Ryan said. The current study is part of an ongoing investigation called CALERIE Comprehensive Assessment of Long-Term Effects of Reducing Intake of Energy that began in Previous studies using CALERIE trial data have shown multiple benefits from cutting calories.

One investigation found that calorie restriction slowed aging-related changes in physiology related to the liver, kidneys, metabolism, blood vessels, and the immune system. Another study showed that calorie restriction reduced risk factors for heart disease and type 2 diabetes and improved cardiovascular and metabolic health.

The new study was designed to explore those earlier findings. Decades of research using animal models have demonstrated that caloric restriction without malnutrition enhances the life span and so-called health span. But does this hold true for humans? In the most recent CALERIE study, researchers randomly assigned participants to the calorie restriction group or a control group that had no restrictions on their food intake.

The calorie restriction group consisted of people 44 men and 99 women , while the control group consisted of 75 people 22 men and 53 women. The study population was 76 percent white, 15 percent African American, and 9 percent Asian, Native American, or Pacific Islander, with an average age of Scientists used sophisticated techniques to estimate how many calories each person needed to maintain their body weight.

That target level of 25 percent was selected because this degree of calorie restriction has had the best results in improving life span and health span in animal models and was found to be feasible in most participants in the original pilot study from , according to the authors.

Individuals who were underweight, depressed, or had a history of diabetes , heart disease, or an eating disorder were not enrolled in the study, he says. Everyone in the study was advised about how to cut calories and still get the recommended nutrients, and they were given options for eating patterns that would accommodate their cultural and individual preferences.

Maintaining a 25 percent calorie reduction proved challenging — the majority fell short of the goal, and the average calorie reduction was 12 percent by the end of the trial. However, study participants still lost an average of 16 pounds over the course of the two-year study.

To measure the impact of calorie restriction on biological aging, investigators analyzed blood samples collected from trial participants at pre-intervention baseline and after 12 and 24 months of follow-up. They found that calorie restriction slowed the pace of biological aging over time. Additionally, there appeared to be a dose-response effect: Participants who reduced their caloric intake to a greater extent had a greater decline in their pace of biological aging, says Parker.

These findings add to what is currently known about calorie intake and biological aging, says Jamie Justice, PhD , researcher and assistant professor of gerontology and geriatric medicine at Wake Forest School of Medicine in Winston Salem, North Carolina.

Justice was not involved in the CALERIE study. Because eating less leads to weight loss which can have many health benefits , more research is needed to strengthen the findings on how calorie restriction directly impacts aging, she says. A follow-up of trial participants is now ongoing to determine if the intervention had long-term effects on healthy aging.

We obtained survival data from graphs 11 , 12 extracted from the retrieved literature by using Gompertz model to fit the survival curve with the maximum-likelihood estimation MLE Finally, we identified studies that fit our inclusion criteria, including 46 case-control pairs of survival curves of C.

elegans and pairs of Drosophila Supplementary Data S1 , Supplementary File S 1. We found that different shift patterns exist among different anti-ageing interventions. For example, salicylic acid extends lifespan through a sharp improvement, as the extension occurs in the early stage Fig.

Conversely, ks61 extends lifespan through a slowly improvement, as most of the extension occurs in the late stage Fig. In addition, N2 GD1 extends lifespan through a parallel pattern, as the improvement is the same in each stage of ageing Fig. Therefore, we can infer that different anti-ageing interventions may extend lifespan in different ways or patterns.

Schematic diagram of the cluster features. A — C Different anti-ageing effect patterns e. D Size improvement. E Anti-ageing type features. F Boxplot of the visualized features of resveratrol and other treatments in C. elegans demonstrating that the survival curves generated using identical factors have similar features.

Because the survival curves assessed here were collected from many published papers, environmental factors such as the temperature, medium, strain and sex were not uniform and varied significantly among different laboratories Supplementary Fig.

We clustered survival curves by sampling certain survival points, and using the above factors to cluster the points. These variations led to significant differences even between normal control samples Supplementary Tables S1 and S2 , hindering the direct comparison of the effects of different anti-ageing interventions.

We therefore proposed a method to extract relative features and measure anti-ageing effects compared to those of the control, including the size improvement and anti-ageing type features Fig. Importantly, survival curves for the same factor from different laboratories had similar feature distributions, thus further supporting the feasibility of our method Fig.

