I've read the ten cached references for HLD14. Next I'm confirming identifiers and the mouse-model papers online, then I'll write the report.

Term lookups are done. Last, I'm checking the NCIT treatment and CHEBI drug IDs in the local caches before writing.

Hypomyelinating Leukodystrophy 14 (HLD14; UFM1-related H-ABC): research report

Prepared: 2026-09-25, for the dismech entry kb/disorders/Hypomyelinating_Leukodystrophy_14.yaml. Most quotes come from the cached references in references_cache/, which are PubMed abstracts or PMC full texts. The rest come from database pages, and those sources are named where used.

Sources. Ten PubMed/PMC items were read in full or as abstracts: PMID:28931644, 29868776, 34573312, 35189806, 39470296, 39846712, 41731076, 42195294, 30626644 and 40315331. Web lookups covered MONDO (via OLS), HGNC REST, Orphanet, ClinicalTrials.gov and PubMed.

How identifiers were checked. Every ontology CURIE below was read from a lookup, never written from memory. Most came from the repo caches (cache/hp, cl, go, uberon, ncit, chebi). UFM1 and UFC1 came from the HGNC REST API, and the MONDO cross-references from OLS. A term marked [lookup needed] had no cached match and has not been verified; it must be searched before it is bound.

Summary of evidence limits

1. The oligodendrocyte defect is inferred. In the only functional test of the founder promoter deletion, the variant reduced promoter activity in neuroblastoma and astroglioma cells. It did not reduce it in the oligodendrocytoma line (PMID:28931644). No study has measured UFM1 or UFMylation in patient oligodendrocytes, or in patient cells carrying the promoter deletion.
2. No human neuropathology exists for UFM1-related disease.
3. Studies disagree on ER stress. Overexpressing UFM1 p.R81C in cell lines did not change ER-stress markers (PMID:29868776). UFM1-knockout mouse neurons do activate the PERK arm of the unfolded protein response (PMID:41731076).
4. One paper assigns a UFC1 variant to "HLD14". PMID:39846712 describes UFC1 p.Arg23Gln as "HLD14-related". OMIM 617899 is the UFM1 disorder; the UFC1 disease is a separate entry. Treat that paper as evidence about the shared pathway, not as evidence on HLD14 itself.
5. It is unconfirmed whether the R81C missense patients are HLD14. The p.Arg81Cys cases (PMID:29868776) had no basal-ganglia abnormality (0/4) but did have delayed myelination (3/4). Before merging them into the entry, check whether OMIM 617899 lists them. A has_subtypes split, or a note, may be the better fit.

1. Disease information

Overview. HLD14 is an autosomal recessive, infantile-onset hypomyelinating leukodystrophy caused by biallelic UFM1 variants. UFM1 is the ubiquitin-like modifier of the UFMylation pathway. Almost all reported patients are homozygous for a 3-bp deletion in the UFM1 promoter, a founder variant in the Roma population. On MRI the disease meets the criteria for hypomyelination with atrophy of the basal ganglia and cerebellum (H-ABC), and patients sit at the most severe end of that spectrum.

"Sixteen patients from 14 families from different countries fulfilling the MRI criteria for H-ABC exhibited a similar, severe clinical phenotype, including lack of development and a severe epileptic encephalopathy." (PMID:28931644)

"Hypomyelinating leukodystrophy type 14 (MIM # 617899) is caused by mutations in the UFM1 gene and is inherited in an autosomal recessive manner. It is characterized by hypotonia, an almost complete lack of motor or cognitive skills, absent language development, spasticity, and intractable seizures." (PMID:34573312)

Identifiers

Synonyms: HLD14; UFM1-related H-ABC; recessive H-ABC (Hamilton et al. title: "recessive variant of H-ABC"); leukodystrophy, hypomyelinating, 14.

Data provenance: aggregated case series and cohorts. There are no EHR-derived data.

2. Etiology

• Cause. Biallelic hypomorphic variants in UFM1 (hgnc:20597; 13q13.3; OMIM gene 610553). The disease is monogenic.
• Genetic risk.
• Roma ancestry, which carries the founder allele.
• Parental consanguinity or endogamy: "Parental consanguinity was reported in 7 families and 2 families were related" (PMID:28931644).
• Environmental risk factors: none described.
• Protective factors: none described. No modifier genes have been reported.
• Gene–environment interaction: none described.

