Aging and the risks of all-cause and cause-specific mortality among diabetics: a prospective cohort study

Abstract

Background

Aging is an important driver for age-related diseases and death, but the evidence regarding the effects of aging on diabetics is limited. This study aims to evaluate the associations of aging with all-cause and cause-specific mortality among diabetics.

Methods

A total of 5,278 diabetics from the National Health and Nutrition Examination Survey 1999-2014 were included. The aging status was measured from different perspectives, including Phenotypic Age, Biological Age, telomere length, and Klotho concentration. Cox proportional hazards models were used to examine the relationships between aging and all-cause, cardiovascular disease (CVD), and cancer mortality. Mediation analysis was performed to elucidate the role of aging in associations of metformin with mortality.

Results 

Over median follow-up for 7.3 years, 1,355 diabetics died. There was a positive and linear association of mortality with Phenotypic Age (hazard ratio (HR)_all-cause 1.05, 95% confidence interval (CI) 1.05-1.06; HR_CVD 1.05, 95%CI 1.04-1.07; HR_cancer 1.05, 95%CI 1.04-1.07; all P<0.001) and Biological Age (HR_all-cause 1.07, 95%CI 1.05-1.08; HR_CVD 1.08, 95%CI 1.05-1.10_; HR_cancer 1.05, 95%CI 1.03-1.08; all P<0.001). Telomere length was inversely associated with all-cause mortality (tertile (T)3 vs. T1: HR 0.67, 95%CI 0.45-0.98; P_trend=0.036). The concentration of Klotho had a U-shaped relationship with mortality (T2 vs. T1: HR_all-cause 0.62, 95%CI 0.43-0.88; HR_CVD 0.48, 95%CI 0.26-0.86; HR_cancer 0.47 95%CI 0.25-0.88). Additionally, metformin users had a lower HR for mortality compared to those without use (HR_all-cause 0.64, 95%CI 0.56-0.73; HR_CVD 0.51, 95%CI 0.37-0.70; HR_cancer 0.65, 95%CI 0.44-0.95; all P<0.001), which was partly mediated by Phenotypic Age and Biological Age.          

Conclusions

These findings suggested aging was a noteworthy risk factor of mortality for diabetics and therapies targeting anti-aging could be encouraged to halt the progression of diabetes. 

Background

Diabetes, as the ninth major cause of death, is a worldwide public health concern, with considerable healthcare and economic burden (1). The number of diabetics has quadrupled in the past 3 decades and is projected to exceed more than 642 million by 2040 (2). Diabetics often suffer from complications and premature death (3). Statistically, the risk of death and cardiovascular events is 2 to 4 times higher among diabetics than in the general population (4). Hence, the identification of risk factors affecting the survival of diabetics may contribute to delaying premature mortality and its development.

Aging characterized by a progressive decline in homeostasis is generally accepted as the leading risk factor for age-related diseases and death (5). Nearly half of diabetics are older adults who are primarily responsible for the growing burden (6, 7). Moreover, diabetics usually develop a premature aging status with dysfunction of multiple systems (8). Compelling animal studies and clinical trials also revealed drugs or gene editing targeting aging improved diabetic outcomes (9, 10). Aging, though, was critical for diabetics, epidemiological evidence is limited. Previously, investigations explored the relationship between short telomere length and all-cause mortality among Chinese (11), Danish (12), and Italian diabetics (13), with the hazard ratio (HR) ranging from 1.87 to 3.45. However, some data showed that aging in the lung (14) and myocardium (15) was independent of telomere length, indicating the limited capacity of it to comprehensively reflect the aging status of the organ or organism. Recently, novel aging measures including multiple readily available clinical biomarkers were applied to capture body’s homeostasis, such as Phenotypic Age and Biological Age. These markers were reported to well predict adverse outcomes. For example, a prospective cohort study in the general population showed that higher Phenotypic Age was positively associated with increased death risk, especially diabetes-caused mortality (HR=1.20) (16). However, studies focused on diabetics are lacking.

Therefore, we measured aging at multiple perspectives and conducted a prospective and representative cohort study to investigate the association between aging and all-cause and cause-specific mortality among U.S. diabetics based on the data from the National Health and Nutrition Examination Survey (NHANES) 1999-2014.

