This Fortified Essay is by computational biologist Daniel Munro

While fighting diseases has long been a major end-goal in biological research, the pursuit of life extension more generally has recently been given serious attention and resources. This essay explores to what degree these and other efforts might benefit man’s best friend? When, for example, might dogs live twice as long as they currently do? Here the unique history and biology of companion animals—as well as the current state of research into human life-extension projects—will provide considerations in predicting progress in dog lifespan extension. 

Anti-aging research

Anti-aging research and clinical testing is already underway at biotechnology companies such as Rejuvenate Bio and Loyal, as well as at the NIH-funded Dog Aging Project. Research efforts are being boosted not only by demand for improved pet health, but also because these therapies are stepping stones to similar therapies for humans, helping to fill scientific, financial, and regulatory gaps. Said Harvard biologist George Church, “[dogs are] not just a big organism close to humans. [Age reversal in dogs is] something people will pay for, and the FDA process is much faster. We’ll do dog trials, and that’ll be a product, and that’ll pay for scaling up in human trials.”

The symbiotic nature of longevity research projects for dogs and for humans is suggestive of a world where major avenues of research currently underway or speculated for humans will be applied to dogs too. Disease research in general will contribute to increases in life expectancy, but this only to a limited degree unless fundamental aging processes are addressed. Likewise, single-molecule drugs meant to slow aging are being studied, but their impact would be constrained by the fact that they cannot perform sophisticated repair. More substantial progress will require the rejuvenation of tissues and organs. This could involve replacing parts with newly grown biological tissue, or rejuvenating tissue from within by repairing accumulated damage.

Fundamentally, DNA is constantly being damaged by many sources, and this corruption of genetic information eventually causes permanent malfunction of cells. It would seem that we would need error detection and genome editing to combat this in the long run, which would constitute a form of advanced nanomedicine. Alternatively, some form of rejuvenation technology could assist healthy cells in outcompeting damaged ones, allowing for radical life extension without the need for genomic perfection. The timelines of these technologies are difficult to speculate on, but success in simple laboratory animals like nematode worms, and then in more complex but still short-lived animals like mice, would serve as bellwethers.

This forecast question asks: Will a senolytic therapy for a companion animal be commercially available before one for humans?

Dog biology

Biologically, domesticated animals may be especially amenable to health improvements. Being bred for long periods under artificial rather than natural selection, their genomes could contain a relatively high number of genes with disease risks. While some of these are a direct result of selected traits, often they are functionally unrelated and are caused by random genetic drift or by their proximity on the chromosome to the selected trait genes. Though this may allow for some relatively easy gains in life expectancy, given that dogs regularly die of “old age” after years of steady decline in overall health, I believe such universal effects of aging would need to be addressed to significantly extend life expectancy.

Could selective breeding itself provide easy gains? Small dogs live around twice as long as the largest dogs, so it may be that further miniaturization would continue this trend. However, we seem to have approached some size limit due to either biology or demand, and selective breeding over the next few decades is not likely to unlock unusual and relevant biological properties not already seen in wild canines.

As with current human gene therapies, dog gene therapies are administered to individuals and do not alter the genomes they pass to their offspring. Will their actual germlines be altered at some point, resulting in inherent lifespan extension in a breed? There would certainly be a market for dogs that live longer without special interventions, though some owners may prefer dogs with shorter lifespans, perhaps to better match their own remaining life expectancy. So there could be a stratification within breeds. However, I am uncertain of the relative importance of genome alteration versus medical treatments for dogs in the future, so my predictions are agnostic with respect to the different means for extending life expectancy.

Predicting lifespan extension

There is no consensus on which current dog breed has the highest life expectancy, but the highest seems to be around 15 years.

This forecast question then explores: When will there be a dog breed with a 30-year life expectancy at birth? 

In the case of lifespan-stratified strains, this increase would only need to apply to one strain. This would not necessarily require heritable alterations of existing breeds; instead, any genetic, medical, lifestyle, or other change that results in a type of dog living to 30 years on average would suffice.

The role of pet longevity research as a pioneering effort for tackling human aging means that progress for pets can inform our forecasts for humans. It also means we can use forecasts for humans to inform our forecasts for pets. According to median Metaculus community predictions as of this writing:

• Maximum human lifespan will increase by 0.75 years per year by 2118.
• A country will attain longevity escape velocity (sustained increase in human life expectancy of at least 1 year per year) in 2112.
• There is a 55% chance that someone born before 2001 will live to be 150. In other words, there is a 55% chance that maximum human lifespan will increase by at least 30 years, or around 25%, by 2150.

Setting aside differences in details of these questions and making a rough assumption that the percent increase in life expectancy and maximum lifespan will be interchangeable over any span of time, what can we infer from this set of predictions?

Taken together, they indicate very little increase in lifespan for about a century followed by rapid progress leading to unpredictable and even unlimited increases within a few decades. (Lifespan extension can reach a kind of singularity where the rate of lifespan increase no longer reflects estimated lifespan due to the expectation of substantial continued increase.) Given the current research strategies, this should happen sooner for dogs, suggesting a doubling in life expectancy for at least some dogs some time in the next century. On one hand, the relative ease of research and of running trials with dogs would have me optimistically suggest two or three decades; on the other hand, dog aging and disease is similar enough to humans that I would not expect their trajectories to differ that much. 

I predict that the first dogs to have a 30-year life expectancy at birth will be around 2072. Given the large variances in predictions for the Metaculus questions on which this prediction is based, I would set the 25th to 75th percentile range of my probability distribution at 2042 to 2102.

It could happen unexpectedly soon if some quirks of domesticated animal genomes allow early trials to show huge progress in one dog breed. It could happen in the unexpectedly distant future if slowing aging turns out to be more fundamentally intractable than we now think, or if biological dogs become so unpopular (unthinkable!) that efforts to improve their health dwindle. In any case, predictions for this question could change significantly in the next decade as the results of early dog aging therapy trials are released.