Is long-term calorie restriction in humans worth it?

Nearly 78 years have passed since the first experiment in the Journal of Nutrition was published that found that the restriction of calories without undernutrition would retard aging and prolong the mean and maximal lifespan of rats. That pioneering work of Clive M. McCay, Mary Crowell, and L. A. Maynard of Cornell led to several subsequent studies by aging researchers using calorie restriction (CR) as a research model with similar findings across several species, mainly in rodents. Now CR is well recognized as having contributed more to the understanding of the biological processes of aging than any other viable research model. Enormous interest in CR continues because of clues it may provide for intervention in aging individuals to prevent chronic disease and increase longevity.

In humans, however, questions remain surrounding the effectiveness, feasibility, and safety of CR. These questions were the topics of discussion on Saturday, April 20, in a symposium organized by the American Society for Nutrition at Experimental Biology 2013. Susan Roberts, director of USDA’s HNRCA’s Energy Metabolism Laboratory at Tufts University, and John Speakman of University of Aberdeen, presented an overview of the state of the evidence on CR from experiments in non-human primates, naturally occurring CR populations, and from randomized clinical trials.

Roberts discussed the few data of CR in humans. She included that from naturally occurring CR population such as those of Okinawa, restrained eaters in the U.S., and the famed accidental experiment of Dr. Roy Walford in Biosphere 2 and gave an overview of the randomized trials in humans including data from phase 1 of CALERIE, which stands for The Comprehensive Assessment of the Long-term Effects of Reducing Intake of Energy. The first phase found that human subjects on CR differing substantially in glycemic load after 12 months resulted in comparable weight loss.

According to Roberts, “what we know so far” about CR in humans can be summarized as the following:

• 50 percent CR is outside of a sustainable range for humans, cannot be maintained voluntarily and has major side effects.
• Several naturally occurring CR populations (such as the Okinawans) that self-documented CR of 14 to 27 percent is likely to be inflated relative to their actual CR.
• No published human intervention study has yet to document more than 12 percent calorie restriction after six months, although several have reported up to 20 to 25 percent before six months. But these achieved values are associated with consistent improvements in variety of health risk factors and few adverse events.

Roberts also introduced the design of phase 2 of CALERIE (currently being prepared for publication) of which was discussed by later speakers involved in the trial.

Speakman presented a comprehensive overview of the animal data including that of rodents and non-human primates. From the rodent studies — including from recent work by Jim Nelson’s lab at University of Texas  — he said what can be gathered is that some rodent strains live longer on CR, but others do not. In addition, extended lifespan on CR in rodents can depend on their gender. For example, he noted, some of the strains of longest-living males on CR were not the same as the longest-living females.

Speakman said that while rodents display a clear linear relation of CR severity to lifespan extension up to 65 percent, recombinant inbred mice show a diversity of response and that appears to affected by gender differences. He also noted that most of the rodent studies began CR immediately after weaning. “That’s never going to happen in humans,” Speakman said. Studies using CR in rodents in midlife have shown only 12 percent (obviously not as appealing as 65 percent when considering what it takes to follow CR in the long-term).

In non-human primates, Speakman said, the data on CR are also mixed as evidenced by the inconsistencies in results from the famed Wisconsin National Primate Research Center in Madison and a parallel study conducted by the National Instiute of Aging on rhesus monkeys. The first showed clear signs of reduced aging as an effect from CR while the other did not. The two studies had important differences in design and, interestingly, Speakman shared that the researchers of these two studies would be compiling and reanalyzing their data for later publication.

CALERIE 2 Preliminary Findings

After Speakman, the following three speakers described how a select group of 220 healthy volunteers across three sites — Tufts University, Pennington Biomedical Research Center, and Washington University School of Medicine —  chose to shun a quarter of their dietary calories in the hope of improving their long-term health and, potentially, extending lifespan. These participants of the CALERIE phase 2 trial were randomized to 25 percent CR or ad libitum eating. The large NIH-funded, mulitcenter, parallel group, randomized controlled trial was designed to evaluate how a calorie restricted diet affected biomarkers of aging and age-related disease over the long-term.