Methods

Aging Markers Measurement

Phenotypic Age was calculated according to the method published previously (18). In brief, 9 aging-related variables, including chronological age, albumin, creatinine, glucose, Ln-C-reactive protein (CRP), lymphocyte percent, mean cell volume, red blood cell distribution width, alkaline phosphatase, and white blood cell count, were evaluated together based on the following equation.

where: xb= -19.907།0.0336 × Albumin + 0.0095 × Creatinine + 0.1953 × Glucose + 0.0954 × LnCRP།0.0120 × Lymphocyte Percent + 0.0268 × Mean Cell Volume + 0.3306 × Red Cell Distribution Width + 0.00188 × Alkaline Phosphatase + 0.0554 × White Blood Cell Count + 0.0804× Chronological Age

Biological Age was calculated using the methods proposed by Klemera and Doubal (19). In this study, 8 biomarkers (CRP, serum creatinine, HbA_1c, serum albumin, serum total cholesterol, serum urea nitrogen, serum alkaline phosphatase, and systolic blood pressure) were estimated based on the following formulas (1-4) (20). The values j and i represent the number of biomarkers and samples respectively. The values k, q, and s are the regression slope, intercept, and the root means squared error of a biomarker regressed on chronological age, respectively. The value r_j² represents the variance explained by regression chronological age on biomarkers.

Additionally, markers of premature aging were defined as Phenotypic/Biological Age Acceleration and ΔPhenotypic/Biological Age (16). Phenotypic/Biological Age Acceleration was the residual resulting from a linear model when regressing Phenotypic/Biological Age on chronological age. ΔPhenotypic/Biological Age represented the value of Phenotypic/Biological Age minus chronological age.

Measurement of telomere length (NHANES 1999-2002) has been reported elsewhere (21). Briefly, blood samples were collected and determined by quantitative polymerase chain reaction (qPCR) assays to assess the telomere length relative to standard reference DNA (the T/S ratio). Klotho (NHANES 2007-2014) was determined by a commercially available ELISA kit (IBL International, Japan). All sample analyses were performed in duplicate strictly following the manufacturer’s instruction and all results were checked to meet the laboratory’s standardized criteria for acceptability before being released for reporting. 

Mortality Ascertainment

All participants from NHANES 1999-2014 were linked to death records in the National Death Index up to 31 December 2015. Follow-up duration is defined as the time from participating in the NHIS to death for decedents or to the censoring date for survivors. Mortality outcomes in the current study included all-cause, cardiovascular disease (CVD)-cause, and cancer-cause determined by ICD-10 codes recorded in NHANES.

Results

During median follow-up for 7.3 years, 1,355 deaths were documented among 5,278 diabetics. Based on the live or death status of the study population, baseline demographic was provided in Table 1. The deaths were older, more likely to be Non-Hispanic White and widowed or divorced or separated, and were less likely to be obese, and had less frequently drinking alcohol. Those who died had a longer duration of diabetes, and a high percentage of anti-diabetic medication or insulin use but had a lower percentage of metformin use. The deaths had a lower median concentration of Klotho (0.76 vs. 0.82 pg/mL), a shorter median telomere length (0.88 vs. 0.97), and an older mean Phenotypic Age (74.55 vs. 57.88 years) and Biological Age (68.39 years vs. 55.05 years).

Table 2 presented the associations of 4 aging markers with all-cause, CVD, and cancer mortality. Each 1-year increase in Phenotypic Age increased the risk of all-cause, CVD, and cancer mortality by 5% (all P<0.001). Similarly, each 1-year increase in Biological Age was associated with an increase in all-cause, CVD, and cancer mortality (7%, 8%, and 5%, respectively, all P<0.001). Premature aging markers were also positively associated with all-cause, CVD, and cancer mortality (Supplementary Table 1). Additionally, for each unit increase in mean T/S ratio (a marker of cellular senescence), HR for all-cause mortality decreased by 47% (HR 0.53, 95%CI 0.29-0.97), consistent with the quantile analysis. The middle concentration of Klotho was associated with a lower risk of all-cause, CVD, and cancer mortality (T2 vs. T1: HR_all−cause 0.62, 95%CI 0.43-0.88; HR_CVD 0.48, 95%CI 0.26-0.86; HR_cancer 0.47, 95%CI 0.25-0.88).

