 haplotype with rare
variants from the other.
Neither allele alone causes disease - the parents carrying one each are unaffected -
and one parent's TYMS coding sequence is entirely wild type. This is what distinguishes
the disease from every single-gene recessive form of dyskeratosis congenita, and it is
why the inheritance only appears recessive.
genetic_context:
variant_origin: GERMLINE
functional_impact_category: LOSS_OF_FUNCTION
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
Although the inheritance appeared to be autosomal recessive, one parent in each
family had a wild-type TYMS coding sequence. Targeted genomic sequencing identified a
specific haplotype and rare variants in the naturally occurring TYMS antisense
regulator ENOSF1 (enolase super family 1) inherited from the other parent.
explanation: >-
The single sentence that establishes the digenic architecture and why it was missed
by a conventional recessive analysis.
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
compound heterozygosity for loss of function variants in TYMS and a specific
haplotype of its antisense regulator ENOSFI (enolase super family 1) causes digenic
DC
explanation: >-
Independent confirmation of the digenic requirement in an unrelated patient, three
years after the defining cohort.
downstream:
- target: ENOSF1-Mediated Silencing of the Wild-Type TYMS Allele
causal_link_type: DIRECT
description: >-
The ENOSF1 haplotype acts on the TYMS allele the other parent contributed.
- name: ENOSF1-Mediated Silencing of the Wild-Type TYMS Allele
role: mechanism
biological_scale: MOLECULAR
description: >-
The epistatic step, and the part of this disease that is genuinely novel. ENOSF1 is
transcribed antisense to TYMS at the same locus. When it is elevated, it silences TYMS
post-transcriptionally - so the coding-intact allele is present, transcribed, and then
prevented from producing enzyme.
This was not inferred from the pedigree structure. It was demonstrated by gene rescue
in cells from affected probands, which is what turns "these two variants co-occur" into
a mechanism.
biological_processes:
- preferred_term: antisense silencing of TYMS by ENOSF1
term:
id: GO:0042868
label: antisense RNA metabolic process
modifier: DYSREGULATED
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
Gene-rescue studies in cells from affected probands revealed that post-transcriptional
epistatic silencing of TYMS is occurring via elevated ENOSF1.
explanation: >-
The experimental demonstration of the epistasis, in patient cells. Graded IN_VITRO
because it is a cell-based rescue experiment, notwithstanding that the cells are of
human patient origin.
downstream:
- target: Severe Thymidylate Synthase Deficiency
causal_link_type: DIRECT
description: >-
One allele lost to coding variant, the other to silencing, leaves almost no enzyme.
- name: Severe Thymidylate Synthase Deficiency
role: mechanism
biological_scale: MOLECULAR
description: >-
The convergence point of the two alleles, and a severe deficiency rather than a partial
one - which is the reason a disease results at all from a genotype in which one coding
allele is intact.
molecular_functions:
- preferred_term: thymidylate synthase activity
term:
id: GO:0004799
label: thymidylate synthase activity
modifier: DECREASED
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
Lymphoblastoid cells from affected probands have severe TYMS deficiency, altered
cellular deoxyribonucleotide triphosphate pools, and hypersensitivity to the
TYMS-specific inhibitor 5-fluorouracil.
explanation: >-
Measures the enzyme deficiency, the nucleotide-pool consequence and the drug
sensitivity in one experiment on patient cells.
downstream:
- target: Deoxyribonucleotide Pool Imbalance
causal_link_type: DIRECT
description: >-
TYMS makes dTMP; without it the thymidine nucleotide pool falls and the balance
between the four dNTPs is distorted.
- target: Antimetabolite Hypersensitivity
causal_link_type: DIRECT
description: >-
A cell with almost no thymidylate synthase has no reserve against a drug that
inhibits thymidylate synthase.
- name: Deoxyribonucleotide Pool Imbalance
role: mechanism
biological_scale: CELLULAR
description: >-
Altered dNTP pools, which is the classical route by which a nucleotide-metabolism
defect becomes a DNA-integrity defect: a replication fork supplied with the wrong
nucleotide ratios misincorporates, stalls and breaks.
biological_processes:
- preferred_term: de novo dTMP biosynthesis
term:
id: GO:0006231
label: dTMP biosynthetic process
modifier: DECREASED
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
Lymphoblastoid cells from affected probands have severe TYMS deficiency, altered
cellular deoxyribonucleotide triphosphate pools
explanation: >-
The measured pool abnormality.
downstream:
- target: Genotoxic Stress and Defective Transcription
causal_link_type: DIRECT
description: >-
Distorted pools damage DNA and impair transcription.
- name: Genotoxic Stress and Defective Transcription
role: mechanism
biological_scale: CELLULAR
description: >-
The cellular injury the pool imbalance produces. Notably the source reports defective
transcription alongside the DNA damage, which is a broader consequence than replication
stress alone and is curated as the authors stated it rather than narrowed to the
replication story that would be easier to tell.
biological_processes:
- preferred_term: DNA damage response to nucleotide misincorporation
term:
id: GO:0006974
label: DNA damage response
modifier: INCREASED
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
These defects in the nucleotide metabolism pathway resulted in genotoxic stress,
defective transcription, and abnormal telomere maintenance.
explanation: >-
Names all three downstream consequences in the order the authors put them, which is
the order this entry's causal chain follows.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
an increase in the phosphorylation levels of proteins ATM, CHK1, CHK2, p53, p21 in the
DNA-damage-response pathway
explanation: >-
The measurement behind the GO binding. This node was previously bound to
base-excision repair, which described a plausible mechanism rather than the one the
authors measured; the pathway they assayed is the DNA damage response, so that is
what the node now binds.
downstream:
- target: Abnormal Telomere Maintenance
causal_link_type: DIRECT
description: >-
Telomere dysfunction here is downstream of genotoxic stress, not the initiating
lesion.
- name: Abnormal Telomere Maintenance
role: mechanism
biological_scale: CELLULAR
description: >-
What makes this a telomere biology disorder clinically, and the finding that places it
alongside the telomerase and shelterin forms of dyskeratosis congenita in diagnosis and
management.
The mechanistic point is the direction of causation, and it has to be stated
precisely. In DKC1, TERT, TERC or TINF2 disease the primary lesion is in the telomere
machinery. Here the primary lesion is in nucleotide supply and the telomere defect
follows from it. That is not the same as the machinery being untouched: in proband
cells telomerase activity is reduced, immature oligoadenylated TERC accumulates, and
DKC1, TERT and SMUG1 are all reduced. What is preserved is shelterin. The authors'
own reading is that impaired telomerase regulation is a consequence of altered
nucleotide metabolism rather than an independent lesion.
An earlier draft of this entry said the telomere machinery was intact. That was too
strong and is corrected here.
biological_processes:
- preferred_term: telomere maintenance
term:
id: GO:0000723
label: telomere maintenance
modifier: ABNORMAL
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
These defects in the nucleotide metabolism pathway resulted in genotoxic stress,
defective transcription, and abnormal telomere maintenance.
explanation: >-
States that the telomere abnormality is a consequence of the nucleotide defect, which
is the ordering claim this node depends on.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: IN_VITRO
snippet: >-
whereas the amount of shelterin protein components (TIN2, TPP1, TRF1 and POT1; Figure
2J), which protect telomeres, remained unchanged, indicating that defects in telomere
length maintenance are a consequence of impaired telomerase regulation in these TYMS
deficient probands as a result of altered nucleotide metabolism
explanation: >-
The sentence that fixes what is and is not preserved. Shelterin is unchanged;
telomerase regulation is impaired, and the authors place that impairment downstream
of the nucleotide defect. This is the evidence for the corrected framing above.
downstream:
- target: Dyskeratosis Congenita Phenotype
causal_link_type: DIRECT
description: >-
Telomere dysfunction in renewing tissues gives the clinical syndrome.
- name: Antimetabolite Hypersensitivity
role: modifier
biological_scale: CELLULAR
description: >-
A branch of the mechanism with direct clinical consequence rather than a downstream
step of the disease process. Thymidylate synthase is the pharmacological target of
5-fluorouracil and capecitabine; a cell that has almost none of it has no margin. The
same cells are separately hypersensitive to hydroxyurea, which inhibits ribonucleotide
reductase rather than thymidylate synthase, so the vulnerability is to antimetabolites
that squeeze dNTP supply generally and not to one drug class alone.
This matters because dyskeratosis congenita carries an elevated risk of squamous cell
carcinoma, for which fluoropyrimidines are a standard option. The genotype that causes
the cancer risk also removes the tolerance for a common treatment of it.
It is not only a prediction. One proband in the defining cohort was treated with
topical 5-fluorouracil for squamous carcinoma and melanoma of the leg and had a severe
adverse response. That is a single topical exposure rather than systemic dosing data,
but it is a clinical observation and this entry previously stated, wrongly, that no
such observation existed.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
has shown a severe adverse response to topical 5-FU when undergoing treatment for her
squamous carcinoma and melanoma in her leg
explanation: >-
A patient with a confirmed TYMS variant who received a fluoropyrimidine and reacted
severely to it. This is the clinical counterpart of the cellular hypersensitivity
below, and it is why the node is not graded as a cell-only finding.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Bilateral ptosis, tooth discolouration, intermediate increased response to
mitomycin-C, basal carcinoma on the chest, squamous carcinoma and melanoma on the
leg, severe response to topical 5-FU treatment.
explanation: >-
The Table 1 footnote for the same proband, which corroborates the 5-FU reaction and
places it alongside her three cutaneous malignancies.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: INDIRECT
evidence_source: IN_VITRO
snippet: >-
hypersensitivity to the TYMS-specific inhibitor 5-fluorouracil
explanation: >-
The measured drug sensitivity in patient-derived cells. Kept as a separate item from
the clinical observation above because it is a different kind of claim: it
establishes the mechanism of the reaction rather than the reaction itself.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: INDIRECT
evidence_source: IN_VITRO
snippet: >-
hypersensitivity to ribonucleotide reductase inhibitors such as hydroxyurea
explanation: >-
Extends the hypersensitivity beyond thymidylate synthase inhibitors, which is why
this node is named for antimetabolites rather than for fluoropyrimidines. INDIRECT
and IN_VITRO: hydroxyurea sensitivity was measured in cells and no patient exposure
to it has been reported.