Sai Krupa Das, an assistant professor at Tufts and who was also the lead author of the phase 1 trial, shared preliminary findings of phase 2. The primary aim of the trial, she said, was to evaluate whether 25 percent CR resulted in a sustained metabolic adaptation. One of the underlying theories of how CR worked is that attenuates the biological aging process by reducing resting metabolic rate (RMR) leading to reduced cumulative oxidative damage from aerobic respiration, she explained. It was for these reasons that they looked at biomarkers of RMR.

However, she said, the calorie restriction did not cause a change in body temperature that would be indicative of reduced resting metabolic rate that would show adaptation. These data (which are not published yet) are currently being evaluated for proper interpretation. She added that overall energy expenditure did show “a little bit of an adaptation” and that there was an “adaptation in the non-resting components related to activity.” The findings are interesting because they are inconsistent with previous studies in animals and a recently in humans showing a metabolic adaptation through RMR and core body temperature in response to CR.

“Basically, these are the primary mechanisms in humans — reduction of metabolic rate and core body temperature — we did not find an adaptation in the resting component, but we did find an adaptation in the non-resting component,” she said. “If there was a reduction, that is supposed to lend support to the oxidative theory. What exactly this means is still being worked out.”

Edward Saltzman, MD, of USDA HNRCA at Tufts University, who presented data on behalf of William Kraus of Duke (who wasn’t able to make it to the event) discussed CR’s effects on cardiovascular disease risk factors. The long-term CR had a significant effect on a variety of measurements including a reduced metabolic syndrome score, reduced systolic bp, reduced LDL, reduced triglycerides, and increased HDL that were maintained over the study. There were no significant differences on glucose measures or inflammatory markers IL-6 and TNF-a. These results are consistent with previous studies related to reductions in body weight.

Dennis Villareal followed Dr. Saltzman with a discussion about the long-term effects of CR on bone metabolism and bone mass. Notably, Villareal said, weight loss consistently increases bone turnover and decreases bone mineral density (BMD). An important finding of CALERIE 2 was that as body weight, fat mass, and fat free mass dropped in the trial, there was also a loss in BMD at clinically relevant sites of the hip and spine, although not the whole body.

It was unclear, Villareal added, whether BMD would stabilize over time and whether it would affect future risk of fractures. The BMD loss is suspected to be due to increased bone resorption as a result of unloading on the skeleton and because of hormonal changes related to CR that led to increasing bone turnover. Calcium and vitamin D intake doesn’t appear to have presented an influence on the subjects who were supplied with supplements (1000mg Ca/d; 1000 IU vitamin D/d).

A short question-and-answer session followed the session. One comment from the audience was about being careful that messages about CR did not lead “young people” to anorexia. Another, from Wayne Campbell of Purdue, was about the definition of CR versus simply “weight loss, followed by weight maintenance.” Roberts told me after the event that the healthy diet component is really what separates this CR study from a general weight-loss-followed-by-weight-maintenance study.

Roberts concluded the session by asking the speakers whether, based on their research, they would themselves consider CR as a personal dietary approach or wait for a CR mimetic drug. As far as Speakman is concerned, he’d prefer waiting for a mimetic because he enjoyed the finer things in life too much such as sex, feeling warm, and eating. According to him, CR is only worth it “if you want to feel cold, hungry all the time, and have a reduced libido.” But intermittent CR may be worth a look, added Roberts.

The CALERIE phase 2 study, a huge undertaking, Roberts said, was “highly successful” because most of the participants — 188 in all — met their weight-loss goals (~15 percent) and followed through by maintaining their weight with 25 percent CR and a healthy diet to the end of the two years. Roberts told me that the subjects’ achievement depended primarily on use of “hunger control” strategies of which she specializes. For example, she told me after the talk that the participants were able to control their hunger by following a diet of which comprised of many high-fiber and low GI foods and around 25 percent of calories from protein.