The dismech unit is one disease-level CMML page, not separate pages for each ontology subclass.
disease_term is anchored to MONDO:0020311 and paired with NCIT:C3178 at mapping level.
Subtypes are modeled as flat facet axes rather than separate disorders.
The first subtype axis is blast percentage: CMML-0, CMML-1, and CMML-2.
The second subtype axis is leukocyte-count phenotype: myelodysplastic CMML and myeloproliferative CMML.
NCIT subtype mappings were added because MONDO currently does not provide corresponding CMML subtype classes for this slice.
The subtype mappings are ontology-grounding only and do not imply separate dismech pages or Not Yet Curated placeholders.
Key literature used in curation
PMID:38450850 for disease overview, revised diagnostic criteria, MP-CMML versus MD-CMML split, and high-level treatment framing.
PMID:34985762 for mutation frequencies, monocytosis plus marrow dysplasia, and hypomethylating-agent response rates.
PMID:35377828 for IDO-positive dendritic-cell aggregates, regulatory T-cell expansion, and T-cell exhaustion in the marrow microenvironment.
PMID:40644618 for RAS-pathway mutation frequency, proliferative phenotype, and transformation risk.
PMID:36455187 for randomized phase III DACOTA treatment data in advanced proliferative CMML.
PMID:33039516 for higher-risk transplant benefit.
PMID:37096522 for CMML-0/1/2 and MD-CMML/MP-CMML clinical subtype usage plus common presenting phenotypes.
PMID:37223910 for GM-CSF hypersensitivity as a CMML hallmark.
Atomic pathophysiology framing
Clonal hematopoietic stem-cell transformation
Early epigenetic and splicing-gene lesions
Persistent monocytosis
RAS pathway-driven proliferative phenotype
IDO-positive dendritic-cell aggregates
Immune tolerance and T-cell exhaustion
AML transformation propensity
Notes
CMML-0/1/2 and MD-CMML/MP-CMML are kept in one file because they reflect subtype facets within a shared CMML causal program rather than fully distinct disease mechanisms.
The curation emphasizes PMID-backed abstract quotes for validator-friendly evidence.