- name: Dyskeratosis Congenita Phenotype
role: outcome
biological_scale: ORGANISM
description: >-
The clinical endpoint: the mucocutaneous triad with progressive bone marrow failure,
clinically indistinguishable from other forms of dyskeratosis congenita.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
Dyskeratosis congenita (DC) is an inherited bone-marrow-failure disorder
characterized by a triad of mucocutaneous features that include abnormal skin
pigmentation, nail dystrophy, and oral leucoplakia.
explanation: >-
Defines the syndrome these patients present with. Note this sentence describes
dyskeratosis congenita in general rather than DKCB8 specifically, which is why the
individual phenotype rows below that need a DKCB8-specific source use the case
report instead.
phenotypes:
- category: Dermatologic
name: Abnormal Skin Pigmentation
frequency: VERY_FREQUENT
description: >-
Part of the mucocutaneous triad. In the confirmed DKCB8 case report it was diffuse
hyperpigmentation with numerous punctate hypopigmented macules, appearing in early
infancy.
phenotype_term:
preferred_term: Reticulated skin pigmentation
term:
id: HP:0007427
label: Reticulated skin pigmentation
evidence:
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
In early infancy, he developed diffuse hyperpigmentation as well as numerous punctate
hypopigmented macules, sparse hair, and nail dystrophy
explanation: >-
The pigmentary findings in a genetically confirmed DKCB8 patient, with their onset.
- category: Dermatologic
name: Nail Dystrophy
frequency: VERY_FREQUENT
phenotype_term:
preferred_term: Nail dystrophy
term:
id: HP:0008404
label: Nail dystrophy
description: >-
Part of the triad, and documented in the genetically confirmed case.
evidence:
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
diffuse hyperpigmentation as well as numerous punctate hypopigmented macules, sparse
hair, and nail dystrophy
explanation: >-
Names the nail dystrophy in a confirmed patient.
- category: Oral
name: Oral Leukoplakia
frequency: FREQUENT
description: >-
The third element of the triad, and the one element of it that is not near-universal in
this cohort: Table 1 records leukoplakia in 4 of the 10 probands, against 10 of 10 for
skin pigmentation and nail dystrophy. Graded from the cohort rather than from the
class-level description of dyskeratosis congenita.
phenotype_term:
preferred_term: Oral leukoplakia
term:
id: HP:0002745
label: Oral leukoplakia
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: INDIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
a triad of mucocutaneous features that include abnormal skin pigmentation, nail
dystrophy, and oral leucoplakia
explanation: >-
Names oral leukoplakia as part of the dyskeratosis congenita triad. Graded INDIRECT
because the sentence characterises dyskeratosis congenita as a class; that these
particular patients had leukoplakia follows from their diagnosis rather than being
separately reported in the abstract.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Leukoplakia | Y | N | N | N | N | N | Y | N | Y | Y
explanation: >-
The Table 1 row, 4 of 10 probands, which is what the FREQUENT band is graded from.
- category: Hematologic
name: Bone Marrow Failure
frequency: OCCASIONAL
description: >-
Graded from the cohort, not from the disease class, and the two disagree sharply. Marrow
failure is definitional for dyskeratosis congenita and dominates its prognosis; in the
eight-family TYMS cohort Table 1 records a haematological abnormality in 1 of 10
probands (anemia at 98 g/L with otherwise normal counts), with three further probands
having normal counts but raised HbF. An earlier version of this entry graded this
VERY_FREQUENT with severity SEVERE on the class-level statement alone.
Two readings are open and the data do not separate them. Most of these probands were
sampled young - six were four years old or less - and marrow failure in dyskeratosis
congenita is progressive, so the cohort may simply be too young. Alternatively the
TYMS-ENOSF1 form may genuinely spare the marrow relative to the telomerase forms, which
would be a real distinction worth knowing. Raised HbF in three probands is a hint of
early marrow stress in either reading. The `severity` qualifier has been dropped because
nothing in either source supports a severity grading for this cohort.
phenotype_term:
preferred_term: Bone marrow failure
term:
id: HP:0005528
label: Bone marrow hypocellularity
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Hematological abnormalities | N a | N | N b | Y c | N | N | N d | N | N | N
explanation: >-
The Table 1 row the OCCASIONAL band is graded from. One affected proband out of ten;
footnotes a, b and d are raised HbF on otherwise normal counts and footnote c is the
anemia.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: INDIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Bone-marrow failure, predisposition to cancer, and pulmonary abnormalities are
reported to be major causes of death in DC-affected individuals.
explanation: >-
The class-level expectation, retained because it is what makes the cohort figure
surprising. INDIRECT: it is a statement about dyskeratosis congenita, not about these
patients.
- category: Immunologic
name: Decreased Circulating Immunoglobulin
frequency: FREQUENT
description: >-
The most under-recognised finding in this cohort and absent from an earlier version of
this entry. Table 1 records immune defects in 6 of 10 probands, with one negative and
three unknown; the footnotes specify the defect as low IgM, low IgA, or low IgA with
low IgG. Three probands additionally had recurrent infections in the first year of
life. Hypogammaglobulinemia is not part of the classical dyskeratosis congenita triad,
so it is worth checking for in a patient with this genotype.
phenotype_term:
preferred_term: Low IgA, IgM or IgG
term:
id: HP:0004313
label: Decreased circulating immunoglobulin concentration
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Immune defects | Y e | Y f | Y g | Y h | Y i | ? | N | Y j | ? | ?
explanation: >-
The Table 1 row the FREQUENT band is graded from: six affected, one unaffected, three
unknown.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
eLow IgM.fLow IgA and IgG.gLow IgM.hLow IgA.iLow IgA.jLow IgA.
explanation: >-
The footnotes behind that row, which is what fixes the binding to decreased
immunoglobulin concentration rather than to immunodeficiency generally.
- category: Neoplastic
name: Cutaneous Malignancy
frequency: OCCASIONAL
description: >-
One proband, aged 26 at sampling, had basal cell carcinoma of the chest plus squamous
carcinoma and melanoma of the leg. That is 1 of 10 in a cohort most of whom were
children, so the figure is a floor rather than a lifetime risk; cancer predisposition
is a recognised feature of dyskeratosis congenita generally.
This row exists because it is load-bearing for the rest of the entry rather than
because one case is much evidence: it is the same proband whose squamous carcinoma was
treated with topical 5-fluorouracil, and the reaction to that treatment is this entry's
one genotype-specific management point.
phenotype_term:
preferred_term: Basal cell carcinoma, squamous carcinoma and melanoma of the skin
term:
id: HP:0008069
label: Neoplasm of the skin
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
basal carcinoma on the chest, squamous carcinoma and melanoma on the leg
explanation: >-
The three cutaneous malignancies in that proband, from her Table 1 footnote.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: INDIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Bone-marrow failure, predisposition to cancer, and pulmonary abnormalities are
reported to be major causes of death in DC-affected individuals.
explanation: >-
The class-level cancer predisposition this sits within. INDIRECT: about dyskeratosis
congenita rather than about TYMS probands.
- category: Dental
name: Abnormal Dentition
frequency: OCCASIONAL
description: >-
Recorded in the Table 1 other-features footnotes for two probands, as abnormal teeth in
one and tooth discolouration in the other, and shown in the paper's clinical
photographs. Dental abnormality is a recognised feature of dyskeratosis congenita and
of the ectodermal dysplasia phenotypes the authors note it overlaps with.
phenotype_term:
preferred_term: Abnormal teeth and tooth discolouration
term:
id: HP:0000164
label: Abnormality of the dentition
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Photographs of affected probands show some of the clinical features: sparse hair, nail
dystrophy, abnormal skin pigmentation, and abnormal dentition.
explanation: >-
Names abnormal dentition among the documented clinical features of these probands.
- category: Dermatologic
name: Sparse Hair
frequency: VERY_FREQUENT
phenotype_term:
preferred_term: Sparse hair and thin eyelashes
term:
id: HP:0008070
label: Sparse hair
description: >-
Near-universal in the defining cohort: Table 1 records hair loss and thin eyelashes in
9 of 10 probands, the tenth being unknown rather than negative. An earlier version of
this entry graded this OCCASIONAL on the strength of the single case report alone,
which understated it by three bands.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
Hair loss and thin eye lashes | Y | Y | Y | Y | Y | Y | Y | ? | Y | Y
explanation: >-
The Table 1 row the VERY_FREQUENT band is graded from. Nine yes, one unknown, no
negatives.
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
numerous punctate hypopigmented macules, sparse hair, and nail dystrophy
explanation: >-
The independent confirmatory case, which agrees with the cohort.
- category: Hematologic
name: Short Telomere Length
frequency: VERY_FREQUENT
description: >-
Very short telomeres are the screening biomarker for the telomere biology disorders and
were what confirmed the diagnosis in the reported case.
phenotype_term:
preferred_term: Short telomere length
term:
id: HP:0031413
label: Short telomere length
evidence:
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
diagnosis of DC was confirmed with a telomere length assay
explanation: >-
Establishes that the telomere assay was abnormal and diagnostic in a confirmed DKCB8
patient.
genetic:
- name: TYMS
gene_term:
preferred_term: TYMS
term:
id: hgnc:12441
label: TYMS
relationship_type: CAUSATIVE
notes: >-
Thymidylate synthase, which catalyses the sole de novo route to dTMP by methylating
dUMP using 5,10-methylenetetrahydrofolate. Patients carry a heterozygous
loss-of-function coding variant; on its own this is not sufficient, and the transmitting
parent is unaffected.
TYMS is also the pharmacological target of the fluoropyrimidines, which is where the
clinical hypersensitivity comes from.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
we have identified a remarkable series of heterozygous germline variants in the gene
encoding thymidylate synthase (TYMS)
explanation: >-
Establishes TYMS as the coding-variant locus and the variants as heterozygous.
- name: ENOSF1
gene_term:
preferred_term: ENOSF1
term:
id: hgnc:30365
label: ENOSF1
relationship_type: COOPERATING
notes: >-
The antisense regulator transcribed from the opposite strand at the TYMS locus.
Recorded as COOPERATING rather than CAUSATIVE because that is precisely its role: the
ENOSF1 haplotype causes nothing by itself, and is pathogenic only in trans to a TYMS
loss-of-function allele.
