A canine biological-age study linked routine blood scores with health and survival. Read the findings, study design and validation limits.
In 2025, researchers at Nestlé Research published “A biological age based on common clinical markers predicts health trajectory and mortality risk in dogs” in GeroScience. An October 16, 2026 report from dvm360 described how the team applied the tool to stored records from the Purina Life Span Study.
Lean-fed Labrador retrievers had a lower estimated biological age relative to their chronological age after age 7 when researchers applied the tool to the feeding-study records. According to dvm360, that difference appeared years before survival differed between the groups. The central longevity finding concerns timing: an age-like blood score separated the groups before their survival records did. This makes the score a candidate early measure to investigate alongside the feeding study’s longer-term health outcomes.
Study setting: Animal models, specifically dogs.
The research brought together two distinct sets of records. The team developed and tested its algorithm using more than 12 years of bloodwork and health histories from 829 dogs at a pet care center. According to dvm360, the algorithm combines 10 measures from routine complete blood count and serum-chemistry panels.
The Purina Life Span Study supplied the smaller feeding-study dataset. It followed 48 Labrador retrievers from seven litters. That distinction matters: 829 refers to the dogs whose records supported algorithm development and testing rather than the number enrolled in the feeding experiment.
The Purina Life Span Study also appeared in an October 12, 2026 retrospective carried by Yahoo Finance. Its role in the biological-age research was specific: it supplied a historical feeding comparison with recorded survival and health outcomes.
These datasets address related but different questions. The larger dataset examines whether a combined blood score relates to health trajectory and mortality. The smaller dataset examines whether that score distinguishes animals with different lifelong feeding histories and outcomes. Keeping those questions separate avoids attributing the feeding trial’s randomized design to the entire algorithm-development dataset.
In the 829-dog dataset, each additional year by which estimated biological age exceeded chronological age was associated with a 75% higher risk of death. The age gap therefore connected a blood-based estimate with an observed survival outcome.
Another finding concerned individual laboratory reference ranges. According to dvm360, nearly half of the dogs whose estimated biological age was more than one year ahead of chronological age had no more than one individual marker outside its reference range. The combined score could therefore draw attention to patterns that were not represented by a long list of individually abnormal results.
That is the rationale for the authors’ proposed clinical role: using information across measurements to identify a health trajectory that warrants closer attention. The relevant research question is whether this combined interpretation adds useful information to an existing assessment.
For readers comparing aging scores with health and survival outcomes, the important distinction is between producing an age-like number and demonstrating what that number contributes. Here, the reported contribution was an association with mortality and separation between feeding groups before their survival diverged.
The original Labrador study established its feeding groups early in life. Dogs were paired by sex and body weight at six weeks. They were then randomly assigned to control or lean-fed groups, with feeding beginning at eight weeks.
As reported by dvm360, lean-fed dogs received 75% of the food their control-fed partners had eaten the previous day. Control dogs initially ate free choice. Their portions were fixed when they were 3.25 years old to prevent serious obesity.
The groups consequently represented different feeding histories rather than a brief dietary intervention introduced near the end of life. Their body-condition measurements also differed. Across ages six to 12, control-fed dogs averaged a score of 6.7 on a nine-point scale compared with 4.6 in the lean-fed group.
Those details help define the comparison behind the biological-age result. Researchers were examining records from animals that had experienced a sustained difference in feeding conditions. The blood-score analysis added another way to describe the trajectories already documented in that experiment.
The lean-fed group lived a median 1.8 years longer than controls. It also reached a median age of 12.0 before long-term treatment for a chronic condition began, compared with 9.9 years in the control group. These figures come from the underlying feeding study as reported by dvm360.
Joint health provided another recorded difference. At age eight, 10% of lean-fed dogs and 77% of control-fed dogs had osteoarthritis in two or more joint types. This placed the later biological-age analysis alongside outcomes involving both survival and chronic health problems.
The value of the historical dataset is that the researchers could compare the score with those longer-term records. They could ask when the score began distinguishing the groups in relation to when survival differed. That ordering is the notable feature of the new analysis rather than simply the observation that the groups had different average scores.
The findings need to be read with several study-specific limitations:
For pet owners, the practical discussion belongs with a veterinarian: routine laboratory results and body condition should be considered alongside diet and any concerning changes. The feeding quantities described above are experimental conditions rather than a feeding plan for an individual animal.
For readers following longevity science, the useful approach is to keep the study’s components separate. Evaluate the blood-score association as a measurement finding and the randomized feeding comparison as evidence about that particular cohort. This preserves the contribution of each without turning them into a broader treatment claim.
The next logical step is independent testing in broader dog populations to determine whether the score adds useful information beyond standard veterinary assessment and supports better clinical decisions.
Weighing a dog blood-score result before treating it as support for a longevity intervention is where AgeAmaze’s research-led explanations become useful. Its coverage gives research-minded adults context for assessing overstated biological-age and therapy claims by making the underlying science and the limits of the evidence easier to understand.
Stay current with research on aging biology, biomarkers, nutrition, therapeutics, peptides and longevity technology. AgeAmaze reports what the evidence shows, where uncertainty remains and which claims still need stronger data.
Follow AgeAmaze for careful reporting on what longevity science can show today and what still needs stronger evidence.
read the Blog