A dose-response function showed Phenotypic and Biological Age had a linear relationship with all-cause, CVD, and cancer mortality (all P linearity <0.001). Telomere length also showed a linear relationship with all-cause (P linearity 0.015), CVD (P linearity 0.152), and cancer (P linearity 0.037) mortality. A U-shaped correlation was observed between Klotho and all-cause (P nonlinear 0.002), CVD (P nonlinear 0.502), and cancer (P nonlinear 0.014) mortality (Fig. 1 and Supplementary Fig. 1).

In the sensitivity analyses, similar results were observed after excluding participants with a diagnosis of diabetes before the age of 20 (Supplementary Table 2), imputing the missing data of covariates with multiple imputations (Supplementary Table 4), or further adjusting for HbA_1c level and the duration of diabetes (Supplementary Table 5). When excluding the deaths within 1 year of follow-up, further adjusting for education levels and family income-poverty ratio, or further adjusting for HDL-cholesterol, total cholesterol, and HOMA of insulin resistance, the results did not materially change (Supplementary Table 3, 6, and 7).

Furthermore, we investigated the relationship of metformin with mortality among diabetics. Metformin users had a lower HR for all-cause, CVD, and cancer mortality compared to non-use (HR_all−cause 0.64, 95%CI 0.56-0.73; HR_CVD 0.51, 95%CI 0.37-0.70; HR_cancer 0.65, 95%CI 0.44-0.95) (Table 3). Consistently, the use duration of metformin was negatively associated with mortality (Supplementary Table 8). Such associations were also presented regardless of whether blood glucose was well controlled. Further, mediation analysis showed that through decreasing Phenotypic Age, metformin use reduced all-cause (mediated proportion=26.52%, P<0.001), and CVD mortality (mediated proportion=14.40%, P=0.018). Among people with HbA_1c greater than 6.5 percent, Phenotypic Age also mediated this association of metformin use with all-cause (mediated proportion=34.18%, P<0.001), and CVD mortality (mediated proportion=26.22%, P=0.048). Biological Age mediated the effect of metformin use on all-cause mortality in total people and people with HbA_1c greater than 6.5 percent, and the mediated proportion was 14.68% and 19.61%, respectively (Fig. 2 and Supplementary Fig. 2).

Discussion

In the current prospective cohort study of the U.S. population with diabetes, we found that aging-related makers were associated with mortality. Specifically, Phenotypic and Biological Age showed positive associations with all-cause, CVD, and cancer mortality, higher telomere length was associated with lower all-cause mortality, and Klotho had a U-shaped relationship with all-cause, CVD, and cancer mortality. Further, metformin decreased all-cause, CVD, and cancer mortality risk. More importantly, such associations were partly mediated by Phenotypic Age and Biological Age. These findings suggested that the associations of aging with mortality were significant, and anti-aging was an important approach to reduce death among diabetics.

Previously, a meta-analysis of 17 cohorts with 5575 diabetics and 6439 control subjects showed shorter telomere length in diabetes (22). An inverse association was noted between telomere length and mortality among Chinese, Danish, and Italian diabetics (HR=1.87-3.45) (23). Consistent findings were also pronounced in our analysis among the U.S. diabetics. Telomere length is accepted as an indicator of cellular aging. Interestingly, removing aging cells or delaying their formation in mice can improve diabetes progress and its complications (10, 24). However, the aging process of mouse cardiomyocytes and human lungs is only linked to the telomere-associated DNA damage, without significant telomere shortening detected (14, 15). The above results may be attributed to the limited role of telomere in interpretation at organs or the whole body level. Importantly, diabetics often develop multiple organ disorders and die by complications. Hence, more comprehensive indicators which can represent the body’s status were needed.