The contributing lesion is a haplotype carrying rare variants rather than a single
variant, which is worth noting for anyone trying to interpret a laboratory report -
there is no one ENOSF1 variant to look up.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
Targeted genomic sequencing identified a specific haplotype and rare variants in the
naturally occurring TYMS antisense regulator ENOSF1 (enolase super family 1)
inherited from the other parent.
explanation: >-
Establishes ENOSF1's contribution, that it is a haplotype rather than a single
variant, and that it comes from the other parent.
inheritance:
- name: Digenic inheritance
inheritance_term:
preferred_term: Digenic inheritance
term:
id: HP:0010984
label: Digenic inheritance
description: >-
Two loci, both required, one contributed by each parent: a TYMS loss-of-function coding
allele and, in trans, an ENOSF1 antisense haplotype with rare variants. Carriers of
either alone are unaffected.
The presentation mimics autosomal recessive inheritance closely enough that the defining
study initially read it that way, and only targeted sequencing of the antisense
regulator revealed otherwise. That is worth recording as a diagnostic caution rather
than as a historical note: a pedigree that looks recessive with an apparently
non-carrier parent is the signal.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
These cell and molecular abnormalities generated by the combination of germline
digenic variants at the TYMS-ENOSF1 locus represent a unique pathogenetic pathway for
DC causation in these affected individuals, whereas the parents who are carriers of
either of these variants in a singular fashion remain unaffected.
explanation: >-
States the digenic requirement and, in the same sentence, that single carriers are
unaffected - which is the definition of digenic rather than modifier inheritance.
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
compound heterozygosity for loss of function variants in TYMS and a specific haplotype
of its antisense regulator ENOSFI (enolase super family 1) causes digenic DC
explanation: >-
Independent statement of the digenic mode in a second report.
diagnosis:
- name: Telomere Length Assay
description: >-
Very short telomeres are the screening test for the telomere biology disorders and were
what confirmed the diagnosis in the reported DKCB8 case. It does not identify the gene,
only the disease class.
evidence:
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
diagnosis of DC was confirmed with a telomere length assay
explanation: >-
The confirmatory test in a genetically established case.
- name: Targeted Sequencing of the TYMS-ENOSF1 Locus
description: >-
The genetic diagnosis needs the antisense regulator, not just the coding gene. A
standard dyskeratosis congenita panel returns a single heterozygous TYMS variant, which
in isolation looks like carrier status rather than a diagnosis, and the ENOSF1
contribution is a haplotype rather than a reportable variant.
This is why the defining cohort needed targeted genomic sequencing over the locus after
conventional analysis had failed.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
Targeted genomic sequencing identified a specific haplotype and rare variants in the
naturally occurring TYMS antisense regulator ENOSF1 (enolase super family 1)
inherited from the other parent.
explanation: >-
Names the assay that made the diagnosis possible.
- name: Consider DKCB8 in 18p Deletions Encompassing TYMS
description: >-
A specific ascertainment route. The reported case had ring chromosome 18 with partial
18p monosomy that deleted TYMS, providing the loss-of-function allele structurally
rather than by point mutation.
The generalisable point is that a patient with an 18p deletion spanning TYMS has one of
the two required lesions already, and should be evaluated for the dyskeratosis congenita
features that would otherwise be attributed to the chromosomal syndrome.
evidence:
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
Our case highlights that individuals with deletions at 18p encompassing TYMS should be
evaluated for features of digenic dyskeratosis congenita.
explanation: >-
The authors' own recommendation, which is the diagnostic point of their report.
treatments:
- name: Avoidance of Fluoropyrimidines and Hydroxyurea
therapeutic_modality: OTHER
treatment_term:
preferred_term: avoidance of thymidylate synthase inhibitors
term:
id: NCIT:C15747
label: Supportive Care
description: >-
The one management point specific to this genotype rather than shared with dyskeratosis
congenita generally. Patient cells are hypersensitive to 5-fluorouracil, which acts on
the enzyme these patients already lack, and separately to hydroxyurea.
This is not a hypothetical. A proband in the defining cohort developed squamous
carcinoma and melanoma of the leg - the cancer risk dyskeratosis congenita carries -
was treated with topical 5-fluorouracil, and had a severe adverse response. The
situation the recommendation covers is therefore one that has already occurred once.
What remains unestablished is the systemic case: that exposure was topical, in a single
patient, with no dose data and no information on how a systemic fluoropyrimidine would
behave. The recommendation to avoid is accordingly firmer than a pure cell-culture
inference and weaker than a dose-adjustment guideline.
target_mechanisms:
- target: Antimetabolite Hypersensitivity
treatment_effect: MODULATES
description: >-
Avoidance does not act on the disease mechanism. It removes an exposure the genotype
cannot tolerate.
evidence:
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: DIRECT
evidence_source: HUMAN_CLINICAL
snippet: >-
has shown a severe adverse response to topical 5-FU when undergoing treatment for her
squamous carcinoma and melanoma in her leg
explanation: >-
The observed clinical reaction that the avoidance recommendation rests on. One
patient, topical route.
- reference: PMID:35931051
reference_title: "Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita."
supports: SUPPORT
directness: INDIRECT
evidence_source: IN_VITRO
snippet: >-
hypersensitivity to the TYMS-specific inhibitor 5-fluorouracil
explanation: >-
The cellular hypersensitivity that explains the reaction and generalises it beyond
the one patient in whom it was seen. INDIRECT because a cell-culture dose response
does not by itself fix a clinical threshold.
notes: >-
Corrected after review. An earlier version of this entry stated in four places that no
DKCB8 patient had ever received a fluoropyrimidine, and built a knowledge gap and a
proposed experiment on that premise. It is not true: the cached full text of
PMID:35931051 reports a proband with a severe adverse response to topical 5-FU during
treatment for squamous carcinoma and melanoma. The claim came from reading the
abstract and the deep-research report, which makes the same omission, rather than the
full text that was already in this pull request.
What the evidence now supports: one observed topical reaction, plus cellular
hypersensitivity to both 5-fluorouracil and hydroxyurea in proband lymphoblastoid
cells. What it does not support: any statement about systemic dosing, about severity
grading, or about whether DPYD-based pharmacogenomic screening would flag such a
patient.
- name: Malignancy Surveillance
therapeutic_modality: BEHAVIORAL
treatment_term:
preferred_term: cancer screening
term:
id: NCIT:C15747
label: Supportive Care
description: >-
Shared with dyskeratosis congenita generally rather than specific to this genotype, but
it is the piece of class-level management this entry cannot omit: the cohort contains a
proband with basal cell carcinoma, squamous carcinoma and melanoma, and this entry's
one genotype-specific recommendation is about what not to treat such a cancer with.
evidence:
- reference: PMID:20301779
reference_title: "Dyskeratosis Congenita and Related Telomere Biology Disorders."
supports: SUPPORT
directness: INDIRECT
evidence_source: OTHER
snippet: >-
For cancer risk: monthly self-examination for oral, head, and neck cancer; annual
cancer screening by an otolaryngologist and dermatologist; annual gynecologic
examination.
explanation: >-
The GeneReviews surveillance schedule for dyskeratosis congenita and the telomere
biology disorders. INDIRECT: GeneReviews predates the TYMS-ENOSF1 entity and does not
mention TYMS, so this applies to DKCB8 by virtue of its being a telomere biology
disorder rather than by having been studied in it.
notes: >-
Recorded here rather than left to the parent entry because the interaction with the
antimetabolite hypersensitivity above is genotype-specific: surveillance finds the
cancer, and the genotype constrains how it can be treated.
- name: Androgen Therapy and Hematopoietic Cell Transplantation
therapeutic_modality: OTHER
treatment_term:
preferred_term: management of bone marrow failure in a telomere biology disorder
term:
id: NCIT:C49236
label: Therapeutic Procedure
description: >-
Class-level management of the marrow failure that defines dyskeratosis congenita.
Recorded for completeness, and with an explicit caveat: the eight-family TYMS cohort
was largely young at sampling and showed little haematological abnormality, so how
often DKCB8 patients reach a transplant decision is not established.
evidence:
- reference: PMID:20301779
reference_title: "Dyskeratosis Congenita and Related Telomere Biology Disorders."
supports: SUPPORT
directness: INDIRECT
evidence_source: OTHER
snippet: >-
Hematopoietic cell transplantation (HCT) is the only curative
treatment for BMF and leukemia, but long-term outcome has historically been poor
due to treatment toxicity; if a suitable donor is not available, androgen
therapy may be considered for BMF.
explanation: >-
The GeneReviews statement of the two options for marrow failure. INDIRECT for the
same reason as the surveillance row: it is class-level guidance, not DKCB8 data.
- reference: PMID:20301779
reference_title: "Dyskeratosis Congenita and Related Telomere Biology Disorders."
supports: SUPPORT
directness: INDIRECT
evidence_source: OTHER
snippet: >-
Of note, cancer therapy may pose an increased risk for prolonged
cytopenias as well as pulmonary and hepatic toxicity.
explanation: >-
Worth carrying because it compounds the genotype-specific problem above: cancer
therapy is already hazardous in a telomere biology disorder before the thymidylate
synthase deficiency is taken into account.
- name: Genetic Counselling
therapeutic_modality: BEHAVIORAL
treatment_term:
preferred_term: genetic counseling
term:
id: NCIT:C15240
label: Genetic Counseling
description: >-
Harder than for a recessive disease, and the difficulty is worth stating. Recurrence
risk is not the standard 25%: it depends on transmission of both the TYMS
loss-of-function allele and the ENOSF1 haplotype, which segregate independently. Testing
a relative for the TYMS variant alone gives an incomplete answer, and no published
recurrence figure exists.
differential_diagnoses:
- name: Telomerase and shelterin forms of dyskeratosis congenita
description: >-
Clinically indistinguishable - same triad, same marrow failure, same short telomeres -
and separable only genetically. The practical consequence is that a negative standard
dyskeratosis congenita panel does not exclude a telomere biology disorder, because this
one is not on it in a form the panel can call.
