Comparative oncology / research surface

Melanoma: the dog as a model of human mucosal disease.

Canine melanoma is predominantly oral and mucosal, not sun-driven, which makes it a natural model of human MUCOSAL melanoma, not the common cutaneous form. Both are non-BRAF, low-mutation, and copy-number-driven, and a single cross-species study sequenced both.

These are somatic tumor alterations, not a germline carrier status. Every number here is a cohort frequency, the fraction of sequenced tumors somatically altered in a gene, reported by a published study. It is not a variant a dog inherits or carries, and it is not a prediction about any individual dog. This surface reports where the biology is shared between the species, model-of, and abstains where a cohort did not report a gene.

This is the molecular side. For how often these cancers actually strike goldens over a lifetime, see the Golden Retriever Lifetime Study cohort → · all cancers →

The conserved core

shared · 5 genes

Driver genes somatically altered in the tumors of both species, the evidence that the dog models the human disease.

NRAS

RAS/MAPK oncogene; the clean shared driver

Dog

11%

of 65 tumors · SNV (Q61, G12)

Human

18%

of 67 tumors · SNV

TP53

tumor suppressor

Dog

of 65 tumors · SNV

Human

of 67 tumors · SNV

MDM2

p53-axis oncogene (amplification); the conserved copy-number driver

Dog

29%

of 37 tumors · focal amplification (35% within the mucosal subset; CDK4 co-amplified)

Human

—

not encoded (see note)

The conserved copy-number driver. Rather than a single MDM2 %, human mucosal melanoma carries a broader amplification axis: ~70% of tumors have CDK4 / CCND1 / CDKN2A-axis alterations (Newell 2019). Copy-number-driven biology, not a BRAF/MAPK one.

PTPRJ

receptor tyrosine phosphatase; tumor suppressor

Dog

19%

of 37 tumors · inactivating / truncating (23% within the mucosal subset)

Human

—

not encoded (see note)

Wong 2019 identifies PTPRJ as a driver shared across canine oral and human mucosal melanoma; the exact human fraction was not extractable from the primary table, so the human cell abstains.

BRAF

MAPK oncogene; the integrity marker

Dog

of 65 tumors · SNV (effectively negative)

Human

16%

of 67 tumors · SNV

The key integrity point: BRAF drives ~50% of human CUTANEOUS melanoma but only ~16% of mucosal and ~0 to 3% of canine oral melanoma. Both diseases here are non-BRAF and non-UV, which is exactly why canine oral melanoma models human mucosal, not cutaneous, melanoma.

Where the cohorts differ

Reported as a recurrent driver in one species' cohort but not the other. An honest asymmetry, a genuine non-report shown as such, never filled with a zero.

KIT · Human-side

15% of 67

receptor tyrosine kinase; a human therapeutic target · SNV + amplification

A human-mucosal-enriched driver and drug target, largely absent in canine oral melanoma (0% in Wong, ~8% in Hendricks) -- a genuine cross-species divergence, stated not papered over.

SF3B1 · Human-side

12% of 67

splicing factor · SNV (R625 hotspot)

A human-mucosal-specific splicing driver (R625 hotspot), absent in the canine cohorts -- a divergence.

NF1 · Human-side

16% of 67

RAS GTPase-activating tumor suppressor · SNV / loss

Recurrent in human mucosal melanoma; not a reported recurrent canine driver.

The human landscape

The shared genome-biology signature is the real model-of evidence: both diseases are non-UV, low-mutation-burden (~2 to 3 mutations/Mb), and copy-number / structural-variant-driven, unlike sun-driven cutaneous melanoma (which is ~15 to 30+ mut/Mb and ~50% BRAF-driven). That is exactly why the honest human comparator is mucosal, not cutaneous, melanoma (Wong 2019).