3. Phenotypes

Onset is early infantile, with medical attention drawn between the neonatal period and about 6 months. Across all cohorts the course is progressive and usually fatal in infancy or early childhood.

Frequency sources. Hamilton 2017 (n=16; frequencies from Nahorski 2018 Table 2), Szűcs 2021 (n=4), Ivanov 2023 (n=9) and Drobňaková 2026 (n=17). Drobňaková states that "All patients presented with" its listed signs.

Quality of life. No EQ-5D, SF-36 or PROMIS data have been published. Patients are totally dependent, many with a tracheostomy, gastrostomy or ventilation.

CSF. One Szűcs patient had low CSF protein (138 mg/L). This is a single observation and should not be curated as a phenotype.

4. Genetic and molecular information

Causal gene: UFM1 (hgnc:20597; Entrez 51569; ENSG00000120686; UniProt P61960). The protein is a 9.1-kDa type I ubiquitin-like modifier.

Pathogenic variants

• The two transcript numberings name the same variant. Szűcs states that the NM_016617.4 c.-155_-153delTCA was "reported as NM_001286704.1: c.-273_-271delTCA" (PMID:34573312). Pick one form for the KB and give the other in notes.
• Classification.
• Szűcs 2021 called it likely pathogenic under ACMG.
• Kaur 2025 classified it "P (PS3, PS4_M, PM2, PP5)" (PMID:39470296).
• Drobňaková 2026 reports it as pathogenic in ClinVar. Confirm the current ClinVar status before citing.
• Population frequency. Absent from gnomAD (PMID:34573312). Among Roma: "Screening of 670 Roma controls revealed 30 carriers with an overall carrier rate of 4.5% ... An additional panel of Eastern Slovak Roma samples revealed a carrier frequency of 3.3% (9 out of 273) ... a carrier rate of approximately 25% (14 out of 57 individuals)" in one endogamous community (PMID:28931644).
• Linkage. "LOD score calculations for the UFM1 variant showed a maximum LOD score of 9.18" (PMID:28931644).
• Functional consequence. Both variants are partial loss of function (hypomorphic).
• "Our results show a reduction rather than abrogation of ufmylation with the activity of UFM1 and UFC1 mutants at 60–75% of their wild-type counterparts" (PMID:29868776).
• R81C may also have non-loss-of-function effects: "UFM1-deficient and UFM1-R81C-expressing neurons display distinct responses to ER stress, indicating that UFM1-R81C is not merely a loss-of-function variant" (PMID:41731076).
• Suggested functional_impact_category: PARTIAL_LOSS_OF_FUNCTION for both; variant_origin: GERMLINE; zygosity: homozygous.
• Biallelic null variants are depleted. "the conspicuous depletion of biallelic null mutations in the components of this pathway in human genome databases suggest that it is necessary for embryonic survival" (PMID:29868776).
• Modifier genes, epigenetics, chromosomal abnormalities: none reported.
• Other genes in the same pathway (differential diagnosis, not HLD14): UBA5 (E1; hgnc:23230) and UFC1 (E2; hgnc:26941). They give overlapping encephalopathies (PMID:29868776). TUBB4A (hgnc:20774) causes dominant H-ABC, which is HLD6.

5. Environmental information

No environmental, lifestyle or infectious contributors are described. This section is not applicable.