- name: Ring chromosome 18 and 18p deletion syndromes
description: >-
Works in both directions. A patient ascertained for an 18p deletion may have DKCB8
hiding inside the chromosomal diagnosis if the deletion spans TYMS and an ENOSF1
haplotype is present in trans; and features of the chromosomal syndrome may be
misattributed to dyskeratosis congenita. The single confirmed case report is exactly
this situation.
evidence:
- reference: PMID:40207375
reference_title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
supports: SUPPORT
evidence_source: HUMAN_CLINICAL
snippet: >-
The patient had physical and developmental features of 18p monosomy, including poor
growth, feeding issues, distinctive facial features, and strabismus.
explanation: >-
Shows the two diagnoses co-occurring, with features attributable to the chromosomal
abnormality rather than to DKCB8 - which is why this entry does not curate those
features as DKCB8 phenotypes.
discussions:
- discussion_id: dkcb8_mondo_gene_association_is_wrong
kind: KNOWLEDGE_GAP
prompt: >-
MONDO asserts DCLRE1B as the causal gene of MONDO:0859319, but OMIM 620133 - to which
MONDO maps this term equivalentTo - is the TYMS-ENOSF1 entity. Which is right, and what
should downstream consumers of the MONDO axiom do?
rationale: >-
This is a data-quality problem rather than a biological one, and it is recorded here
because it actively misled this curation.
MONDO contains the edge `MONDO:0859319 RO:0004003 HGNC:17641`, naming DCLRE1B. The
dismech curation stub for this disease was generated by `just enrich-stubs`, which reads
exactly that edge, so the stub carried DCLRE1B; the claim issue written from the stub
repeated it; and the curation began by assembling DCLRE1B literature. The error was
caught only when the deep-research report came back about TYMS.
The evidence that TYMS-ENOSF1 is correct is direct: OMIM 620133 is mapped equivalentTo
MONDO:0859319, and the entity OMIM 620133 describes is the one Tummala et al. defined
across eight families, confirmed independently in 2025. DCLRE1B is a genuine telomere
gene with a genuine human phenotype - Apollo is the nuclease that processes leading-end
telomeres, and a dominant-negative splice variant was reported in a Hoyeraal-Hreidarsson
patient - but that is a different entity from DKCB8.
Two things follow. Upstream, the MONDO axiom should be corrected. Downstream, nothing in
the dismech stub records that `genes:` is a machine-copied single ontology assertion
rather than a curated fact, so it reads with the same authority as the rest of the file.
That is the generalisable gap: a provenance or confidence marker on the enriched fields
would have made this checkable without a deep-research run.
attaches_to:
- genetic#TYMS
- genetic#ENOSF1
- discussion_id: dkcb8_fluoropyrimidine_toxicity_beyond_the_index_case
kind: KNOWLEDGE_GAP
prompt: >-
One DKCB8 proband had a severe adverse response to topical 5-fluorouracil. Does that
extend to systemic fluoropyrimidines, at what dose, and would standard DPYD-based
pharmacogenomic screening identify such a patient?
rationale: >-
Re-scoped after review. This discussion previously asked whether the cellular
hypersensitivity translates to patients at all, and asserted that no DKCB8 patient had
ever received a fluoropyrimidine. That assertion was false: the defining cohort reports
a proband with a severe adverse response to topical 5-fluorouracil during treatment for
squamous carcinoma and melanoma of the leg. The error came from working off the
abstract and the deep-research report rather than the cached full text.
What is actually open is narrower and more useful. The reported exposure was topical
and in one patient, so there is no systemic dosing experience, no severity grading
against a standard toxicity scale, and no attenuated-dose protocol. Separately, nothing
is known about whether DPYD-based screening would flag such a patient; on mechanism it
should not, because the lesion is in the drug's target rather than in its catabolic
pathway, and that prediction has never been tested.
The asymmetry that made this worth recording is unchanged and is now better supported:
the patient who most needs a fluoropyrimidine is the one whose dyskeratosis congenita
gave her the squamous cell carcinoma, and she is the patient least able to tolerate
it.
attaches_to:
- pathophysiology#Antimetabolite Hypersensitivity
- treatments#Avoidance of Fluoropyrimidines and Hydroxyurea
proposed_experiments:
- experiment_id: dkcb8_dpd_screening_adequacy
name: Retrospective review of fluoropyrimidine exposure in telomere biology disorder patients with TYMS variants
description: >-
Starting from the one reported topical exposure, identify further patients with
dyskeratosis congenita and a TYMS loss-of-function allele who have received a
fluoropyrimidine by any route, and review toxicity against matched dyskeratosis
congenita patients without one. Systemic exposures are the ones that matter, since
the reported case is topical. In parallel, test whether standard DPYD genotype-based
screening would have identified any of them as at risk.
readouts:
- name: Grade 3-4 fluoropyrimidine toxicity rate
target: pathophysiology#Antimetabolite Hypersensitivity
direction: INCREASED
interpretation: >-
Excess toxicity in TYMS-variant carriers would convert the cell-based prediction into
a clinical contraindication.
would_support:
- pathophysiology#Antimetabolite Hypersensitivity
supporting_outcome:
- >-
Severe toxicity at standard systemic doses in TYMS-variant patients who passed DPYD
screening, extending the reported topical reaction to systemic exposure and
establishing the inadequacy of current screening for it.
would_refute:
- pathophysiology#Antimetabolite Hypersensitivity
refuting_outcome:
- >-
Systemic fluoropyrimidine tolerance comparable to other dyskeratosis congenita
patients, which would confine the hazard to topical high-local-concentration exposure
rather than establishing a systemic contraindication.
references:
- reference: PMID:35931051
title: Germline thymidylate synthase deficiency impacts nucleotide metabolism and causes dyskeratosis congenita.
findings:
- statement: >-
Heterozygous germline TYMS loss-of-function variants, inherited in trans with a
specific ENOSF1 haplotype from an unaffected parent with a wild-type TYMS coding
sequence, cause dyskeratosis congenita across eight independent families.
supporting_text: >-
In a cohort of eight independent DC-affected families, we have identified a remarkable
series of heterozygous germline variants in the gene encoding thymidylate synthase
(TYMS).
- statement: >-
The telomere defect is downstream of nucleotide metabolism: shelterin is unchanged
while telomerase regulation is impaired.
supporting_text: >-
whereas the amount of shelterin protein components (TIN2, TPP1, TRF1 and POT1; Figure
2J), which protect telomeres, remained unchanged, indicating that defects in telomere
length maintenance are a consequence of impaired telomerase regulation in these TYMS
deficient probands as a result of altered nucleotide metabolism
- statement: >-
A proband treated with topical 5-fluorouracil for cutaneous malignancy had a severe
adverse response, which is the only reported clinical drug exposure in this disease.
supporting_text: >-
has shown a severe adverse response to topical 5-FU when undergoing treatment for her
squamous carcinoma and melanoma in her leg
- reference: PMID:40207375
title: "TYMS-ENOSF1 Dyskeratosis Congenita in a Patient With Ring Chromosome 18: A Case Report."
findings:
- statement: >-
Independent confirmation of the TYMS-ENOSF1 entity in a patient whose ring chromosome
18 supplied the TYMS deletion, with the classical mucocutaneous triad and a confirmed
short telomere length.
supporting_text: >-
In early infancy, he developed diffuse hyperpigmentation as well as numerous punctate
hypopigmented macules, sparse hair, and nail dystrophy
- reference: PMID:20301779
title: Dyskeratosis Congenita and Related Telomere Biology Disorders.
tags:
- GeneReviews
findings:
- statement: >-
Class-level management of the telomere biology disorders: transplantation or androgens
for marrow failure, and a defined cancer-surveillance schedule. This GeneReviews
chapter predates the TYMS-ENOSF1 entity and does not mention TYMS, so it is cited here
only for management shared across the telomere biology disorders and never for a
DKCB8-specific claim.
supporting_text: >-
Hematopoietic cell transplantation (HCT) is the only curative
treatment for BMF and leukemia, but long-term outcome has historically been poor
due to treatment toxicity; if a suitable donor is not available, androgen
therapy may be considered for BMF.
notes: >-
Identity. This entry curates the disease OMIM 620133 describes: digenic TYMS-ENOSF1
dyskeratosis congenita. MONDO's causal-gene axiom for MONDO:0859319 names DCLRE1B and is
wrong; the discussion above records the evidence and how the error propagated into this
repository's curation stub. The entry's filename and `name` retain the MONDO label so the
identifier and the record stay aligned.
Why there is no DCLRE1B `genetic:` record. A later reviewer will reasonably ask why the
disagreement is recorded only in prose, when a `genetic:` entry graded DISPUTED would make
it machine-readable - which is what the sibling Embryonal_Rhabdomyosarcoma entry does for
the gene MONDO wrongly attributes to it. The answer is the KGX exporter. It hardcodes the
predicate to `biolink:contributes_to` and never reads `relationship_type`
(monarch-initiative/dismech#12320), so a DISPUTED DCLRE1B record here would export as a
plain causative gene-disease edge asserting exactly the relationship this entry says is
false. The rhabdomyosarcoma case is different only because the gene there is genuinely
reported in that disease, however thinly, whereas DCLRE1B belongs to a different entity
altogether. Revisit this once #12320 is fixed.
Evidence base. Two clinical references: the eight-family defining cohort and one
independent confirmatory case report. Eight families is unusually strong for an ultra-rare
entity and is what makes the digenic claim credible; but the individual phenotype detail
comes largely from the single case report, and two phenotype rows are graded INDIRECT
because their support is a statement about dyskeratosis congenita as a class rather than
about these patients.
What is graded IN_VITRO and why it matters. The epistatic silencing, the enzyme
deficiency, the nucleotide pool imbalance, the genotoxic stress and the drug
hypersensitivities were all measured in proband-derived lymphoblastoid cells. That is
strong evidence about mechanism and it is not, by itself, evidence about patients.
The fluoropyrimidine correction. An earlier version of this entry said in four places
that no DKCB8 patient had ever received a fluoropyrimidine and that the avoidance
recommendation was purely this entry's inference from cell data. That was wrong. The
cached full text of PMID:35931051 reports a proband with a severe adverse response to
topical 5-FU during treatment for her squamous carcinoma and melanoma, and the Table 1
footnote for that proband records the same. The claim was written from the abstract and
from a deep-research report that makes the same omission, while the full text sat in the
same pull request. The entry now carries that observation as HUMAN_CLINICAL evidence on
the pathophysiology node and on the treatment, and the associated discussion asks the
narrower question that is genuinely open: whether a topical reaction in one patient
extends to systemic dosing.
Phenotype frequencies are graded from Table 1 of the defining cohort, not from the
disease class. That matters most for bone marrow failure, which is definitional for
dyskeratosis congenita and appears in only 1 of 10 probands here; see that row for the
two readings the data cannot separate. Leukoplakia is likewise 4 of 10 rather than
near-universal, while hair loss is 9 of 10, and immunoglobulin deficiency is present in 6
of 10 despite not being part of the classical triad.
