Comparative oncology / research surface
The dog is a natural model of human cancer.
Dogs get the same cancers we do, driven by the same genes, and unlike a lab mouse they get them naturally, sharing our homes and environment. That makes canine cancer one of the strongest natural models in comparative medicine. Each map below pairs a canine cancer with its human counterpart and shows the somatically-altered driver genes they share, from peer-reviewed cohorts, cited.
Start here
Four paths into the map: a classic model, the golden genesis cancer, a human-rare model, and a signature divergence.
Classic conserved model (shared tumor-suppressor losses)
Golden cohort · genesis cancer for the project
Human-rare cancer with a canine-enriched driver story
Signature honest divergence (BRAF high in dog, rare in human bladder)
Where dog and human diverge
dissonance · 10Both sides measured for a commensurable lesion, and they disagree. The honest limit of the model (INV-81). Not darkness, and not a rate filled with zero.
-
Dog: 87.9% BRAF V595E in canine invasive urothelial carcinoma (Decker 2015) · Human: <1% BRAF V600E in human bladder cancer (Decker 2015)
-
Dog: 2.1% (1/47) in canine HSA (Megquier 2019) · Human: >20% KDR (VEGFR2) alteration in human angiosarcoma (Megquier 2019)
-
Dog: not recurrently mutated across canine HS sequencing cohorts (Takada 2019, Yang 2024, Asada 2023) · Human: 62.5% BRAF V600E (5/8, Go 2014)
-
Dog: 29% exon-11 internal tandem duplication in canine MCT (Montanucci 2024) · Human: human mastocytosis is defined by the kinase-domain exon-17 KIT D816V; canine MCT carries NO exon-17 mutation (Montanucci 2024)
-
Dog: 0% (Wong 2019), ~8% (Hendricks) in canine oral melanoma · Human: 15% in human mucosal melanoma (a driver and drug target)
-
Dog: not a reported recurrent canine driver (Wong 2019) · Human: 16% recurrent in human mucosal melanoma
-
Dog: absent in the canine cohorts (Wong 2019) · Human: 12% human-mucosal-specific R625 splicing driver
-
Dog: not reported recurrently altered in the canine OS cohort (Gardner 2019) · Human: 29% in human OS (Sakthikumar; full text paywalled but the rate is cited)
-
Dog: reported rare / not recurrently mutated in cPAC (Lorch 2019, Mariotti 2014); erbB activated via HER2 instead · Human: 14% in human lung adenocarcinoma (TCGA n=230)
-
Dog: 37.8% (28/74), 93% V659E transmembrane hotspot (Lorch 2019) · Human: 1.7% (25/1478), exon-20 kinase-domain insertions (Arcila 2012)
Concordant drivers live on each cancer page under “the conserved core.” Full instrument framing on Spectra.
All cancers
Bladder Cancer (Invasive Urothelial Carcinoma)
Naturally-occurring canine invasive urothelial carcinoma (also called invasive transitional cell carcinoma) is a recognized model of human invasive bladder cancer, similar in how it looks and how it behaves. There is an honest twist worth stating plainly: the gene that drives it in dogs is not the one that drives it in people. Below is the canine somatic driver, cited, with that divergence made explicit. These are somatic tumor alterations reported as cohort frequencies, a portrait of the disease across dogs, never a germline carrier status and never an individual-dog prediction.
driver landscape by concordance class →Glioma
Canine glioma is a natural model of human glioma, and Amin et al. 2020 showed it resembles human PEDIATRIC glioma most closely, by mutation rate, aneuploidy, and DNA-methylation class (78% classify as pediatric). Both share recurrent drivers across the receptor-tyrosine-kinase, PI3K, and cell-cycle pathways. Below is the canine somatically-altered driver landscape, cited to that cohort. These are somatic tumor alterations reported as cohort frequencies, a portrait of the disease across dogs, never a germline carrier status and never an individual-dog prediction.
driver landscape by concordance class →Hemangiosarcoma
Hemangiosarcoma is one of comparative oncology's most important natural models. Naturally-occurring canine hemangiosarcoma, studied here in golden retrievers, closely mirrors the genomic landscape of human angiosarcoma of the breast and viscera, both converging on the PI3K and MAPK pathways. Below is the shared somatically-altered driver landscape, drawn from a peer-reviewed golden-retriever cohort and cited to its source. These are somatic tumor alterations reported as cohort frequencies, a portrait of the disease across dogs, never a germline carrier status and never an individual-dog prediction.
driver landscape by concordance class →Histiocytic Sarcoma
Canine histiocytic sarcoma is concentrated in specific breeds and reaches MAPK activation through PTPN11 rather than the BRAF that defines the human disease. Human histiocytic sarcoma is very rare, which is exactly why the dog is its premier genetic model. The comparison is convergent pathway, divergent gene.
driver landscape by concordance class →Lymphoma
Canine B-cell lymphoma, especially diffuse large B-cell lymphoma, is one of the most-studied natural models of human DLBCL. The canine driver landscape is unusually well characterized across five cohorts, and it shares its core with the human disease.
driver landscape by concordance class →Mammary Carcinoma
Naturally-occurring canine mammary carcinoma is a clinically similar model of human breast cancer, and here the driver genetics line up too: the PIK3CA H1047R hotspot, one of the defining driver mutations of human breast cancer, is the single most common mutation in canine mammary tumors. Below is the canine somatically-altered driver, cited, with that shared hotspot made explicit. These are somatic tumor alterations reported as cohort frequencies, a portrait of the disease across dogs, never a germline carrier status and never an individual-dog prediction.
driver landscape by concordance class →Mast Cell Tumor
Mast cell tumor is one of the most common skin cancers in dogs, and it is driven by the same gene that drives human mast cell disease, KIT. There is an honest twist worth stating plainly: dogs and people break KIT in different places. Below is the canine driver, cited, with that divergence made explicit. These are somatic tumor alterations reported as cohort frequencies, a portrait of the disease across dogs, never a germline carrier status and never an individual-dog prediction.
driver landscape by concordance class →Melanoma
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.
driver landscape by concordance class →Osteosarcoma
Osteosarcoma is one of the clearest cases of comparative oncology: naturally-occurring canine OS is a conserved genetic model of human, especially pediatric, osteosarcoma. Both are driven by the same core tumor-suppressor losses. Below is the shared somatically-altered driver landscape, drawn from peer-reviewed cohorts and cited to each.
driver landscape by concordance class →Pulmonary Adenocarcinoma
Canine pulmonary adenocarcinoma is a low-mutation-burden, never-smoker-like lung cancer. It reaches the erbB axis through HER2 (the V659E transmembrane hotspot) where human lung adenocarcinoma uses EGFR, a reciprocal-driver divergence. TP53 and KRAS are shared drivers but occur far less often in the dog.
driver landscape by concordance class →The reverse-lookup lane: candidates and gaps
Where a human cancer has known drivers but no canine tumor has been sequenced, we bridge from the human driver to its dog ortholog and name the gap. Candidate models, incomplete, and each gap is a study target the field could close.
This map grows. Each cancer is a single cited data file, added whenever it is worth sharpening the picture, so the map deepens over time without a rebuild. New verticals need a concordance call per driver before they claim agree or diverge (INV-81).
The molecular side pairs with the population side: for how often these cancers strike goldens over a lifetime, see the Golden Retriever Lifetime Study →