We examined not only the degree of the total size improvement but also the pattern of delayed ageing. For a population, a pattern occurring mainly in early life stages could improve the population structure, whereas a pattern effect occurring later in life would likely present more of a burden to society.

Therefore, the parallel shift pattern, with both a large ln size and an ln type close to 0, may be the ideal effect of the lifespan-extending method because it improves survival over the whole lifespan of the population and does not change the lifespan structure, preserving the stability of the population.

In general, anti-ageing interventions in these two organisms are mainly classified into three categories: CR, medication administration, and genetic manipulations Supplementary Data S1.

The differences among these three types of anti-ageing interventions were extracted and expressed as features of the survival curves of each control and case cohort; we also compared the effects of different interventions on the survival curves from a biological perspective.

The visualized feature scatter and corresponding cumulative distribution plot of C. elegans indicated that the degrees of the size improvement due to CR and genetic manipulations were slightly larger than that due to medications, even though the P -value was not statistically significant Fig.

However, genetic manipulations mainly increased the maximum lifespan, whereas CR improved the total lifespan Fig. As illustrated by the average difference curves see the curve comparison in the Methods , the mode of improvement due to CR seemed to benefit types of individuals in the group, whereas genetic manipulations appeared to benefit only a few long-lived individuals.

Thus, CR is more beneficial for a population because more individuals live longer, in contrast to genetic manipulations, which allows a few long-lived individuals to expend more resources to sustain life Fig. Similarly, the meta-analysis of Drosophila indicated a genetic transitivity Fig.

We therefore concluded that CR and genetic manipulations resulted in a large degree of improvement in lifespan compared to medications, although the underlying mechanism of this improvement is unknown.

Nevertheless, the effect pattern of CR is superior to that of genetic manipulations in C. Analysis of different classes indicating that CR is more effective at changing the shape and scale of the survival curves. A Cumulative distribution of the size improvement in C.

B Cumulative distribution of the type ratio in C. C Average case-control difference curves of C. D Cumulative distribution of the size improvement in Drosophila. E Cumulative distribution of the type ratio in Drosophila. F Average case-control difference curves of Drosophila.

CR and genetic manipulations have greater effects on extending lifespan than the use of medications. However, genetic manipulation appears to have limited potential for direct application in humans 14 , 15 , 16 , and most people would not comply with such a rigorous CR programme because it may reduce quality of life 17 , 18 , For example, medications that inhibit glycolysis 2-deoxyglucose , enhance insulin action metformin , or affect stress-signalling pathways resveratrol , are being assessed as CR mimetics We therefore examined the lifespan effects of different classes of medications, which revealed some interesting patterns of differences.

Medications were classified by their pharmacological action, clinical application or anti-ageing-related pathway Supplementary Data S1. The KS test was applied to assess the pattern differences and the cumulative distribution of each feature, and a significant pattern was observed for classification by pharmacological action Fig.

The average survival curve differences indicated that the improvement due to antiepileptic medications was relatively large but mainly affected the later stage of life rather than the early stage and that other medications, including antioxidants and hypoglycaemic agents, can preserve the survival across the entire lifespan Fig.

A cross-study comparison for Drosophila produced similar trends Fig. That is, though the total improvements due to hypoglycaemic agents and antioxidants were not as large as those due to antiepileptic medications, the former two types of medications shifted the survival curves in parallel, which might be a healthier way to extend lifespan of a population.

Reportedly, the effects of hypoglycaemic agents and antioxidants on ageing, health, and lifespan are similar to those of CR 25 , Therefore, we can conclude that CR mimetics tend to be the most robust candidate among all the anti-ageing medications. Analysis of different medications demonstrating that antioxidants and hyperglycaemic agents extend the lifespan of both models more effectively than other medications.

A Visualized scatter distribution of different medications classified by the size improvement and type in C. B Cumulative distribution of the size improvements due to medications in C.

C Cumulative distribution of the type features due to medications in C. D Average case-control difference curves of C.

E Visualized scatter distribution of different medications classified by the size improvement and type in Drosophila. F Cumulative distribution of the size improvement due to medications in Drosophila. G Cumulative distribution of the type features due to medications in Drosophila.

H Average case-control difference curves of Drosophila. Understanding the genetic basis of CR is of great importance not only to the biology of ageing but to the understanding of how diet can influence ageing, longevity, health and age-related diseases.

Pharmaceutical interventions that target CR-associated genes are an emerging area with enormous potential. The results showed that the enrichment pathways of the selected genes are largely associated with the determination of adult lifespan GO: , oxidative stress GO: and nutrients GO: Fig.