One confounder worth carrying forward. The only individually described patient also has
ring chromosome 18 with partial 18p and 18q monosomy. His poor growth, feeding
difficulties, facial features and strabismus are attributed by his own authors to the 18p
monosomy and are deliberately not curated here as DKCB8 phenotypes.
Deep research results are used as seeds for research; they do not undergo the same validation as the main records and may contain errors. How we use deep research.
Record notes
Identity. This entry curates the disease OMIM 620133 describes: digenic TYMS-ENOSF1 dyskeratosis congenita. MONDO's causal-gene axiom for MONDO:0859319 names DCLRE1B and is wrong; the discussion above records the evidence and how the error propagated into this repository's curation stub. The entry's filename and `name` retain the MONDO label so the identifier and the record stay aligned. Why there is no DCLRE1B `genetic:` record. A later reviewer will reasonably ask why the disagreement is recorded only in prose, when a `genetic:` entry graded DISPUTED would make it machine-readable - which is what the sibling Embryonal_Rhabdomyosarcoma entry does for the gene MONDO wrongly attributes to it. The answer is the KGX exporter. It hardcodes the predicate to `biolink:contributes_to` and never reads `relationship_type` (monarch-initiative/dismech#12320), so a DISPUTED DCLRE1B record here would export as a plain causative gene-disease edge asserting exactly the relationship this entry says is false. The rhabdomyosarcoma case is different only because the gene there is genuinely reported in that disease, however thinly, whereas DCLRE1B belongs to a different entity altogether. Revisit this once #12320 is fixed. Evidence base. Two clinical references: the eight-family defining cohort and one independent confirmatory case report. Eight families is unusually strong for an ultra-rare entity and is what makes the digenic claim credible; but the individual phenotype detail comes largely from the single case report, and two phenotype rows are graded INDIRECT because their support is a statement about dyskeratosis congenita as a class rather than about these patients. What is graded IN_VITRO and why it matters. The epistatic silencing, the enzyme deficiency, the nucleotide pool imbalance, the genotoxic stress and the drug hypersensitivities were all measured in proband-derived lymphoblastoid cells. That is strong evidence about mechanism and it is not, by itself, evidence about patients. The fluoropyrimidine correction. An earlier version of this entry said in four places that no DKCB8 patient had ever received a fluoropyrimidine and that the avoidance recommendation was purely this entry's inference from cell data. That was wrong. The cached full text of PMID:35931051 reports a proband with a severe adverse response to topical 5-FU during treatment for her squamous carcinoma and melanoma, and the Table 1 footnote for that proband records the same. The claim was written from the abstract and from a deep-research report that makes the same omission, while the full text sat in the same pull request. The entry now carries that observation as HUMAN_CLINICAL evidence on the pathophysiology node and on the treatment, and the associated discussion asks the narrower question that is genuinely open: whether a topical reaction in one patient extends to systemic dosing. Phenotype frequencies are graded from Table 1 of the defining cohort, not from the disease class. That matters most for bone marrow failure, which is definitional for dyskeratosis congenita and appears in only 1 of 10 probands here; see that row for the two readings the data cannot separate. Leukoplakia is likewise 4 of 10 rather than near-universal, while hair loss is 9 of 10, and immunoglobulin deficiency is present in 6 of 10 despite not being part of the classical triad. One confounder worth carrying forward. The only individually described patient also has ring chromosome 18 with partial 18p and 18q monosomy. His poor growth, feeding difficulties, facial features and strabismus are attributed by his own authors to the 18p monosomy and are deliberately not curated here as DKCB8 phenotypes.
Review round 1: correct the false no-patient-exposure claim; mine Table 1 · 2026-09-12T12:48:58Z · View source
Addressed the REQUEST_CHANGES review on PR #11719. CRITICAL, and a real error. The entry stated in four places that no DKCB8 patient had ever received a fluoropyrimidine, and built a knowledge gap plus a proposed experiment on that premise. The cached full text of PMID:35931051 reports a proband with 'a severe adverse response to topical 5-FU' during treatment for squamous carcinoma and melanoma of the leg, corroborated by her Table 1 footnote. The claim was written from the abstract and from a deep-research report that makes the same omission, while the full text was already committed in the same PR. Fixed by adding two HUMAN_CLINICAL evidence items (pathophysiology node and treatment), regrading the treatment's supporting evidence into a DIRECT/HUMAN_CLINICAL item plus a separate INDIRECT/IN_VITRO item, and rewriting the treatment description, treatment notes, pathophysiology description, top-level notes, and the discussion so none asserts the observation is absent. The discussion was re-scoped from 'is this untested in patients' to the narrower open question: whether one topical reaction extends to systemic dosing, and whether DPYD screening would flag such a patient. Table 1 of the cached full text had never been mined. Oral leukoplakia regraded VERY_FREQUENT to FREQUENT (4/10). Sparse hair regraded OCCASIONAL to VERY_FREQUENT (9/10, one unknown). Bone marrow failure regraded VERY_FREQUENT/SEVERE to OCCASIONAL (1/10) with the severity qualifier dropped and both readings of the discrepancy stated: a young cohort versus a genuinely marrow-sparing form. Three phenotypes added: decreased circulating immunoglobulin (6/10, HP:0004313), cutaneous malignancy (HP:0008069), abnormal dentition (HP:0000164). Each new frequency carries the Table 1 row as a DIRECT evidence item. Ontology correction. pathophysiology#Genotoxic Stress and Defective Transcription was bound to GO:0006284 base-excision repair, which describes a plausible mechanism rather than the one measured. Rebound to GO:0006974 DNA damage response, with the ATM/CHK1/CHK2/p53/p21 phosphorylation sentence added as the supporting measurement. Telomerase framing corrected. The entry claimed the telomere machinery is intact. The paper reports reduced telomerase activity, accumulating oligoadenylated immature TERC, and reduced DKC1, TERT and SMUG1; only shelterin is unchanged. Rewritten to shelterin-intact with the authors' own downstream-of-nucleotide-metabolism reading, and the shelterin sentence added as evidence. Node renamed. pathophysiology#Fluoropyrimidine and Antifolate Hypersensitivity to #Antimetabolite Hypersensitivity, because antifolate hypersensitivity is unsupported by either source while hydroxyurea hypersensitivity is reported. Hydroxyurea evidence added. The treatment renamed correspondingly. All intra-entry references updated. Management added from the GeneReviews cache, which contains zero TYMS mentions and is therefore cited only for class-level telomere-biology-disorder management, graded INDIRECT: malignancy surveillance and androgen therapy / hematopoietic cell transplantation. A top-level references block was added, with the GeneReviews entry tagged. Validation: just validate passes, 42/42 snippets verified (was 25/25). check-duplicate-keys, check-entity-refs, check-causal-targets, check-qualifier-terms, check-enum-values all OK.
Create: DKCB8 / TYMS-ENOSF1 digenic dyskeratosis congenita (MONDO:0859319) · 2026-09-11T22:12:35Z · View source
De-novo curation from claim issue #11715. The curation began on the wrong gene: MONDO asserts MONDO:0859319 RO:0004003 HGNC:17641 (DCLRE1B), just enrich-stubs copied that into the stub, and the claim issue repeated it. The OpenScientist deep-research report came back entirely about TYMS-ENOSF1; just preflight-dr flagged the conflict correctly (DCLRE1B mentioned 0 times, TYMS 70) and pointed at the OMIM 620133 xref as the independent identity anchor, which resolves it in favour of TYMS-ENOSF1. Verified against the primary sources (Tummala 2022 AJHG, eight families; Shams 2025 confirmatory case) before rewriting. Curated the digenic architecture explicitly with HP:0010984 and relationship_type COOPERATING on ENOSF1, since the ENOSF1 haplotype is pathogenic only in trans to a TYMS loss-of-function allele. Ordered the causal chain nucleotide-supply-first (TYMS deficiency -> dNTP pool imbalance -> genotoxic stress -> abnormal telomere maintenance) because that ordering is what distinguishes this from the telomerase and shelterin forms it is clinically indistinguishable from. The 5-fluorouracil hypersensitivity is graded IN_VITRO and INDIRECT throughout and the avoidance recommendation is labelled as this entry's inference, since no patient has been reported to receive a fluoropyrimidine. Validated: linkml-validate, linkml-term-validator, 25/25 snippets verified, check-duplicate-keys, check-entity-refs, check-causal-targets, check-qualifier-terms, validate-history.
Disease: Dyskeratosis Congenita Autosomal Recessive 8 (DKCB8) MONDO ID: MONDO:0859319 · OMIM: 620133 · Category: Mendelian (telomere biology disorder) Causal locus: TYMS–ENOSF1 (digenic), chromosome 18p11.32
Dyskeratosis Congenita, Autosomal Recessive 8 (DKCB8) is an ultra-rare, childhood-onset inherited bone marrow failure and telomere biology disorder (TBD) defined by a distinctive digenic genetic architecture at the TYMS–ENOSF1 locus. Unlike classic single-gene recessive dyskeratosis congenita, DKCB8 arises when an individual inherits a loss-of-function coding variant in TYMS (thymidylate synthase) from one parent and a specific haplotype with rare variants in the antisense regulator ENOSF1 (enolase superfamily 1) from the other parent. Because ENOSF1 post-transcriptionally silences the remaining wild-type TYMS allele, the net effect is severe thymidylate synthase deficiency — even though neither parent alone is affected and one parent carries an entirely wild-type TYMS coding sequence. This "pseudo-recessive" inheritance was established by Tummala et al. (Blood, 2022) across eight independent DC families and confirmed by an independent 2025 case report [PMID: 35931051; PMID: 40207375].
Mechanistically, thymidylate synthase catalyzes the sole de novo route to dTMP (dUMP → dTMP, using 5,10-methylenetetrahydrofolate). Its deficiency in DKCB8 depletes the dTMP/dTTP pool and distorts the balance of cellular deoxyribonucleotides, promoting uracil misincorporation into DNA, base-excision-repair–mediated strand breaks, replication-fork collapse, and genotoxic stress, which in turn produce abnormal telomere maintenance and stem-cell attrition. The clinical consequence is the classic mucocutaneous triad of dyskeratosis congenita (reticulate skin pigmentation, nail dystrophy, oral leukoplakia) together with progressive bone marrow failure, pulmonary and hepatic fibrosis, and an elevated risk of myelodysplastic syndrome, leukemia, and squamous cell carcinoma.
