Aging in Mice, Worms and Humans: What Transfers

What the animal findings are

Worms. In the roundworm C. elegans, mutations in the gene daf-2 (a relative of the human insulin and insulin-like growth factor receptors) made the worms live more than twice as long, and this needed a second gene, daf-16. The same paper notes that wild worms age and die in under three weeks. [1] These were genetic changes in a short-lived animal, not a treatment given to people.

Mice. The US National Institute on Aging runs the Interventions Testing Program (ITP), which tests compounds in genetically varied mice at several sites. In 2009 it reported that rapamycin, fed from 600 days of age, extended median and maximum lifespan of male and female mice; the lifespan gain based on age at 90% mortality was 14% in females and 9% in males. [2] A review of the first two decades of the program says it tested 54 agents in more than 30,000 mice, and that most compounds that extended lifespan worked mainly or only in male mice. [3] Dose and the age at which treatment began also changed results. [3]

Where results fail to carry over

Even a closely related primate gave mixed answers. Calorie restriction (eating less without malnutrition) extends lifespan in many species. In rhesus monkeys, the University of Wisconsin study reported a survival benefit while the National Institute on Aging study found no significant survival effect. [4] The NIA authors said differences in study design, housing and diet composition may strongly affect the life-prolonging effect. [5] One animal question, two answers, in the same species.

Animal benefit does not guarantee human benefit. A BMJ systematic review compared six treatments with clear results in human trials against animal experiments. For example, corticosteroids showed no benefit in human head injury trials but a benefit in animal models, and tirilazad improved outcomes in animal stroke models but was linked to worse outcomes in patients. [6] The authors said the mismatch may come from bias or from animal models failing to mimic the human disease. [6] That review was not about aging, but the lesson applies.

Sex differences. If mouse benefits depend on sex, dose and start age, a single lifespan headline is hard to apply to any person. [3]

What human trials add

Human trials of calorie restriction measure risk factors, not lifespan. In CALERIE, 218 healthy, non-obese adults aged 21 to 50 were randomly assigned to a 25% calorie restriction diet or usual eating for two years. The restriction group achieved a mean 11.9% cut in intake, and improved blood pressure, cholesterol and C-reactive protein measures. [7] Those are biomarker changes: they do not show that anyone lived longer. A separate safety report from the same trial found bone mineral density was lower in the restriction group, and three participants were withdrawn for safety concerns. [8] This is a good example of why a biomarker gain and an animal lifespan gain are not the same as a health benefit.

How to read a headline about an animal study

• Which model? Worms, mice, monkeys and people are different. Name the species before judging.
• Which outcome? Lifespan, a blood marker and a clinical event are not interchangeable. See Healthspan and Lifespan.
• Which sex, dose and age? Results can depend on all three. [3]
• Has it been repeated, and in humans? Replication and human trials move a finding up the evidence ladder. See Reading Longevity Studies.
• Is a product being sold on it? A mouse paper is not evidence that a supplement works. See Longevity Interventions: What Human Evidence Exists.