Phenotypic and Biological Age incorporated composite clinical biomarkers, such as inflammation (e.g. CRP), immunity (e.g. number of immune cells), and organ function (e.g. albumin and serum urea nitrogen) (25). They are useful to identify the unhealthy condition of patients caused by multiple organ disorders. Herein, we found that Phenotypic Age and Biological Age were significantly associated with all-cause and cause-specific mortality among diabetics. Moreover, we defined premature aging status based on chronological age and found that diabetics who appeared older than expected physiologically had a higher death risk. The previous study also reported similar associations of Phenotypic and Biological Age with all-cause mortality, but in the general population (HR=1.09, 1.10, respectively). Notably, among the disease-specific mortality, the risk of diabetes-caused mortality increased mostly (HR=1.20) (16). These researches indicated the importance of the whole body’s aging for death among diabetics.

On the other hand, as the aging-related molecule, Klotho has been reported to be involved in the aging process for over 30 years, where it regulates phosphate homeostasis, insulin, and Wnt signaling (26). Interestingly, in our study population, Klotho concentration had a U-shaped relationship with all-cause, CVD, and cancer mortality. Previous epidemiological studies in the old or chronic kidney disease population revealed that Klotho concentration was negatively associated with death (27, 28). Indeed, overexpression of soluble Klotho up-regulated fibroblast growth factors-23 levels, which may lead to hypovitaminosis D (29). More importantly, a recent prospective cohort study including 6,329 diabetics reported that the level of serum 25-Hydroxyvitamin D was negatively associated with all-cause and cause-specific mortality (30). These studies supported our findings that circulating Klotho may have a U-shaped relationship with mortality risk in diabetics. Collectively, from many perspectives, we observed the effect of aging on promoting diabetics’ death.

Metformin was first introduced to the world in 1957, as an anti-hyperglycemic agent. Consistent with previous studies, we found that metformin users had a lower risk of all-cause and cause-specific mortality, which were positively correlated with the number of days taken among diabetics. Recently, accumulating evidence has revealed the gerotherapeutic effect of metformin on lowering the incidence of multiple age-related diseases and all-cause mortality in diabetics (31). A meta-analysis indicated that the anti-aging function of metformin is possibly independent of its effect on diabetes control (32). Hence, we further investigated whether anti-aging was involved in the role of metformin in preventing mortality among diabetics. We found that metformin use decreased mortality risk among diabetics, regardless of whether HbA_1C was well controlled. Of note, our mediation analyses showed that metformin reduced mortality partly by decreasing Biological Age and Phenotypic Age, emphasizing the importance of anti-aging. Mechanism-related studies in multiple models have elucidated metformin participated in various pathways of aging, including deregulated nutrient sensing, altered intercellular communication, genomic instability, and loss of proteostasis (33, 34). Therefore, treatments targeting anti-aging are greatly promising in improving the mortality of diabetics. Some interventions have been well validated in animal or cells models (35). For example, senolytic drugs (e.g. ABT263) improved glucose tolerance and insulin sensitivity in diabetes mice by reducing senescent cell burden (10). Another combination senolytic agents, dasatinib and quercetin (DQ), offered a good therapeutic effect on mice with age-related diseases (36). In particular, preliminary clinical trials presented the positive effects of DQ on patients with idiopathic pulmonary fibrosis and diabetic kidney disease (9, 37). The current study provided evidence that anti-aging was effective to delay the progression of diabetes and mortality.

This study has some strengths. Firstly, the study was a prospective cohort study based on data from a large nationally representative survey among U.S. diabetics. Secondly, the associations reported in this study were relatively robust by adjustment for a variety of confounders and several sensitivity analyses. Finally, we examined the relationship between aging and death outcomes among diabetics from multiple perspectives. There are also limitations. Firstly, considering the cross-sectional nature of the NHANES data, the time-varying changes in aging markers could not be investigated. Secondly, the study lacked further information on the severity of diabetes, though adjusting for diabetes medication use, duration of diabetes, HbA_1c levels, and some self-reported comorbidities. Thirdly, unmeasured confounders and measurement errors may bias our analyses. Finally, whether anti-aging is involved in the effect of metformin on mortality still needs further research. In addition, we only analyzed the role of one anti-aging drug in mortality, which limited the interpretation of our conclusions.

