A Dog Blood Test That Reads Several Diseases at Once, Without a Sequencer

One blood draw, multiple diseases, over 90% accuracy, and no genomic sequencing involved. A peer-reviewed study out of Oxford lays down an early marker in a race that IDEXX and PetDx are already running with a very different technology.

LatusPet, a University of Oxford biotech startup, published clinical evidence on July 1 that its platform can screen dogs for cancer and cardiovascular disease and assess overall health from a single sample. The science is promising. The harder question is who gets it to the exam room.

LatusPet's SINO platform screens dogs for cancer and heart disease from one sample

The study, "Application of NMR-Based Metabolomics and Machine Learning for Non-Invasive Disease Screening in Dogs," appeared in Frontiers in Veterinary Science (DOI: 10.3389/fvets.2026.1830153). It is the first published evidence that blood NMR metabolomic analysis combined with machine learning can screen companion animals for several diseases at once.

The proprietary platform, called SINO, reads the chemical signature in a dog's blood and reports a diagnostic accuracy (ROC AUC) above 90% for the conditions tested. The work was led by Prof. Riccardo Finotello of Link University in Rome, a veterinary oncologist, alongside Oxford scientists and Italian clinical partners.

The design deserves attention. Rather than the usual shortcut of comparing sick dogs against healthy ones, the researchers separated each condition from a realistic mix of other diseases and healthy animals together. That is a harder, more honest test, and closer to what a vet actually sees. The study ran across 139 dogs at Italian veterinary centres. The company says it plans to add conditions each year and extend the platform to other species.

LatusPet is pre-commercial. Founder Dr Bobo Nazarov framed the milestone plainly: the science is proven, and the mission now is getting the test into vets' hands. The company is expanding to larger cohorts and actively seeking veterinary partners.

Why multiplex screening threatens per-test lab economics

Reference labs make money one analyte at a time. A wellness panel is a bundle of discrete tests, each with its own line item. A single blood draw that reads cancer risk, cardiovascular disease, and general health from one sample collapses that model into one screen, which is either a threat to per-test revenue or a premium upsell, depending entirely on how it is priced and sold.

The technology choice sharpens the point. The best-known blood test for canine cancer, PetDx's OncoK9, uses cell-free DNA and next-generation sequencing to detect cancer signals, and reaches vets through a distribution deal with IDEXX. It is cancer-focused and sequencing-dependent, which carries a cost and turnaround profile that comes with running an NGS pipeline.

LatusPet is taking a different road. NMR metabolomics does not require a sequencer, which points toward lower per-sample cost and a simpler lab workflow, and its pitch is breadth: multiple disease categories from one sample rather than a single cancer readout. If those cost and breadth advantages hold up at scale, the addressable use case is broader than cancer screening. It is the annual wellness visit itself.

That is the operator angle. For clinic groups, an affordable multiplex screen bundled into routine check-ups could become a new margin-positive service line and a reason to pull wellness visits forward. For the diagnostic incumbents who own the in-clinic and reference-lab stack, a credible metabolomics challenger is a reason to watch the category, not dismiss it.

Restraint is warranted. A total of 139 dogs at Italian centres is a proof of concept, not a commercial validation. Accuracy on a curated research cohort is not the same as performance across breeds, geographies, and the messy baseline of everyday practice.

What LatusPet must prove before vets adopt it

Three gates decide whether this becomes a product or a paper. First, cohort scale and diversity: hundreds of dogs across multiple countries and breeds, not one region. Second, the condition roadmap and species extension the company has promised, which is what turns a single study into a platform. Third, and most decisive, distribution.

The science is the easy part. OncoK9 needed IDEXX to reach the vet channel, because the vet relationship is the moat, and it is owned by IDEXX, Antech, and Zoetis. LatusPet says it is seeking partners, which is the polite version of the hardest problem it faces. Whoever controls the blood draw controls adoption.

Watch for a named distribution or lab partner, a larger multi-country validation, and any signal on price versus a standard wellness panel. Those, not the AUC number, will determine whether a dog's next routine check-up quietly starts screening for diseases before the symptoms show.

This piece discusses veterinary diagnostic research. It is industry analysis, not veterinary or medical guidance.

Source: Blood Test Screens Dogs for Multiple Diseases at Once, Peer-Reviewed Study Shows, via PR Newswire