Clinically, DKCB8 is diagnosed and managed as part of the broader dyskeratosis congenita / TBD spectrum: very short telomeres (flow-FISH below the 1st percentile) provide the screening biomarker, and targeted sequencing of the TYMS–ENOSF1 locus (rather than standard single-gene panels) is required to capture the digenic lesion. Management centers on androgens (danazol/oxymetholone) for cytopenias and fludarabine-based reduced-intensity allogeneic hematopoietic stem cell transplantation (HSCT) as the only cure for marrow failure, while non-hematopoietic complications (pulmonary/hepatic fibrosis, malignancy) remain the principal drivers of late mortality. A mechanistically-inferred, DKCB8-specific caution is that patient cells are hypersensitive to fluoropyrimidines (5-fluorouracil, capecitabine) and antifolates, which target thymidylate synthase — these agents should be avoided.
Overview. DKCB8 is a Mendelian, autosomal recessive (digenic) subtype of dyskeratosis congenita, itself a prototypical telomere biology disorder. Dyskeratosis congenita is a progressive bone-marrow-failure syndrome classically presenting with the ectodermal/mucocutaneous triad of reticulate skin pigmentation, nail dystrophy, and oral leukoplakia, with a wide spectrum of multisystem complications and cancer predisposition [PMID: 42625322; PMID: 35097237].
"Classically, it presents with the ectodermal triad of reticulate skin pigmentation, nail dystrophy, and oral leukoplakia" — [PMID: 42625322]
Key identifiers. Cross-references retrieved from EBI OLS4 / MONDO (Finding F008):
| Resource | Identifier |
|---|---|
| MONDO | MONDO:0859319 ("dyskeratosis congenita, autosomal recessive 8") |
| OMIM | 620133 (equivalentTo) |
| GARD | 0026695 |
| MedGen | C1824030 |
| UMLS | C5774257 |
| Abbreviation | DKCB8 |
| Parent concept | dyskeratosis congenita (MONDO:0015780; MeSH D019871; Orphanet ORPHA:1775) |
The causal locus per OMIM 620133 is TYMS (with ENOSF1), referencing Tummala et al. 2022 [PMID: 35931051].
Synonyms / alternative names. "DKCB8"; "dyskeratosis congenita, autosomal recessive 8"; within the literature, "TYMS-ENOSF1 dyskeratosis congenita" and "thymidylate synthase deficiency dyskeratosis congenita" [PMID: 40207375].
Source of information. The DKCB8 entry is derived from aggregated disease-level resources (OMIM, MONDO) built on a small number of individual-patient reports — the original eight-family cohort [PMID: 35931051] and subsequent single case reports [PMID: 40207375]. There is no EHR-scale dataset for this ultra-rare entity.
Primary cause — genetic, digenic. DKCB8 is caused by germline digenic TYMS–ENOSF1 variants producing thymidylate synthase deficiency (Finding F001). Tummala et al. identified heterozygous germline TYMS variants across eight independent DC families; crucially, in each family one parent carried a wild-type TYMS coding sequence, while the other transmitted a specific ENOSF1 haplotype plus rare ENOSF1 variants:
"we have identified a remarkable series of heterozygous germline variants in the gene encoding thymidylate synthase (TYMS). Although the inheritance appeared to be autosomal recessive, one parent in each family had a wild-type TYMS coding sequence. Targeted genomic sequencing identified a specific haplotype and rare variants in the naturally occurring TYMS antisense regulator ENOSF1 (enolase super family 1) inherited from the other parent" — [PMID: 35931051]
An independent 2025 report confirmed the mechanism:
"compound heterozygosity for loss of function variants in TYMS and a specific haplotype of its antisense regulator ENOSFI (enolase super family 1) causes digenic DC" — [PMID: 40207375]
Genetic risk factors. The obligate risk determinants are (i) a TYMS loss-of-function coding allele and (ii) a permissive ENOSF1 antisense haplotype carrying rare variants in trans. Both are required; neither alone is sufficient (Findings F001, F007).
Environmental risk factors. No environmental cause is required for disease. However, a clinically important gene–environment interaction exists: because thymidylate synthase is the pharmacologic target of fluoropyrimidines, DKCB8 confers hypersensitivity to 5-fluorouracil and folate-antagonist chemotherapy (Finding F009):
"hypersensitivity to the TYMS-specific inhibitor 5-fluorouracil" — [PMID: 35931051]
TYMS uses 5,10-methylenetetrahydrofolate as cofactor, linking enzyme activity to dietary folate / one-carbon metabolism [PMID: 37183313; PMID: 28461497]. This provides a plausible (though not clinically demonstrated for DKCB8) axis of modifiable risk.
Protective factors. No specific protective variants or environmental protective factors have been reported for DKCB8. Avoidance of fluoropyrimidine/antifolate exposure is a mechanistically-inferred protective action (not a natural protective factor).
Gene–environment interaction. The principal, evidence-based GxE relationship is pharmacogenomic (TYMS deficiency × fluoropyrimidine/antifolate exposure), detailed in Sections 5 and 12 [PMID: 35931051; PMID: 27569869; PMID: 22496803].
The phenotypic spectrum of DKCB8 mirrors classic dyskeratosis congenita (Finding F003). Reported features, with suggested HPO terms:
| Phenotype | Type | HPO term (suggested) | Notes / frequency |
|---|---|---|---|
| Reticulate skin hyperpigmentation | Physical manifestation | HP:0007441 (reticulate skin pigmentation) | Core triad; childhood onset |
| Nail dystrophy | Physical manifestation | HP:0008404 (nail dystrophy) | Core triad |
| Oral leukoplakia | Clinical sign | HP:0002745 (oral leukoplakia) | Core triad; may progress to SCC |
| Diffuse hyperpigmentation + punctate hypopigmented macules | Physical manifestation | HP:0007441 / HP:0001010 | Documented in confirmed DKCB8 case |
| Sparse hair | Physical manifestation | HP:0008070 (sparse hair) | Reported in DKCB8 case |
| Bone marrow failure / cytopenias | Laboratory abnormality | HP:0005528; HP:0001903; HP:0001873 | Often first–second decade; drives mortality |
| Poor growth / failure to thrive | Clinical sign | HP:0001508 (failure to thrive) | Reported in DKCB8 case |
| Feeding difficulties | Symptom | HP:0011968 (feeding difficulties) | Reported in DKCB8 case |
| Strabismus | Clinical sign | HP:0000486 (strabismus) | Reported in DKCB8 case |
| Oral lichenoid lesions | Clinical sign | HP:0030955 / related | Broader DC oral spectrum |
| Pulmonary fibrosis | Physical manifestation | HP:0002206 (pulmonary fibrosis) | Adult / late; major late-mortality driver |
| Pulmonary arteriovenous malformations | Physical manifestation | HP:0002638 / related | DC complication |
| Liver fibrosis | Physical manifestation | HP:0001395 (hepatic fibrosis) | Frequently subclinical |
| GI telangiectasias | Physical manifestation | HP:0004389 / related | AR/XLR-predominant |
Onset, severity, progression. Mucocutaneous features typically appear in childhood; marrow failure often follows in the first–second decade; pulmonary and hepatic fibrosis and malignancy are later complications. The course is progressive and multisystem (Findings F003, F011).
Specific documentation in a genetically confirmed DKCB8 patient:
"he developed diffuse hyperpigmentation as well as numerous punctate hypopigmented macules, sparse hair, and nail dystrophy, and diagnosis of DC was confirmed with a telomere length assay" — [PMID: 40207375]
Oral involvement extends beyond leukoplakia to lichenoid lesions (reticular, plaque, erosive-ulcerative), and one DC patient developed tongue squamous cell carcinoma at age 25 [PMID: 42625322].
Quality-of-life impact. No DKCB8-specific EQ-5D/SF-36 data exist. By extension from DC/TBD: marrow failure imposes transfusion dependence and infection risk; pulmonary and hepatic fibrosis impair function and survival; malignancy risk requires lifelong surveillance. Impaired reproductive function has been documented in DC (reduced anti-Müllerian hormone, oocyte yield, and fertilization/euploidy rates) [PMID: 32405899].
Causal genes. - TYMS — thymidylate synthase (HGNC:12441; OMIM 188350), chromosome 18p11.32. Catalyzes dUMP → dTMP. - ENOSF1 — enolase superfamily member 1 (HGNC:24338), the natural antisense regulator of TYMS, reversely oriented and overlapping (Finding F007).
Pathogenic variants and classification. In DKCB8 the operative lesions are (i) a loss-of-function TYMS coding variant (the affected 2022 cohort carried heterozygous germline TYMS variants) and (ii) rare variants on a specific ENOSF1 haplotype in trans [PMID: 35931051]. A distinct DKCB8 case arose from a structural lesion: a TYMS deletion within a ring chromosome 18 (partial 18p/18q monosomy) combined with the ENOSF1 haplotype [PMID: 40207375]. Thus variant classes span point loss-of-function, structural deletion, and regulatory-haplotype variation. Formal ACMG/AMP classification of individual DKCB8 alleles is not standardized because the digenic architecture falls outside conventional single-gene rules.
Functional consequence. Net loss of function of thymidylate synthase activity. The key mechanistic twist is epistatic post-transcriptional silencing: elevated ENOSF1 suppresses the remaining wild-type TYMS allele (Finding F007):
"post-transcriptional epistatic silencing of TYMS is occurring via elevated ENOSF1" — [PMID: 35931051]
"TYMS expression is regulated by its antisense mRNA, ENOSF1. Disrupted regulation may promote uncontrolled DNA synthesis" — [PMID: 30134598]
Allele frequency. DKCB8-causing configurations are ultra-rare; the ENOSF1 haplotype-based mechanism means population allele frequencies of individual SNPs do not straightforwardly predict disease. The TYMS–ENOSF1 region is well studied pharmacogenomically (e.g., 5′-UTR VNTR; rs495139; rs3819102) [PMID: 22496803; PMID: 30134598; PMID: 35631247].
Somatic vs germline. The DKCB8 lesions are germline. (Somatic clonal hematopoiesis can arise secondarily in DC marrow — see Sections 6 and 11 [PMID: 32736377].)