6. Mechanism and pathophysiology

Causal chain

1. A homozygous UFM1 promoter deletion (or the R81C missense variant) leads to reduced UFM1 function. The deletion reduces UFM1 transcription, measured in neural cell lines only: "The deletion significantly reduced promoter activity in SY-5Y and U373 but not in HeLa and HOG-F2 cell lines" (PMID:28931644). R81C instead weakens UFM1's binding to UBA5 and its thioester transfer to UFC1 (PMID:29868776).
2. Caveat: HOG-F2 is the oligodendroglial line, so a neuron- or astrocyte-selective effect is what was shown. Reduced UFM1 in patient tissue has not been measured for the promoter variant.
3. Reduced UFM1 results in reduced protein UFMylation (GO:0071569 protein ufmylation, DECREASED). This was shown for R81C in patient lymphoblasts: "the level of two UFM1-conjugates with cellular proteins in patient-derived lymphoblasts was significantly reduced" (PMID:29868776). For the promoter variant it is inferred.
4. Reduced UFMylation leads to loss of ribosomal uL24/RPL26 UFMylation at the ER. This step is inferred from cell biology in non-neural cells. "a largely uncharacterized ribosomal protein, RPL26, is the principal target of UFM1 conjugation ... UFMylated RPL26 and the UFMylation machinery are in close proximity to the SEC61 translocon" (PMID:30626644; HEK/K562 cells). Branches:
5. 3a. Failed ER ribosome-associated quality control (ER-RQC) at stalled translocon-bound 60S subunits: "ER-RQC differs from cytosolic RQC by requiring covalent conjugation of the ubiquitin-like protein UFM1 (UFMylation) to uL24" (PMID:40315331). Arrest peptides are then released into the ER lumen instead of being degraded: "In the absence of RQC or UFMylation machinery, ER-APs are not properly extracted into the cytosol for degradation but are instead released into the ER lumen" (PMID:40315331; HEK293). See also PMID:36945571 (PNAS 2023) and PMID:31595041 (Cell Res 2020); both were found by search and are not yet cached.
6. 3b. ER stress with PERK-branch UPR activation and reduced global translation in neurons. This was demonstrated in mouse Ufm1-KO neurons: "UFM1 loss is associated with induction of ER stress, activation of the unfolded protein response (UPR) pathway, and reduced protein translation" (PMID:41731076). The paper shows PERK specifically, with ATF6 and IRE1α unchanged. Suggested GO terms: GO:0036499 PERK-mediated unfolded protein response (INCREASED); GO:0006412 translation (DECREASED).
   • Conflicting evidence: overexpressing R81C in HeLa or SH-SY5Y cells did not change CHOP/BiP or tunicamycin-induced apoptosis (PMID:29868776).
7. Steps 3a/3b lead to impaired neuronal development and synapse formation, demonstrated in mouse. "UFM1-deficiency confounds neuron development and synapse function" (PMID:41731076). Dendrite complexity (GO:0016358) and synapse number (GO:0007416) were reduced, excitatory transmission was reduced, and R81C rescued these only partially.
8. Selective neuron loss. This is demonstrated in mouse; in humans it is inferred from MRI. "CNS-specific knockout of Ufm1 in mice caused neonatal death with microcephaly and apoptosis of neurons in specific brain regions" (PMID:28931644, citing Muona 2016, PMID:27545674). GO:0051402 neuron apoptotic process (INCREASED).
9. In patients this appears as progressive atrophy of the caudate, putamen, cerebellar vermis and cortex, and as postnatal progressive microcephaly.
10. The authors infer apoptosis from imaging: "signal abnormality of the lateral part of the head of the caudate nucleus suggestive of local apoptosis" (PMID:28931644).
11. Hypomyelination, the step with the weakest evidence. The mechanism is unresolved: it may be a primary oligodendrocyte defect or secondary to neuronal and axonal failure.
12. The only oligodendroglial data come from a UFC1 R23Q variant in the mouse FBD-102b line. There, mutant UFC1 aggregated in lysosomes, Akt phosphorylation fell, and MBP/PLP1 expression and morphological differentiation were reduced (PMID:39846712; IN_VITRO, a different gene).
13. Suggested GO terms: GO:0022010 central nervous system myelination (DECREASED); GO:0048709 oligodendrocyte differentiation (DECREASED, and label it hypothesis).
14. Steps 4–6 result in the clinical picture:
15. basal-ganglia and cortical failure → dystonia, opisthotonus, spastic tetraparesis;
16. cortical dysfunction → epileptic encephalopathy and absent development;
17. brainstem and bulbar involvement → inspiratory stridor, dysphagia, bradypnea and apnea, leading to respiratory insufficiency and death (clinical inference);
18. sensory pathway failure → visual and hearing loss.

