Predictive Medicine
Challenges and Opportunities for the Pet Food Industry

What is "Predictive Medicine?" Predictive Medicine is a modern medical paradigm incorporating "disease risk assessment" followed by treatments/strategies to offset that estimated "risk." The "genomic era" has ushered in the hope of unraveling the genetic basis of disease. However, the associations between the genes and the ultimate manifestation of many of the most common diseases (for example cancer, diabetes, or hip dysplasia) are complex; meaning not simply the result of single-gene effects. It will likely be decades, if ever, before all the genes and all gene interactions for a particular complex disease are discovered. Yet, almost daily new gene-disease associations are being discovered. These discoveries, while not permitting an absolute knowledge of when or even whether a patient will succumb to a disease, nevertheless enable an estimation of a patient's risk or susceptibility for the disease. Based on this estimated risk, physicians and veterinarians can implement evidence-based "preventive" strategies to offset that risk.

The predictive test need not be a DNA test. It can be any test; radiographic, MRI, ultrasound, CT, blood test, among others, that can be shown to key on factors correlating with the risk or susceptibility for a specific disease. The logic is the same irrespective of mode of testing: knowing the risk, one can implement measures early to offset that risk. In this context, canine hip dysplasia and a hip screening test known as PennHIP serve as an excellent example.

Canine Hip Dysplasia (CHD) is an extremely common genetic disease affecting most breeds of dogs, but particularly large-breed dogs. It affects millions of dogs and leads to osteoarthritis (OA) of the hip causing pain and decreased performance for both pet dogs and for dogs in sport and work. Treating the OA of hip dysplasia is big business. The pharmaceutical industry manufactures several NSAIDs (nonsteroidal anti-inflammatory drugs) to treat the pain of OA and many of the leading pet food companies offer premium, prescription diets having additives (eg. nuetraceuticals) aimed at relieving joint discomfort and slowing the rate of cartilage degradation in osteoarthritic joints. The "traditional paradigm" for treating hip OA is shown in the Illustration 1. The "treatment interval" for administering the medication or ration begins shortly after the onset of the pain and disability of OA and continues as indicated for the life of the dog. In one study of 48 Labrador retrievers kept in a kennel setting, hip pain was not evident and therefore pain medications were not administered until affected dogs were 7 years of age or older, hence the onset of the treatment interval in Illustration 1 at 7 years of age.4

But Predicitve Medicine is the new paradigm. As mentioned previously CHD is a disease of complex inheritance meaning that its expression is influenced by both genetic and environmental factors. A landmark study of diet restriction in Labrador Retrievers demonstrated the profound influence of calorie restriction (an environmental factor) on delaying the onset of OA and lessening its severity.1-5 This study proved definitively that a genetic disease could be mitigated by appropriate environmental manipulation. Surely other forms of environmental manipulation will be discovered to act similarly to offset the risk of OA. Candidate strategies include physical therapy, appropriate exercise regimen, pharmaceuticals, neutraceuticals, and nutrigenomics however, to date there are few studies to document the efficacy of these strategies. Preliminary evidence, however, is encouraging.

The concept of "risk reduction" requires a valid test of disease risk. The PennHIP radiographic test was developed at the University of Pennsylvania in 1983 and a large body of research has been published to validate the test in its ability to determine the "risk" for osteoarthritis at an age as young as 16 weeks.6-27 It is the only test capable of estimating the risk for hip OA. In the lifespan study mentioned above, the PennHIP test of Labs at 2 years of age indicated all dogs to be "at risk" for OA. By the end of the dogs' natural lives 98% had definitive evidence of hip OA (98% positive predictive value). From another study of an entirely different sample of dogs, Illustration 2 is a plot of OA risk for 4 breeds of dogs (mean age 39 months) relative to PennHIP score (0.0-1.1). All 4 breeds of dogs demonstrated a similar pattern of with respect to OA risk: Radiographic OA risk was found to increase as PennHIP score (DI) increased. From earlier research, it was found that this risk for OA could be estimated as young as 16 weeks of age.

The dogs shown in illustration 2 were from the general pet population and no strategies for risk reduction were attempted, however, if diet restriction were imposed on these dogs early in life (say 16 weeks) the evidence in the literature suggests that the OA probability curves would flatten out. That is, fewer dogs at the mean age of 39 months would have expressed radiographic evidence of OA. The PennHIP method, therefore is a powerful tool in determining which dogs are at risk for OA and which dogs could benefit from preventive measures.

The rationale of implementing the "Predictive Medicine" paradigm is shown in Illustration 3. With the knowledge that a dog is at risk for a disease such as OA, one is justified to implement strategies to offset the risk. The net effect to the Pet Food Industry is to lengthen considerably the "treatment interval". The advantages to the dog and the dog owner are obvious: later onset of OA and reduced severity of the disease, a better quality of life.

In conclusion, Predictive Medicine is here now. The veterinarian can assess the risk of hip OA using the PennHIP method and based on this information can prescribe treatments as simple as diet restriction to lower significantly the risk of expressing hip OA.