Modifier genes / epigenetics. ENOSF1 itself functions as the principal modifier via antisense regulation. No additional DKCB8-specific modifier genes or DNA-methylation signatures have been reported.
Chromosomal abnormalities. A ring chromosome 18 with partial 18p/18q monosomy encompassing TYMS has been reported as one route to DKCB8 [PMID: 40207375].
Environmental factors. No toxin, radiation, or occupational exposure is required for DKCB8. The clinically relevant environmental interaction is pharmacologic: fluoropyrimidines (5-FU, capecitabine, FdUMP metabolites) and antifolates (raltitrexed, methotrexate) inhibit thymidylate synthase and would be expected to compound the pre-existing deficiency (Finding F009) [PMID: 35931051; PMID: 27569869; PMID: 15930305; PMID: 25245820].
Lifestyle factors. Dietary folate / one-carbon metabolism is mechanistically linked because TYMS uses 5,10-methylenetetrahydrofolate; folate status modulates thymidylate biosynthesis and genome integrity in model systems [PMID: 37183313; PMID: 28461497]. Whether folate supplementation modifies DKCB8 severity is untested. As with all DC/TBD, smoking is a general risk factor for squamous carcinogenesis and should be avoided.
Infectious agents. Not applicable — DKCB8 is a genetic disorder with no infectious etiology. (Recurrent infections occur secondary to marrow failure/immune dysfunction.)
BER intermediates and thymineless stress lead to replication-fork stalling/collapse and DNA double-strand breaks, activating homologous recombination (RAD51), RPA2 and γ-H2AX → results in genotoxic replication stress. (inferred from thymidylate-stress models) [PMID: 18773878; PMID: 25245820; PMID: 2839770; PMID: 15930305]
"Thymidylate deprivation increases dUTP and uracil in DNA, which is removed by base excision repair (BER)" — [PMID: 18773878]
Genotoxic stress and defective transcription lead to abnormal telomere maintenance → results in accelerated telomere attrition and replicative senescence in high-turnover stem/progenitor compartments. (demonstrated) [PMID: 35931051]
"These defects in the nucleotide metabolism pathway resulted in genotoxic stress, defective transcription, and abnormal telomere maintenance" — [PMID: 35931051]
Stem-cell attrition in bone marrow and epithelia leads to the clinical phenotype: bone marrow failure and the mucocutaneous triad; over time → pulmonary/hepatic fibrosis. (demonstrated clinically) [PMID: 34852175; PMID: 31754622]
Cell types (CL suggestions): hematopoietic stem cell (CL:0000037), common myeloid progenitor (CL:0000049), keratinocyte (CL:0000312), oral mucosal epithelial cell, hepatic stellate cell (CL:0000632, fibrosis), type II pneumocyte (CL:0002063).
Organ level. - Primary: bone marrow / hematopoietic system (UBERON:0002371, bone marrow), skin (UBERON:0002097), nails, oral mucosa (UBERON:0003729). - Secondary: lungs (pulmonary fibrosis, AVMs; UBERON:0002048), liver (fibrosis; UBERON:0002107), gastrointestinal tract (telangiectasias; UBERON:0001555), eyes (strabismus; UBERON:0000970). - Body systems: hematopoietic/immune, integumentary, respiratory, digestive/hepatobiliary.
Tissue and cell level. Predominantly rapidly proliferating epithelial and hematopoietic tissues. Affected cell populations (CL): hematopoietic stem/progenitor cells (CL:0000037), keratinocytes (CL:0000312), oral mucosal epithelium, hepatic stellate cells and pulmonary fibroblasts (fibrotic remodeling), type II pneumocytes (CL:0002063).
Subcellular level (GO Cellular Component). Nucleus (GO:0005634) and cytoplasm (GO:0005737, site of thymidylate synthase and dNTP synthesis); telomeric chromosomal ends (GO:0000781, chromosome, telomeric region).
Localization / lateralization. Skin pigmentation is typically diffuse/reticulate; oral leukoplakia and lichenoid lesions affect the tongue and buccal mucosa; pulmonary and hepatic fibrosis are bilateral/diffuse. No consistent lateralization.
Onset. Congenital predisposition with childhood-onset clinical manifestations. Mucocutaneous features usually appear first (childhood), followed by marrow failure (Finding F011). The reported DKCB8 ring-18 case presented in early childhood [PMID: 40207375].
Progression. Chronic, progressive, multisystem. Typical trajectory: mucocutaneous triad (childhood) → bone marrow failure (first–second decade) → pulmonary/hepatic fibrosis and malignancy (later). Age-at-diagnosis data for DC/TBD: median 19.4 years (range 0–71.6) in the NCI cohort [PMID: 34852175] and 9 years in the Canadian pediatric registry [PMID: 42267950].
"median age at diagnosis 19.4 years [range 0 to 71.6]" — [PMID: 34852175]
Patterns. No spontaneous remission. Marrow failure can be transiently stabilized by androgens and cured (hematologically) by HSCT. Critical windows: early recognition of marrow failure to time HSCT before severe non-hematopoietic organ damage accrues; avoidance of TYMS-inhibiting chemotherapy at all times.
Epidemiology. DKCB8 is ultra-rare: originally 8 independent families [PMID: 35931051] plus rare subsequent case reports [PMID: 40207375]. Dyskeratosis congenita overall is estimated at ~1 per million (Finding F011).
Inheritance. Autosomal recessive but digenic — a TYMS coding LOF allele plus a trans ENOSF1 antisense haplotype. Classic single-gene recessive segregation is not observed; one parent carries a wild-type TYMS coding sequence [PMID: 35931051].
"In a cohort of eight independent DC-affected families, we have identified a remarkable series of heterozygous germline variants in the gene encoding thymidylate synthase (TYMS)" — [PMID: 35931051]
Penetrance / expressivity. Presumed high penetrance when both genetic requirements are met; expressivity is variable, consistent with the broader DC/TBD spectrum. Formal penetrance estimates are unavailable given the small case count.
Anticipation / mosaicism / founder effects. Not established for DKCB8. (Anticipation is a general feature of telomere biology disorders due to progressive telomere shortening across generations but has not been specifically quantified for DKCB8.)
Consanguinity / carrier frequency. Not specifically reported for DKCB8; the digenic mechanism complicates carrier-frequency estimation. The permissive ENOSF1 haplotype is common in the population, whereas the TYMS LOF allele is rare — so disease requires the specific trans combination.
Population demographics. No defined ethnic predilection reported. Sex ratio not established. Within DC/TBD broadly, autosomal-recessive/X-linked forms tend to present earlier and more severely (relevant to DKCB8) [PMID: 34852175].
Diagnostic strategy (Finding F006). Two-step: (1) telomere length screening, then (2) molecular confirmation.
Telomere length. Measured by flow-FISH; very short telomeres (<1st percentile of age-matched controls, or age-modified thresholds such as <6.5 kb in patients >40 y) are the key screening biomarker.
"TL was considered suspicious once below the 10th percentile of normal individuals (standard screening) or if below 6.5 kb in patients >40 years (extended screening). In cases with shortened TL, next generation sequencing (NGS) for TBD-associated genes was performed" — [PMID: 37096215]
For the confirmed DKCB8 case, a telomere length assay confirmed DC and the TYMS deletion was identified genetically [PMID: 40207375].
Genetic testing. WES/NGS with segregation analysis is standard for DC/TBD [PMID: 42557666]. Because DKCB8 is digenic (TYMS coding + ENOSF1 antisense haplotype in trans), standard single-gene panels may miss it — targeted genomic sequencing of the TYMS–ENOSF1 locus is required.
"Targeted genomic sequencing identified a specific haplotype and rare variants in the naturally occurring TYMS antisense regulator ENOSF1" — [PMID: 35931051]
Chromosomal microarray / karyotyping is warranted when a structural lesion (e.g., ring chromosome 18) is suspected [PMID: 40207375].
Ancillary testing / organ surveillance. - Liver: transient elastography detects subclinical fibrosis in ~88.8% of TBD patients [PMID: 34565437]. - Lung: restrictive spirometry and reduced DLCO in 42% of DC patients [PMID: 31754622]. - Marrow: CBC, bone marrow aspirate/biopsy for cytopenias and MDS surveillance. - In vitro corroboration: patient lymphoblastoid cells show TYMS deficiency, altered dNTP pools, and 5-FU hypersensitivity [PMID: 35931051].
Clinical criteria / differential diagnosis. Diagnosis rests on the mucocutaneous triad + marrow failure + very short telomeres + molecular confirmation. Differential: other DC genotypes (DKC1, TERT, TERC, RTEL1, TINF2, PARN), Fanconi anemia, Shwachman–Diamond syndrome, Hoyeraal–Hreidarsson syndrome, and acquired aplastic anemia [PMID: 35605178; PMID: 36091172; PMID: 37507252].
Screening. Cascade telomere-length + targeted molecular testing of at-risk relatives; no newborn screening exists for this ultra-rare entity [PMID: 36286734].
Survival / mortality (DC/TBD context; Finding F005). In 231 DC/TBD individuals (NCI IBMFS study): 42% deceased, median overall survival 52.8 years (95% CI 45.5–57.6); transplant-free median survival 45.3 years (95% CI 37.4–52.1).
"42% of patients were deceased with a median overall survival (OS) of 52.8 years (95% confidence interval [CI] 45.5-57.6)" — [PMID: 34852175]
AR/XLR forms (relevant to DKCB8) carry the worst prognosis:
"Severe bone marrow failure (BMF), severe liver disease, and gastrointestinal telangiectasias were more prevalent in AR/XLR or TINF2 disease... After adjusting for age at DC/TBD diagnosis, we observed the highest cancer risk in AR/XLR individuals" — [PMID: 34852175]
Cancer risk. Increased risk of MDS, AML, and solid tumors — especially head/neck and anogenital squamous cell carcinoma; clonal hematopoiesis contributes to leukemia risk [PMID: 36091172; PMID: 32736377].
Morbidity / disease course. Complications include marrow failure, pulmonary fibrosis, pulmonary AVMs, liver fibrosis, hepatopulmonary syndrome, and GI telangiectasias [PMID: 40356079; PMID: 31754622; PMID: 34565437]. Even after curative HSCT for marrow failure, late mortality from pulmonary and hepatic fibrosis remains high (e.g., 4/7 died at median 10 years post-HSCT) [PMID: 40356079].