Branch-point summary

Cell types (CL, cache-verified): - neuron CL:0000540 - medium spiny neuron CL:1001474 (striatum; inferred from caudate/putamen atrophy) - oligodendrocyte CL:0000128 - oligodendrocyte precursor cell CL:0002453 - astrocyte CL:0000127 (the promoter effect was seen in the U373 astroglioma line) - Purkinje cell CL:0000121 and cerebellar granule cell CL:0001031 are possible but not demonstrated.

Other GO terms: GO:0071566 UFM1 activating enzyme activity (UBA5 step). Terms for ribosome-associated quality control, rescue of stalled ribosomes, and UFM1 conjugating/transferase activity are [lookup needed].

Omics. Bulk RNA-seq of Ufm1-KO mouse neurons found 539 differentially expressed genes. Synaptic genes went down, and translation and RNA-processing genes went up (PMID:41731076). There are no human transcriptomic, proteomic, metabolomic, single-cell or spatial data.

7. Anatomical structures affected

• Primary organ: brain (UBERON:0000955) and CNS (UBERON:0001017).
• Structures on imaging:
• cerebral white matter (white matter UBERON:0002316)
• putamen (UBERON:0001874)
• caudate nucleus (UBERON:0001873)
• basal ganglion (UBERON:0002420)
• cerebellum, mainly the vermis (UBERON:0002037)
• corpus callosum (UBERON:0002336)
• cerebral cortex (UBERON:0000956)
• medulla oblongata, thin in one case (UBERON:0001896)
• optic nerves, hypoplastic in one case
• Secondary: larynx (UBERON:0001737), through abnormal tone or innervation producing stridor (PMID:34573312); respiratory control; growth.
• Subcellular: the endoplasmic reticulum membrane (cytosolic face) and ER-bound ribosomes / the 60S subunit at the SEC61 translocon (PMID:30626644, 40315331). GO CC terms: [lookup needed].
• Lateralization: bilateral and symmetric.

8. Temporal development

• Onset: neonatal to early infantile.
• "Presentation is no later than 2 months" (PMID:35189806).
• Median age at first symptoms is 2 months (Hamilton, as summarized in PMID:34573312).
• Szűcs gives an average of 2.5 months (range 1 week–6 months).
• Course: progressive and neurodegenerative. Minimal development is followed by regression at 2–5 months. MRI shows progressive cerebral atrophy and cerebellar atrophy developing over 3–15 months (PMID:28931644). Myelination stays severely deficient, with only slight gains in 3/5 patients on follow-up scans.
• Remission: none.
• Critical period: the prenatal and early postnatal window. Carrier testing and prenatal diagnosis are the only interventions that change outcome.

9. Inheritance and population

• Inheritance: autosomal recessive. Penetrance was assumed complete in the linkage analysis, and segregation was "in perfect agreement".
• Expressivity: fairly uniform and severe for the promoter variant.
• Anticipation and germline mosaicism: not reported.
• Founder effect: a Roma founder haplotype (PMID:28931644) and a separate Sudanese R81C founder haplotype (PMID:29868776).
• Geography: Slovakia, Hungary, Bulgaria, and other European Roma communities. The promoter variant has also been reported in Pakistani and Indian patients (PMID:34573312, 39470296).
• The shared South Asian ancestry of the Roma (PMID:42195294) may explain the Indian and Pakistani cases, but no haplotype study has shown this.
• Carrier frequency: 4.5% across European Roma panels, 3.3% in Eastern Slovak Roma, and about 25% in one endogamous community.
• Prevalence and incidence: no published estimate.
• My own derivation, not a sourced figure: Hardy–Weinberg with 4.5% carriers (q ≈ 0.0225) gives an expected birth prevalence of about 5 per 10,000 among Roma if mating were random. Endogamy would raise it locally.
• For the KB, use prevalence_class: NOT_YET_DOCUMENTED or RARE with measure_type: UNKNOWN, and record the carrier data as CARRIER_FREQUENCY (PMID:28931644).
• Case count: Szűcs counted 23 published patients in 2021. Adding Ivanov's 9 (2023), Kaur's 2 (2025) and Drobňaková's 17 (2026) gives more than 50. That total is my own tally, and the cohorts may overlap.
• Sex ratio: 11 males to 6 females in the Slovak cohort (PMID:42195294) and 1:1 in Szűcs. There is no evidence of a sex effect.