Conclusions

In summary, our results of nationally representative samples from a prospective cohort study showed significant associations of aging with total and cause-specific mortality among diabetics. Moreover, we observed the improvement in mortality by metformin use among diabetics and further highlighted the mediating effects of aging. These findings suggested that aging accelerated death among diabetics and anti-aging treatments as a promising approach for diabetics.

Abbreviations

CVD: cardiovascular disease; HR: hazard ratio; CI: confidence interval; T: tertile; HbA_1c: glycated hemoglobin; CRP: C-reactive protein; SE: standard error; NCHS: National Center for Health Statistics.

Declarations

Acknowledgements

The authors thank all participants and all investigators.

Author’s contributions

Li Chen and Tianqi Tan conducted analyses. Li Chen, Ying Zhao, and Huimin Chen wrote the draft of the article. Ping Yao and Yuhan Tang conceived of the study design. All authors contributed to the interpretation of the results and critical revision of the manuscript for important intellectual content and approved the final version of the manuscript. Yuhan Tang is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. 

Funding 

This study was supported by the National Key Research and Development Project (2018YFC1604000).

Availability of data and materials 

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

The survey was approved by National Center for Health Statistics (NCHS) Ethics Review Board.

Consent for publication

The authors have reviewed the manuscript and consent for publication.

 Competing interests

No potential conflicts of interest relevant to this article were reported.