The determination of adult lifespan category includes the JNK pathway, the insulin-signalling pathway, and the target of rapamycin dTOR , which are important signalling pathways in CR In addition, the oxidative stress and nutritional response pathway is the basis for the lifespan-extending effect of CR 31 , 32 , reflecting the potential benefits of CR.

Analysis of genetic manipulations in Drosophila indicates that CR-associated genes extend lifespan in a better pattern than other analysed genes. A Visualized feature distribution of genes: selected genes orange and other genes cyan.

It is widely accepted that CR, medications and genetic manipulations extend lifespan in a diversity of species ranging from yeast to primates. However, no study has focused on the general differences and effect patterns of these interventions in lifespan experiments.

For example, a recent study of the effects of a variety of lifespan interventions on C. elegans developed a good algorithm with an accelerated failure time AFT model to remove differences in the timescale to the same extent, but the question remained unresolved Here, we introduced a useful methodology that enables the variations in the scale and shape of survival curves to be measured separately, and we performed three procedures to reanalyse survival curves in two classic model organisms.

First, we extracted relative features to measure the effects of distinct interventions, despite variations in the controls.

We combined statistical analyses to validate the reliability of the results. Finally, we calculated the P- value of the KS test of each feature among different subtypes of anti-ageing interventions from a biological perspective.

Our study indicated that CR and genetic manipulations are effective ways in delaying senescence. The effect pattern of CR is superior to that of genetic manipulation in C. elegans but similar to that of genetic manipulation in Drosophila Fig.

Genetic manipulation in mammals faces many problems and risks, and CR, including changes in diet composition, time-restricted feeding or CR mimetics, could be a more feasible approach for humans. These considerations and our results support CR as a feasible and effective anti-ageing intervention.

Graphical summary of the main results of our study. The effects of different anti-ageing interventions exhibited fairly strong species transitivity from C. elegans to Drosophila.

Ageing is an inevitable aspect of life. Although the mechanism of ageing has not been fully elucidated, there are several general hypotheses for the ageing mechanism, including the free radical theory, decreased immune function, the telomere theory, and the brain ageing center doctrine 34 , 35 , Our research suggests that hypoglycaemic agents and antioxidants can obviously preserve the age structure of a population and delay senescence.

These medications protect the cell membrane and organelles from free radical damage and can mimic the effects of CR. In summary, antioxidants and CR mimetics are key regulators for extending the lifespan of C.

elegans and Drosophila. One limitation of our study is the difficulty in obtaining unbiased data. We collected data from recently published papers and cannot control the bias of publishing. However, as most published papers have focused on the amount of the improvement e. Hence, we constructed a new method based on data recovery and model fitting.

In addition, a SurvCurv database and online analysis platform for animal survival data was created, and all the survival records from the database originate from 60 publications 39 , 40 , Therefore, we checked all these publications to ensure that all the relevant survival data were included in our analysis.

We used numerical survival data for C. elegans and Drosophila and specific analysis scripts to address these questions, which were otherwise very difficult to approach. Our research is currently based on a limited number of studies. The extension of these results to other organisms, such as mice or rats, requires a separate examination in future studies.

Survival data from lifespan experiments were extracted from the literature. Papers published on any date up to were included. From the more than studies that this search yielded, papers were selected that contained a graphical survival curve or, for some older studies, provided the complete data set from which a survival curve could be constructed.

Our inclusion criteria were as follows: i Studies were conducted with the original empirical data using real animals and were not reviews or computer simulations. ii The experiment contained both a control group and a treatment group.

iii Survival was reported until all animals died and was extractable from figures or tables in at least five binned time intervals. iv Only strains of control groups that were wild type were included in our analysis. v When multiple experimental groups e.

Accordingly, the following exclusion criteria were also used: i Strains of control groups that were transgenic or mutants were excluded.

The survival curves in the various papers were generated from different numbers of samples and time intervals, preventing a direct comparison of the raw curves. Thus, we used parametric models to fit the raw data to smoothed curves. Six common mortality models, including the exponential model, logistic model, logistic-Makeham model, Weibull model, Gompertz model and Gompertz-Makeham model 13 , were applied to fit each cohort.

The relative goodness of fit was measured by the Akaike 43 , 44 and Bayesian 45 information criterion AIC and BIC, respectively values and the classic MLE together with the parameter values provided.