10. Diagnostics

• MRI pattern recognition is the key clinical clue:
• hypomyelination
• a very small or absent putamen
• a small caudate with high T2 signal in the lateral part of the caudate head (distinguishes it from TUBB4A H-ABC)
• cerebellar vermis atrophy
• progressive cerebral atrophy (PMID:28931644)
• MR spectroscopy: "almost equally high peaks of choline, creatine, and N-acetyl-aspartate" in one advanced case (PMID:34573312).
• EEG: diffuse cortical dysfunction and hypsarrhythmia.
• Genetic testing:
• Targeted Sanger testing of the founder variant is fast and suited to Roma infants with the phenotype (PMID:34573312). The Slovak center diagnosed most patients this way, with a mean of 5.7 months from symptom onset to diagnosis (PMID:42195294).
• Exome sequencing can miss the variant: "As the coverage of the promoter region of the UFM1 gene is generally poor in the case of exome sequencing data, and in many cases only the coding region of the genes are analyzed, this pathogenic founder variant could be easily missed" (PMID:34573312). Solo exome sequencing did detect it in two Indian patients (PMID:39470296).
• Genome sequencing covers the promoter.
• Chromosomal microarray, karyotype, FISH, mtDNA and repeat testing are not relevant.
• Clinical rule of thumb: "Roma patients with severe encephalopathy in early infancy with stridor, opisthotonus, bradypnea, severe hearing and visual impairment should be tested for the Roma founder mutation" (PMID:35189806).
• Biomarkers: none validated. UFM1-conjugate immunoblots of lymphoblasts are research-only (PMID:29868776).
• Differential diagnosis:
• TUBB4A H-ABC / HLD6 (dominant, de novo; normal caudate signal)
• UBA5- and UFC1-related encephalopathies (same pathway)
• PCH1B (EXOSC3, also a Roma founder variant, also with stridor)
• Pelizaeus–Merzbacher disease
• POLR3-related leukodystrophy
• other HLDs (PMID:39470296, 42195294)
• Screening: there is no newborn screening. Carrier and cascade testing in Roma communities is recommended (PMID:28931644, 42195294).

11. Outcome and prognosis

• Causes of death: respiratory insufficiency, apnea, cardiopulmonary arrest.
• Morbidity: total dependence; no motor or cognitive milestones; need for tracheostomy, tube feeding or ventilation.
• Recovery potential: none.
• Prognostic factors: none established. The later-surviving Hamilton patients (up to 7 years) suggest some variability.

12. Treatment

No disease-modifying therapy exists. A ClinicalTrials.gov search for "UFM1" (API query, 2026-09-25) returned zero studies.

Supportive care

Preclinical only (not for treatments; better as a discussion or hypothesis): - Trazodone (UPR/PERK inhibitor): "Trazodone, an inhibitor of the UPR, restores protein translation solely in UFM1-R81C-expressing neurons, and increases synapse numbers in both UFM1-KO and UFM1-R81C-expressing neurons" (PMID:41731076; mouse neurons). It did not rescue dendrite complexity. The trazodone CHEBI ID is [lookup needed]. - Hesperetin: restored oligodendroglial differentiation in UFC1-R23Q FBD-102b cells (PMID:39846712). It is a different gene, and the authors themselves call the effect "limited".

There is no gene therapy, ASO or cell therapy, and no pharmacogenomic data.

13. Prevention

• Primary prevention: carrier screening and preconception or prenatal testing in at-risk Roma communities.
• "The identification of the disease-causing variant enables better clinical and genetic counseling, the possibility of prenatal testing, and the option of carrier testing in populations with a high carrier frequency" (PMID:28931644).
• "Prenatal and newborn screening, along with voluntary carrier testing for couples, is essential" (PMID:42195294).
• Secondary prevention: early targeted testing of symptomatic Roma infants shortens the diagnostic delay.
• Tertiary prevention: airway management, nutrition and seizure control.
• Not applicable: immunization, behavioral interventions, prophylaxis.

14. Other species and natural disease

• No naturally occurring UFM1 disease was found. OMIA was not searched directly.
• UFM1 orthologs are conserved across metazoans: "each of its components having a corresponding orthologue in all multicellular organisms" (PMID:29868776). UFMylation is metazoan-specific (PMID:30626644).
• Mouse and human UFM1 proteins are identical (PMID:41731076).
• No zoonotic or transmission relevance.