Prognostic factors. Younger age at diagnosis and severe BMF predict worse outcome (pediatric registry: severe BMF associated HR 7.5 for mortality) [PMID: 42267950]. Baseline PFT abnormalities predict pulmonary outcomes [PMID: 31754622].
Treatment-related prognostic note. Androgen therapy improves cytopenias but creates an atherogenic lipoprotein profile (↓HDL-C, HDL particle number/size; ↑LDL-C, apoB; all p<0.001), warranting cardiovascular monitoring [PMID: 34929494].
No disease-specific/curative therapy exists for the underlying TYMS deficiency; management follows DC/TBD principles (Finding F004).
Pharmacotherapy — androgens (first-line for cytopenias). Danazol/oxymetholone can improve hematologic parameters (NCIT: androgen therapy C1516; danazol C494; oxymetholone C716).
"Although hematological defects can respond to danazol/oxymetholone, the only current curative treatment for these is hematopoietic stem cell transplantation (HSCT) using fludarabine-based conditioning protocols" — [PMID: 35929966]
Monitor lipids/cardiovascular risk on androgens [PMID: 34929494].
Hematopoietic stem cell transplantation (only cure for marrow failure). Fludarabine-based reduced-intensity conditioning (RIC) — often with alemtuzumab, minimizing/avoiding radiation and alkylators — is standard because of mucosal, vascular, pulmonary, and hepatic fragility (NCIT: hematopoietic stem cell transplantation C15431; fludarabine C1094).
"Because of toxicity after myeloablative conditioning, RIC is becoming standard for HCT in DKC. These results suggest that RIC regimen is feasible and safe for patients with DKC and does not accelerate pulmonary damage in the short-to-medium term after HCT" — [PMID: 34086408]
A prospective single-arm trial found TBI is dispensable for DC/TBD-associated marrow failure [PMID: 39002862]. HSCT does not correct non-hematopoietic complications, and late pulmonary/hepatic fibrosis remains the leading cause of post-HSCT mortality [PMID: 40356079].
Pharmacogenomics — a DKCB8-specific caution. Because thymidylate synthase is the target of fluoropyrimidines (5-FU, capecitabine) and antifolates, and DKCB8 cells are hypersensitive to 5-FU, these agents should be avoided or used with extreme caution (Finding F009):
"hypersensitivity to the TYMS-specific inhibitor 5-fluorouracil" — [PMID: 35931051]
"variants in genes of the 5-FU metabolic pathway, including TYMS, MTHFR and DPYD also influenced capecitabine efficacy and toxicity" — [PMID: 27569869]
This is a mechanistically-inferred recommendation of high clinical relevance, especially given the elevated malignancy risk in DC (where fluoropyrimidines might otherwise be considered).
Supportive/experimental. Transfusion support, infection prophylaxis, malignancy surveillance, organ-specific management (pulmonary, hepatic). Emerging DC therapeutics under study include PAPD5 inhibitors and other telomere-directed agents (broader DC pipeline) [PMID: 35605178]. Lung transplantation may be considered for pulmonary failure [PMID: 28407835]. No DKCB8-specific gene, cell, or RNA therapy exists.
Primary model (Finding F010): patient-derived lymphoblastoid cell lines (LCLs) — the established DKCB8 experimental system:
"Lymphoblastoid cells from affected probands have severe TYMS deficiency, altered cellular deoxyribonucleotide triphosphate pools, and hypersensitivity to the TYMS-specific inhibitor 5-fluorouracil" — [PMID: 35931051]
These cells recapitulate TYMS deficiency, dNTP-pool imbalance, 5-FU hypersensitivity, genotoxic stress, and abnormal telomere maintenance, and gene-rescue experiments confirmed ENOSF1-mediated silencing of TYMS.
Related/supporting models. - Chemically induced thymidylate stress: mouse FM3A cells and 5-FdUrd/raltitrexed models reproduce thymineless DNA damage [PMID: 2839770; PMID: 15930305; PMID: 25245820]. - Drosophila gene–nutrient (vitamin B6 / SHMT / TS) models link one-carbon metabolism to genome integrity [PMID: 37183313]. - Broader DC telomere biology has been modeled in mice and zebrafish for other DC genes (e.g., dkc1), but no TYMS-specific animal model of DKCB8 exists.
Limitations. No animal model captures the human-specific digenic TYMS–ENOSF1 antisense mechanism; LCLs do not reproduce tissue-level fibrosis or malignancy. Development of a humanized TYMS–ENOSF1 model is an open need.
[TYMS LOF coding/deletion allele] [ENOSF1 permissive haplotype + rare variants]
(parent A) (parent B, in trans)
\ /
\____________ DIGENIC GENOTYPE _________/
|
Elevated ENOSF1 antisense activity
|
Post-transcriptional silencing of WT TYMS
|
SEVERE THYMIDYLATE SYNTHASE DEFICIENCY
|
dUMP up / dTMP down , imbalanced dNTP pool
|
Uracil misincorporation into DNA -> BER strand breaks
|
Replication-fork collapse, gamma-H2AX/RPA2, HR (RAD51)
|
GENOTOXIC STRESS + defective transcription
|
ABNORMAL TELOMERE MAINTENANCE -> stem-cell attrition
/ | \
Bone marrow failure Mucocutaneous triad Pulmonary/hepatic fibrosis
\ /
Clonal hematopoiesis -> MDS/AML
Epithelial dysplasia -> SCC (head/neck, anogenital)
Upstream drivers are the digenic lesion → TYMS deficiency → nucleotide imbalance; downstream effects are telomere dysfunction → multisystem stem-cell failure and malignancy. The pharmacogenomic branch (fluoropyrimidine/antifolate hypersensitivity) intersects the upstream node directly.
| PMID | Study (abbrev.) | Role in this report |
|---|---|---|
| 35931051 | Germline thymidylate synthase deficiency… causes DC | Foundational — digenic TYMS–ENOSF1 mechanism, LCL model, 5-FU hypersensitivity, telomere defect (F001, F002, F007, F008, F009, F010, F011) |
| 40207375 | TYMS-ENOSF1 DC in ring chromosome 18 | Independent confirmation; structural-deletion route; phenotype (F001, F003) |
| 18773878 | DNA damage/HR from thymidylate deprivation | Uracil/BER/strand-break step (F002) |
| 25245820 | Raltitrexed/TS inhibition, DNA damage | Thymidylate-stress DNA-damage mechanism (F002) |
| 30134598 | rs495139 in TYMS-ENOSF1 region | ENOSF1 antisense regulation of TYMS (F007) |
| 42625322 | Oral lichenoid lesions in DC | Mucocutaneous triad; oral spectrum; tongue SCC (F003) |
| 35929966 | Biology and management of DC | Androgens + fludarabine-based HSCT only cure (F004) |
| 34086408 | RIC-based HCT for DC | RIC standard; feasible/safe (F004) |
| 39002862 | RIC without radiation trial | TBI dispensable for DC/TBD BMF (F004) |
| 40356079 | Late complications post-HSCT in DC | High late mortality from PF/LF (F004, F005) |
| 34852175 | Disease progression/outcomes in TBD | AR/XLR worst prognosis; survival; age at diagnosis (F005, F011) |
| 36091172 | FA and DC/TBD genomic instability | Cancer spectrum (F005) |
| 32736377 | Clonal hematopoiesis in IBMFS | Clonal hematopoiesis → leukemia risk (F005) |
| 34565437 | Transient elastography in cryptic DC | Subclinical liver fibrosis 88.8% (F006) |
| 31754622 | PFTs in DC | Restrictive/DLCO abnormalities 42% (F006) |
| 37096215 | Telomere length screening, age-modified | TL-first, then-NGS diagnostic strategy (F006) |
| 42557666 | TERT-associated DC characterization | WES + segregation + flow-FISH paradigm (F006) |
| 27569869 | Pharmacogenetics of capecitabine | TYMS as fluoropyrimidine PGx determinant (F009) |
| 22496803 | TYMS genetic region polymorphisms | TYMS pharmacogenetics (F009) |
| 34929494 | Lipoprotein alterations from androgens in DC | Androgen cardiovascular risk (F005) |
| 42267950 | Canadian Inherited Marrow Failure Registry — DC | Pediatric outcomes; median dx age 9 y; severe BMF HR 7.5 (F011) |
| 37183313 | B6/SHMT gene-nutrient interaction (Drosophila) | One-carbon/folate link (F009) |
| 2839770 | Chromosomal DNA degradation from thymidylate stress | Conserved thymineless-death model (F002) |
Report compiled from 11 confirmed findings across 5 investigation iterations and 44 reviewed papers. Evidence source types are indicated inline: human clinical (patient cohorts/case reports), in vitro (patient LCLs), model organism (mouse/Drosophila thymidylate-stress systems), and computational/database (MONDO, OMIM cross-references).
Checked with linkml-reference-validator 0.2.1.
| Outcome | Count |
|---|---|
| References checked | 33 |
| Resolved | 33 |
| Unresolved (possible confabulation) | 0 |
| Unverifiable | 0 |
| Quoted claims checked | 21 |
| Quoted claims found in source | 19 |
| Quoted claims not found in source | 2 |
| References weighed for topical relevance | 33 |
| On topic | 7 |
| Off topic | 0 |
Searched the abstract, any retrieved full text, and the title. A quote drawn from a part of the paper that was not retrieved will appear here too, so check before treating one as invented:
Every one of these was searched against an abstract alone, with no full text retrieved - marked abstract only below. Where full text can be fetched, re-running with it will settle them; where the source publishes only a summary to PubMed, as GeneReviews chapters do, it will not, and the quote has to be checked by hand against the chapter itself.
PMID:40207375 (abstract only): "thymidylate synthase deficiency dyskeratosis congenita"PMID:34852175 (abstract only): "42% of patients were deceased with a median overall survival (OS) of 52.8 years (95% confidence interval [CI] 45.5-57.6)"Checked with linkml-term-validator 0.4.5, through the ols: adapter.
| Outcome | Count |
|---|---|
| Terms checked | 42 |
| Resolved | 39 |
| Unresolved (possible confabulation) | 0 |
| Obsolete | 0 |
| Unverifiable | 3 |
| Terms whose name was checked | 16 |
| Terms named correctly | 15 |
| Terms named as a different term | 1 |
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:
HP:0007441 (2 mentions) - the report calls it "reticulate skin pigmentation"; HP calls it Hyperpigmented/hypopigmented maculesTerms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: ORPHA.