In general, the Gompertz model yielded good fit results for both the C. elegans and Drosophila survival data Supplementary Data S1 and the Gompertz parameters provided a meaningful explanation This model was therefore used for the analyses.

Because many factors affect survival, considerable variation was observed even between the survival curves for the normal control samples from the different papers, hindering a direct comparison of the curves for different treatments from different papers.

We therefore extracted the features that measured the relative effects of each interaction compared to the matched control samples reported in the same paper. The survival curve for the transition time was approximated to be linear, and the features were extracted on this basis.

We used case-control plots of C. elegans and Drosophila to study the differences between the case and control survival curves. This method allowed us to compare different anti-ageing strategies using the transformed difference survival curves on the same plot. As those lifespan strategies that have a longer extension and a parallel pattern seem to be better strategies, we analysed the biological functions of the genes in the regions affected by such strategies by GO analysis using DAVID 27 , 28 , From this analysis, we can determine the function of these genes and can postulate whether the beneficial effects are associated with CR.

We analysed the biological pathways and compared the results among these better regions and other genes that we collected.

We used two-sample KS tests to determine if two subgroups had different anti-ageing patterns and applied this test to the different features studied 47 , P -values were calculated from the KS statistic and were determined to be more or less significant using a one-sided test.

We also used the KS test to determine which of the parameters in the Gompertz model produced significant improvements in survival as we tested the differences in the parameters between each control and case cohort.

Fontana, L. Extending healthy life span-from yeast to humans. Science , — Article CAS PubMed PubMed Central ADS Google Scholar. Hunt, P. et al. Extension of lifespan in C. elegans by naphthoquinones that act through stress hormesis mechanisms. PloS One 6 , e Lin, Y.

Diacylglycerol lipase regulates lifespan and oxidative stress response by inversely modulating TOR signaling in Drosophila and C. Aging Cell 13 , — Article CAS PubMed PubMed Central Google Scholar.

Martin-Montalvo, A. Metformin improves healthspan and lifespan in mice. Article PubMed PubMed Central Google Scholar. Mattison, J.

Impact of caloric restriction on health and survival in rhesus monkeys from the NIA study. Nature , — Article CAS PubMed ADS Google Scholar. Verdaguer, E. Aging biology: a new frontier for drug discovery. Expert Opin. Drug Discov. Article CAS PubMed Google Scholar.

Combescure, C. Meta-analysis of single-arm survival studies: a distribution-free approach for estimating summary survival curves with random effects.

Article MathSciNet PubMed Google Scholar. Swindell, W. Dietary restriction in rats and mice: a meta-analysis and review of the evidence for genotype-dependent effects on lifespan.

Ageing Res. Article PubMed Google Scholar. Nakagawa, S. Comparative and meta-analytic insights into life extension via dietary restriction. Aging Cell 11 , — Jensen, K. Sex-specific effects of protein and carbohydrate intake on reproduction but not lifespan in Drosophila melanogaster.

Aging Cell 14 , — Guyot, P. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves.

BMC Med. Liu, Z. Recovering the raw data behind a non- parametric survival curve. Pletcher, S. Why do life spans differ? Partitioning mean longevity differences in terms of age-specific mortality parameters. A Biol. Testa, G. Calorie restriction and dietary restriction mimetics: a strategy for improving healthy aging and longevity.

Weinert, B. Theories of aging. Tosato, M. The aging process and potential interventions to extend life expectancy. Aging 2 , — PubMed PubMed Central Google Scholar.

Anton, S. Psychosocial and behavioral pre-treatment predictors of weight loss outcomes. Eat Weight Disord. Dirks, A. Calorie restriction in humans: potential pitfalls and health concerns.

Ageing Dev. Redman, L. Effect of caloric restriction in non-obese humans on physiological, psychological and behavioral outcomes. Lane, M. The serious search for an anti-aging pill. Ingram, D. Development of calorie restriction mimetics as a prolongevity strategy.

Everitt, A. Caloric restriction versus drug therapy to delay the onset of aging diseases and extend life. Age Dordr 27 , 39—48 Article PubMed Central Google Scholar.

Roth, G. Caloric restriction mimetics: The next phase. Calorie restriction mimetics: an emerging research field. Aging Cell 5 , 97— Hadley, E. Poehlman, E. Caloric restriction mimetics: physical activity and body composition changes. Huang, D. Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources.

Article CAS Google Scholar. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists.