| Domain | Uveal melanoma key entity/fact | Conjunctival melanoma contrast | Suggested ontology term(s) |
|---|---|---|---|
| Disease entity | Uveal melanoma is the main intraocular melanoma in adults; MONDO confirmed as **MONDO:0006486**. Often treated as the dominant subtype within “ocular melanoma,” but biologically distinct from conjunctival melanoma (pqac-00000000, pqac-00000007) | Conjunctival melanoma is an ocular-surface/mucosal melanoma, not a uveal tumor; review evidence emphasizes it is embryologically, biologically, and clinically distinct from UM (pqac-00000002, pqac-00000016) | **MONDO:0006486** uveal melanoma; conjunctival melanoma: **suggest MONDO mapping needed** (do not infer exact ID) |
| Synonym/scope | “Uveal melanoma (UM)”; arises from melanocytes in iris, ciliary body, or choroid (pqac-00000007, pqac-00000014) | “Conjunctival melanoma (Co-M)”; ocular surface melanoma, often grouped historically with ocular melanoma but should be separated in KB design (pqac-00000002, pqac-00000016) | MeSH/ICD/Orphanet exact cross-maps: **suggest curator lookup** |
| Anatomy | Most UM arises from choroid (~90%), then ciliary body (~7%), iris (~2–3%) (pqac-00000014, pqac-00000020) | Usually bulbar conjunctiva near limbus, but can involve any conjunctival region and adjacent tissues (pqac-00000016) | **UBERON:** uvea; choroid; ciliary body; iris; conjunctiva; bulbar conjunctiva; limbus (**exact IDs should be curated if required**) |
| Epidemiology | Incidence ~5–6 per million/year in US/Europe; much higher in fair-skinned/Caucasian populations (pqac-00000020, pqac-00000005) | Incidence ~0.46 per 1,000,000 persons/year; increasing, especially in older adults (pqac-00000002, pqac-00000016) | MONDO:0006486; phenotype annotation may use “adult onset” HPO term |
| Cell of origin | Malignancy of **uveal melanocytes**; early oncogenic events arise in melanocytes of choroid/ciliary body/iris (pqac-00000007, pqac-00000014) | Malignancy of **conjunctival epithelial/basal melanocytes**; often from C-MIN/PAM with atypia (pqac-00000016) | **CL:** melanocyte; conjunctival epithelial cell; immune infiltrates incl. macrophage, T cell (**exact CL IDs to curate**) |
| Initiating gene driver | **GNA11** activating mutation, ~55% in one 2024 summary; mutually exclusive with GNAQ; early/initiating driver (pqac-00000004, pqac-00000015) | GNA11 is not a canonical frequent Co-M driver in recent reviews (pqac-00000016) | HGNC:GNA11; **GO suggestions:** G protein-coupled receptor signaling pathway; MAPK cascade |
| Initiating gene driver | **GNAQ** activating mutation, ~40% in one 2024 summary; with GNA11 accounts for ~85–94% of UM across stages; early driver, not strongly prognostic by itself (pqac-00000004, pqac-00000015) | Not a typical major Co-M driver in current review summaries (pqac-00000016) | HGNC:GNAQ; **GO:** MAPK cascade; phospholipase C-activating GPCR signaling pathway |
| Initiating gene driver | **CYSLTR2** mutation in ~2–4% of UM, usually in GNAQ/GNA11-wild-type tumors; initiating event (pqac-00000015) | Not emphasized as a common Co-M driver in recent clinical reviews (pqac-00000016) | HGNC:CYSLTR2; **GO:** leukotriene signaling / GPCR signaling (**exact process term to curate**) |
| Initiating gene driver | **PLCB4** mutation ~2.5% of UM; activating PLC/PKC/MAPK signaling (pqac-00000015) | Not a defining frequent Co-M driver in 2024 review evidence (pqac-00000016) | HGNC:PLCB4; **GO:** phosphatidylinositol-mediated signaling; protein kinase C signaling |
| Prognostic gene | **BAP1** loss/inactivating mutation: associated with aggressive disease, monosomy 3, high metastatic risk; ~38% primary and ~84% metastatic in one 2024 review summary (pqac-00000004, pqac-00000005) | BAP1 is not the hallmark frequent Co-M driver pattern emphasized in current review summaries (pqac-00000016) | HGNC:BAP1; **GO:** DNA repair; chromatin organization; deubiquitination |
| Prognostic gene | **SF3B1** mutation ~25%; intermediate/later metastasis risk and distinct molecular subgroup (pqac-00000005, pqac-00000015) | Not a headline common Co-M mutation in recent clinical review summaries (pqac-00000016) | HGNC:SF3B1; **GO:** mRNA splicing via spliceosome |
| Prognostic gene | **EIF1AX** mutation ~13%; associated with favorable prognosis and younger patients (pqac-00000004, pqac-00000005) | Not a major defining Co-M driver in recent clinical review summaries (pqac-00000016) | HGNC:EIF1AX; **GO:** translation initiation |
| Chromosomal alteration | **Monosomy 3** strongly linked to poor prognosis/BAP1-mutant disease (pqac-00000005, pqac-00000015, pqac-00000020) | Copy-number variation occurs in Co-M, but chromosome-3-centric prognostic framework is mainly UM-focused (pqac-00000016) | Cytogenetic annotation: monosomy 3 (**formal ontology/NCIt code should be curated**) |
| Chromosomal alteration | **8q gain/amplification** linked to metastatic risk; often with monosomy 3 in poor-risk classes (pqac-00000015, pqac-00000020) | CNVs also occur in Co-M, but specific UM class system is not directly transferable (pqac-00000016) | Cytogenetic annotation: 8q gain (**exact code to curate**) |
| Chromosomal alteration | **6p gain** seen in better-risk UM classes; 6q loss may accompany SF3B1-related structural patterns (pqac-00000005, pqac-00000015) | No analogous standard clinical class scheme highlighted for Co-M (pqac-00000016) | Cytogenetic annotation: 6p gain / 6q loss (**exact code to curate**) |
| Molecular class | DecisionDx-UM/GEP classes used for prognostic stratification: class 1A, 1B, 2 with increasing 5-year metastatic risk; transcriptomic classes 1–4/A–D also used (pqac-00000015) | No comparably established routine prognostic GEP system highlighted in the 2024 Co-M review (pqac-00000016) | **NCIT/diagnostic concept:** gene expression profiling (**exact NCIt term to curate**) |
| Core pathway | Gαq/Gα11 signaling activates **PKC**, **MAPK/ERK**, **PI3K/mTOR** networks driving proliferation and survival (pqac-00000014, pqac-00000015) | Co-M more often resembles cutaneous melanoma genetics, especially UV-related **BRAF/NRAS/NF1** alterations (pqac-00000016) | **GO:** MAPK cascade; PI3K signaling; TOR signaling; cell proliferation |
| Immune microenvironment | UM is immune-privileged/immune-cold; lymphocytic inflammatory phenotype, macrophages, HLA class I/II upregulation and NF-κB activity correlate with poor prognosis (pqac-00000007, pqac-00000014) | Co-M transcriptomic studies show **high PD-L1 expression** and immune-enriched subtypes (pqac-00000016) | **CL:** T cell, CD8-positive T cell, macrophage, endothelial cell; **GO:** immune response, antigen processing/presentation, NF-kappaB signaling |
| Multi-omics/single-cell | scRNA-seq of **37,660 malignant cells from 17 UM tumors** revealed heterogeneous malignant programs and 2 intratumoral subtypes with prognostic/immune differences (pqac-00000013) | Equivalent single-cell evidence base for Co-M is less mature in the retrieved set (pqac-00000016) | **NCIT/assay:** single-cell RNA sequencing (**exact term to curate**) |
| Metastatic tropism | About half of UM patients ultimately metastasize; liver is dominant metastatic site (~89% or more than 90% across sources) (pqac-00000020, pqac-00000021) | Co-M more often spreads first to regional lymph nodes (~25%), but can also involve liver, lungs, brain (pqac-00000016) | **HPO suggestions:** Hepatic metastasis; Lymph node metastasis; Pulmonary metastasis; Brain metastasis (**exact IDs to curate**) |
| Clinical phenotype | Up to ~30% asymptomatic; symptomatic disease can cause visual impairment/vision loss, exudation, retinal detachment; iris melanoma may present with heterochromia and corectopia (pqac-00000003, pqac-00000004) | Visible pigmented or amelanotic conjunctival lesion; may cause sight loss, eye loss, local invasion, disfigurement (pqac-00000016) | **HPO suggestions:** decreased visual acuity; retinal detachment; heterochromia iridis; corectopia; conjunctival pigmentation; amelanotic melanoma (**exact IDs to curate**) |
| Histopathology/prognostic phenotype | Epithelioid or mixed cell type, extra-scleral extension, larger tumor size and chromosome 3/8q abnormalities increase metastatic risk (pqac-00000020) | High postoperative recurrence (33–45%) and lack of standardized therapy are emphasized (pqac-00000016) | **HPO suggestions:** extrascleral extension; recurrent neoplasm; epithelioid morphology (**exact mappings to curate**) |
| Diagnostics | Ophthalmic exam plus ocular imaging and tissue/molecular prognostication; liquid biopsy, ctDNA, extracellular vesicles and AI-assisted methods are active research areas (pqac-00000014, pqac-00000020) | Histopathology is critical; clinical misdiagnosis/late diagnosis remains common; molecular pathology increasingly relevant (pqac-00000016) | **NCIT/assay suggestions:** ultrasonography; fundus photography; biopsy; gene expression profiling; liquid biopsy |
