In Biogerontology, Shanley and Kirkwood pour some water on the blaze of enthusiasm for calorie restriction (CR):

Calorie restriction is known to increase lifespan in many but not all species and may perhaps not do so in humans. Exceptions to life extension have been identified in the laboratory and others are known in nature. Given the variety of physiological responses to variation in food supply that are possible, evolutionary life history theory indicates that an increased investment in maintenance in response to resource shortage will not always be the strategy that maximises Darwinian fitness. Additionally, for the well-studied species in which life extension is observed, there is considerable variation in the response. This suggests that it is not an ancient ancestral response, which has been conserved across the species range. Although calorie restriction does not increase lifespan in all species, it remains a fascinating and valuable tool to study ageing at the whole organism level.

When I first saw the title (“Caloric restriction does not enhance longevity in all species and is unlikely to do so in humans”) and then read the abstract, I was expecting (dreading?) a scathing and decisive indictment of the CR field. In the event, the dent inflicted is quite a bit smaller than I had anticipated.

The basic argument of the review is as follows:

  1. There are sound evolutionary reasons to believe that the maintenance/reproduction balance is adjusted differently in response to famine (presumably the real-world event mimicked by CR) in short-lived vs. long-lived organisms.
  2. All of the well-studied laboratory model organisms are short-lived, indeed were chosen in part because they are, so we might have been misled by the coherence of the findings in these species.
  3. Among rodents, even among different strains of mice, there are dramatic differences in the response to CR.
  4. Among primates, the natural history of a species (particularly regarding the frequency of food shortages over evolutionary time) will likely determine to what extent if at all the species responds favorably to CR.
  5. Regardless of the ultimate result, the jury is still out on primate CR.

Ultimately it’s a bit of theory of the sort that sounds very reasonable, but could be refuted by an equally reasonable-sounding piece of theory (e.g., once the famine response is established in an ancestral species, famine wouldn’t have to be common over evolutionary time for the response to be preserved, especially if contributing genes confer other advantages), followed by a warning against overgeneralization from early results.

I appreciated the caution. CR seems to have it all these days: a good story, a host of clear results in model organisms, some very promising indications in our favorite primate, drug development projects, even its own fan club — everything except the healthy skepticism that the pending results of ongoing experiments rightly deserve.

I can certainly understand the enthusiasm of CR proponents, especially those who are calorie restricting themselves: after the first sip of (sugar-free) Kool-Aid, one has a clear vested interest in the general truth of the proposition that what’s good for a mouse is very likely good for oneself. I can also appreciate the logic of those who have adopted CR as a lifestyle: the best evidence at the moment is that CR does extend lifespan, but if one were to wait for definitive results in higher primates, it might be too late to reap its benefits — so why not start today?

But as a scientist working in the aging field, doing my best to frame hypotheses around what is known and demonstrated rather than what is hoped and extrapolated, I also valued the authors’ caution, and the implicit reminder that we should always remain skeptical.