15. Model organisms

Gaps in models: - No animal or iPSC model carries the founder promoter deletion. - No oligodendrocyte-specific Ufm1 model exists. - No model reproduces hypomyelination together with the basal-ganglia pattern.

In the KB, the modeled_mechanisms links for the mouse cKO should record model_scale and a limitations or SPECIES_MISMATCH/BOUNDARY_OMISSION divergence. Hypomyelination is outside every current model.

Curation notes for the entry

• Mappings:
• disease_term: MONDO:0033486
• Orphanet ORPHA:139441 as skos:broadMatch (H-ABC includes TUBB4A)
• The TUBB4A GeneReviews chapter is not a baseline for this entry
• Gene: hgnc:20597 (UFM1). It is not yet in cache/hgnc/terms.csv, so run just validate-terms after binding it.
• Before binding, look up: drug-resistant epilepsy, bradypnea, tracheostomy (NCIT), trazodone (CHEBI), ER-RQC and ER-membrane GO terms.
• Fetch before citing: PMID:27545674 (Muona), PMID:36945571 and PMID:31595041 (RPL26 ER-RQC). None of the three is cached yet.
• Evidence grading for PMID:34573312: its opening paragraph quoting Hamilton's statistics restates another paper, so mark it quote_role: BACKGROUND. The same applies to the Drobňaková discussion paragraphs that restate Hamilton.

Sources

• Local cache: references_cache/PMID_28931644.md, PMID_29868776.md, PMID_30626644.md, PMID_34573312.md, PMID_35189806.md, PMID_39470296.md, PMID_39846712.md, PMID_40315331.md, PMID_41731076.md, PMID_42195294.md; cache/bookshelf/genereviews.csv
• OLS MONDO:0033486
• HGNC UFM1 · HGNC UFC1
• Orphanet ORPHA:139441 (H-ABC)
• Muona et al. 2016, PMID:27545674 · AJHG full text
• Wang et al. 2020 Cell Res, PMID:31595041 · Scavone et al. 2023 PNAS, PMID:36945571
• Zhang et al. 2022 Mol Neurobiol
• Perdigão et al. 2026 EMBO Mol Med
• ClinicalTrials.gov API query "UFM1"

Reference Validation

Checked with linkml-reference-validator 0.3.0rc1.

Quotes not found in the cited source

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:27545674 (abstract only): "CNS-specific knockout of Ufm1 in mice caused neonatal death with microcephaly and apoptosis of neurons in specific brain regions"
• closest text in source: "Finally, we show that the CNS-specific knockout of Ufm1 in mice causes neonatal death accompanied by microcephaly and apoptosis in specific neurons, further suggesting that the UFM1 system is essential for CNS development and function"

Term Validation

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

Terms the report names something else

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

• DOID:0080296 (1 mention) - the report calls it "DOID"; DOID calls it hypomyelinating leukodystrophy 14
• HP:0001332 (1 mention) - the report calls it "Dystonia, extrapyramidal signs"; HP calls it Dystonia
• CL:1001474 (1 mention) - the report calls it "striatum; inferred from caudate/putamen atrophy"; CL calls it medium spiny neuron
• UBERON:0000955 (1 mention) - the report calls it "Primary organ: brain"; UBERON calls it brain**

Terms whose name is worth a second look

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

• HP:0001272 (1 mention) - the report calls it "Cerebellar atrophy (vermis)"; HP calls it Cerebellar atrophy
• UBERON:0002037 (1 mention) - the report calls it "cerebellum, mainly the vermis"; UBERON calls it cerebellum
• UBERON:0001896 (1 mention) - the report calls it "medulla oblongata, thin in one case"; UBERON calls it medulla oblongata
• UBERON:0001737 (1 mention) - the report calls it "Secondary: larynx"; UBERON calls it larynx**
• CHEBI:2972 (1 mention) - the report calls it "Baclofen for spasticity"; CHEBI calls it baclofen, and lists "baclofeno" among its other names

Prefixes with no resolver

Terms 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.