| Prognosis | Historical metastatic median OS about 1 year; 5-year survival overall often 50–70%; metastatic prognosis poor (pqac-00000020, pqac-00000021) | ~27% 5-year disease-specific mortality and recurrence 33–45% in review summary (pqac-00000016) | **HPO suggestions:** reduced life expectancy; recurrent neoplasm; metastasis |
| Local treatment | Plaque brachytherapy and enucleation remain standard local therapies; globe-preserving radiotherapy common (pqac-00000004, pqac-00000020) | Surgical excision ± cryotherapy, topical chemotherapy, brachytherapy, proton/photon radiotherapy; exenteration for advanced invasion (pqac-00000016) | **NCIT suggestions:** Plaque Brachytherapy; Enucleation; Cryosurgery; Topical Chemotherapy; Proton Radiation Therapy; Orbital Exenteration |
| Systemic/metastatic treatment | **Tebentafusp** for HLA-A*02:01-positive unresectable/metastatic UM improved OS: median 21.6 vs 16.9 months; 3-year OS 27% vs 18% (phase 3) (pqac-00000021) | No standard targeted/immunotherapy established; anti-BRAF/anti-MEK/anti-PD(L)1 evidence remains limited and often case-series level (pqac-00000016) | **NCIT suggestions:** Tebentafusp; Pembrolizumab; Ipilimumab; Dacarbazine |
| Tebentafusp toxicity | Common AEs: rash 83%, pyrexia 76%, pruritus 70%, hypotension 38%; discontinuation low (2%) in phase 3 follow-up (pqac-00000021) | Not directly applicable; Co-M systemic therapy toxicities depend on regimen used | **HPO/AE suggestions:** rash; pyrexia; pruritus; hypotension; cytokine release syndrome (**exact IDs to curate**) |
| Liver-directed treatment | Liver-directed therapy remains central for metastatic UM because liver is the dominant metastatic site (pqac-00000020, pqac-00000021) | Co-M metastasis pattern is less liver-dominant than UM and more nodal at presentation of spread (pqac-00000016) | **NCIT suggestions:** Hepatic Perfusion; Radiofrequency Ablation; Embolization; Hepatic-directed Therapy (**exact preferred term to curate**) |
| Current trial example | **Darovasertib (IDE196/LXS196)** neoadjuvant/adjuvant phase 2 for localized UM; PKC inhibitor; outcomes include eye salvage, dose reduction to critical structures, recurrence and metastasis follow-up (NCT05907954) (pqac-00000019) | No matched conjunctival trial in retrieved evidence | **NCIT suggestions:** Darovasertib; Protein Kinase C Inhibitor Therapy |
| Current trial example | **Belzupacap sarotalocan (AU-011 / bel-sar)** phase 3 randomized sham-controlled trial for indeterminate lesions/small choroidal melanoma using suprachoroidal administration plus laser photoactivation (NCT06007690) (pqac-00000017) | Not a conjunctival melanoma protocol | **NCIT suggestions:** Belzupacap sarotalocan; Suprachoroidal Injection; Laser Therapy |
| Current trial example | **Adjuvant melatonin** phase 3 prevention-oriented trial in high-risk primary UM with 5-year metastasis endpoint (NCT05502900) (pqac-00000018) | No analogous Co-M adjuvant prevention trial in retrieved set | **NCIT suggestions:** Melatonin; Adjuvant Therapy |
| Model systems | PDX, zebrafish xenografts, and single-cell functional studies are active UM platforms; zebrafish UM PDX reproduced disseminating UM and enabled drug testing with navitoclax/everolimus (pqac-00000012, pqac-00000013) | No equally developed Co-M preclinical evidence highlighted in retrieved set | **NCIT/model suggestions:** Patient-Derived Xenograft Model; Zebrafish Model; Single-Cell Sequencing |
| Evidence note | UM ontology, molecular classes, and treatment evidence are substantially more mature than for Co-M in the retrieved 2023–2024 literature (pqac-00000015, pqac-00000016, pqac-00000021) | Co-M should be represented as a distinct KB entity with separate genetics, anatomy, and management pathways (pqac-00000002, pqac-00000016) | **Curation note:** exact IDs for uncertain ontology mappings should be validated before production use |


*Table: Compact knowledge-base mapping table contrasting uveal melanoma with conjunctival melanoma across disease identity, genes, anatomy, phenotypes, mechanisms, and treatments. It highlights confirmed and suggested ontology mappings while avoiding invention of uncertain IDs.*