Cancer kills dogs. It kills roughly 50% of dogs over 10 years old, and in large breeds like Golden Retrievers, that figure approaches 60%. The problem is not that treatment options are absent — canine oncology has advanced substantially over the past decade. The problem is timing. Most canine cancers are identified when the dog is symptomatic, at a stage where surgical and chemotherapeutic intervention has reduced efficacy and shorter median survival times.
Blood-based biomarker screening does not solve this problem entirely. But it changes the detection window, and that window change is clinically meaningful. Today we are describing the four biomarker signatures added to our canine neoplasia screening panel and the data behind each addition.
What Was Already in the Panel
Our original canine neoplasia module launched with five biomarker signals focused on hemangiosarcoma, lymphoma, and nonspecific neoplastic inflammatory patterns. Those five markers — thymidine kinase 1 (TK1), C-reactive protein, haptoglobin, and two proprietary circulating biomarker ratios — provided a sensitivity of 71% for detecting active neoplasia in dogs with confirmed cancer diagnoses.
Sensitivity of 71% sounds reasonable until you consider what the 29% miss rate means in practice: roughly 3 in 10 dogs with active cancer going home from an annual wellness visit without a flag. That rate drove our expansion work.
The Four New Biomarkers
1. Circulating Tumor DNA (ctDNA) — Methylation Signature
Tumor cells shed DNA fragments into the bloodstream. In dogs with active neoplasia, these fragments carry methylation patterns that differ from normal somatic DNA. We validated a 12-marker methylation signature across 3 tumor types: hemangiosarcoma (n=94 samples), multicentric lymphoma (n=71), and osteosarcoma (n=58).
Sensitivity across the three types ranged from 68% (osteosarcoma, early stage) to 89% (lymphoma, active disease). Specificity across healthy controls was 96.2%. The osteosarcoma sensitivity figure is lower because early osteosarcoma sheds less ctDNA into peripheral blood — it improves to 84% at Stage II and above.
2. Osteopontin
Osteopontin is a phosphoprotein involved in cell adhesion, immune modulation, and — critically — tumor invasion and metastasis. It is elevated in the serum of dogs with osteosarcoma, transitional cell carcinoma, and some forms of mammary gland tumor. Its primary value in our panel is as an orthogonal signal to ctDNA for osteosarcoma detection in large breeds over 5 years old.
When combined with our existing TK1 signal and the new ctDNA methylation marker, sensitivity for osteosarcoma in the validation dataset (Rottweilers, Great Danes, Saint Bernards, and Irish Wolfhounds) reached 81% — up from 52% with TK1 alone.
3. VEGF (Vascular Endothelial Growth Factor) Isoform Ratio
Tumor angiogenesis — the formation of new blood vessels to supply a growing tumor — is driven in part by VEGF signaling. The ratio of VEGF-A to VEGF-C in serum shows tumor-specific patterns that differ from inflammatory-only VEGF elevation. A dog with active hemangiosarcoma or mast cell tumor has a VEGF isoform ratio that is distinguishable from a dog with arthritis-related VEGF elevation.
Adding VEGF isoform ratio to the panel improved hemangiosarcoma sensitivity from 73% to 88% in our validation set. The specificity held at 93%. This is the marker we are most confident in from a clinical standpoint, because hemangiosarcoma is the neoplasm where early detection has the most dramatic impact on survival outcomes.
4. miR-21 and miR-155 (microRNA Signature)
Circulating microRNAs are small non-coding RNA molecules detectable in plasma that carry tumor-specific expression patterns. miR-21 is overexpressed in canine diffuse large B-cell lymphoma and is detectable in plasma before lymph node enlargement becomes palpable in approximately 60% of cases. miR-155 similarly elevates early in some carcinoma types.
The microRNA extraction and quantification step adds processing time — samples require additional preparation compared to standard protein biomarkers. This means our neoplasia panel with miRNA signatures has a slightly longer turnaround: 6-8 hours rather than under 4 hours. For high-risk breed annual wellness panels where the goal is early detection rather than urgent diagnosis, that tradeoff is appropriate.
Combined Panel Performance
With all four additions integrated and the model retrained on the expanded dataset (1,847 dog samples, 412 with confirmed neoplasia diagnoses), overall neoplasia panel sensitivity improved from 71% to 86.4%. Specificity improved from 89% to 93.1% due to the improved discriminative power of multi-marker interaction modeling.
These numbers are from internal validation. External validation through partner clinic prospective enrollment is ongoing, with 6-month follow-up data expected in Q3 2026.
Which Breeds Should Be Prioritized
We recommend annual neoplasia screening for any dog over 5 years in the following high-incidence breeds:
- Golden Retriever, Labrador Retriever (hemangiosarcoma, lymphoma)
- Rottweiler, Great Dane, Saint Bernard, Irish Wolfhound (osteosarcoma)
- Boxer, Bulldog, Boston Terrier (mast cell tumor, brain tumor)
- Bernese Mountain Dog (histiocytic sarcoma — unusually high prevalence, ~25%)
- Scottish Terrier (transitional cell carcinoma — approximately 20x breed average risk)
For other breeds, we suggest annual screening from age 7 onward as routine practice. The incremental cost of adding a neoplasia screen to a wellness panel is currently $28 at partner clinic pricing — modest relative to the outcome difference between Stage I and Stage III diagnosis.