References

1. Rawshani A, Rawshani A, Franzen S, Sattar N, Eliasson B, Svensson AM, et al. Risk Factors, Mortality, and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2018;379(7):633–44.
2. Perreault L, Skyler JS, Rosenstock J. Novel therapies with precision mechanisms for type 2 diabetes mellitus. Nature Reviews Endocrinology. 2021;17(6):364–77.
3. Zheng Y, Ley SH, Hu FB. Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nature Reviews Endocrinology. 2018;14(2):88–98.
4. Zimmet P, Alberti KG, Magliano DJ, Bennett PH. Diabetes mellitus statistics on prevalence and mortality: facts and fallacies. Nature Reviews Endocrinology. 2016;12(10):616–22.
5. Di Micco R, Krizhanovsky V, Baker D. d'Adda di Fagagna F. Cellular senescence in ageing: from mechanisms to therapeutic opportunities. Nature reviews Molecular cell biology. 2021;22(2):75–95.
6. Bellary S, Kyrou I, Brown JE, Bailey CJ. Type 2 diabetes mellitus in older adults: clinical considerations and management. Nature Reviews Endocrinology. 2021;17(9):534–48.
7. Huang ES, Laiteerapong N, Liu JY, John PM, Moffet HH, Karter AJ. Rates of Complications and Mortality in Older Patients With Diabetes Mellitus The Diabetes and Aging Study. Jama Internal Medicine. 2014;174(2):251–8.
8. Palmer AK, Gustafson B, Kirkland JL, Smith U. Cellular senescence: at the nexus between ageing and diabetes. Diabetologia. 2019;62(10):1835–41.
9. Hickson LJ, Prata LGPL, Bobart SA, Evans TK, Giorgadze N, Hashmi SK, et al. Senolytics decrease senescent cells in humans: Preliminary report from a clinical trial of Dasatinib plus Quercetin in individuals with diabetic kidney disease. Ebiomedicine. 2019;47:446–56.
10. Aguayo-Mazzucato C, Andle J, Lee T, Midha A, Talemal L, Chipashvili V, et al. Acceleration of β Cell Aging Determines Diabetes and Senolysis Improves Disease Outcomes. Cell Metabol. 2019;30(1):129–42.e4.
11. Cheng FF, Luk AO, Wu HJ, Lim CKP, Carroll L, Tam CHT, et al. Shortened relative leukocyte telomere length is associated with all-cause mortality in type 2 diabetes- analysis from the Hong Kong Diabetes Register. Diabetes Res Clin Pract. 2021;172:108649.
12. Astrup AS, Tarnow L, Jorsal A, Lajer M, Nzietchueng R, Benetos A, et al. Telomere length predicts all-cause mortality in patients with type 1 diabetes. Diabetologia. 2010;53(1):45–8.
13. Bonfigli AR, Spazzafumo L, Prattichizzo F, Bonafe M, Mensa E, Micolucci L, et al. Leukocyte telomere length and mortality risk in patients with type 2 diabetes. Oncotarget. 2016;7(32):50835–44.
14. Birch J, Anderson RK, Correia-Melo C, Jurk D, Hewitt G, Marques FM, et al. DNA damage response at telomeres contributes to lung aging and chronic obstructive pulmonary disease. American Journal of Physiology-Lung Cellular Molecular Physiology. 2015;309(10):L1124-L37.
15. Anderson R, Lagnado A, Maggiorani D, Walaszczyk A, Dookun E, Chapman J, et al. Length-independent telomere damage drives post-mitotic cardiomyocyte senescence. Embo Journal. 2019;38(5).
16. Liu Z, Kuo P, Horvath S, Crimmins E, Ferrucci L, Levine M. A new aging measure captures morbidity and mortality risk across diverse subpopulations from NHANES IV: A cohort study. PLoS Med. 2018;15(12):e1002718.
17. Han T, Gao J, Wang L, Li C, Qi L, Sun C, et al. The Association of Energy and Macronutrient Intake at Dinner Versus Breakfast With Disease-Specific and All-Cause Mortality Among People With Diabetes: The U.S. National Health and Nutrition Examination Survey, 2003-2014. Diabetes Care. 2020;43(7):1442–8.
18. Levine ME, Lu AT, Quach A, Chen BH, Assimes TL, Bandinelli S, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging-Us. 2018;10(4):573–91.
19. Klemera P, Doubal S. A new approach to the concept and computation of biological age. Mech Ageing Dev. 2006;127(3):240–8.
20. Levine M. Modeling the rate of senescence: can estimated biological age predict mortality more accurately than chronological age? The journals of gerontology Series A. Biological sciences. 2013;68(6):667–74.
21. Banach M, Mazidi M, Mikhailidis DP, Toth PP, Jozwiak J, Rysz J, et al. Association between phenotypic familial hypercholesterolaemia and telomere length in US adults: results from a multi-ethnic survey. Eur Heart J. 2018;39(40):3635–40.
22. Wang JF, Dong X, Cao L, Sun YY, Qiu Y, Zhang Y, et al. Association between telomere length and diabetes mellitus: A meta-analysis. J Int Med Res. 2016;44(6):1156–73.
23. Cheng FF, Carroll L, Joglekar MV, Januszewski AS, Wong KK, Hardikar AA, et al. Diabetes, metabolic disease, and telomere length. Lancet Diabetes Endocrinology. 2021;9(2):117–26.
24. Chen L, Mei GB, Jiang CJ, Cheng XE, Li D, Zhao Y, et al. Carbon monoxide alleviates senescence in diabetic nephropathy by improving autophagy. Cell Prolif. 2021;54(6):e13052.
25. Belsky D, Caspi A, Houts R, Cohen H, Corcoran D, Danese A, et al. Quantification of biological aging in young adults. Proc Natl Acad Sci USA. 2015;112(30):E4104-10.
26. Xu YC, Sun ZJ. Molecular Basis of Klotho: From Gene to Function in Aging. Endocr Rev. 2015;36(2):174–93.
27. Semba RD, Cappola AR, Sun K, Bandinelli S, Dalal M, Crasto C, et al. Plasma Klotho and Mortality Risk in Older Community-Dwelling Adults. Journals of Gerontology Series a-Biological Sciences and Medical Sciences. 2011;66(7):794–800.
28. Charoenngam N, Ponvilawan B, Ungprasert P. Lower circulating soluble Klotho level is associated with increased risk of all-cause mortality in chronic kidney disease patients: a systematic review and meta-analysis. Int Urol Nephrol. 2020;52(8):1543–50.
29. Torres PU, Prie D, Molina-Bletry V, Beck L, Silve C, Friedlander G. Klotho: An antiaging protein involved in mineral and vitamin D metabolism. Kidney Int. 2007;71(8):730–7.
30. Wan ZZ, Guo JY, Pan A, Chen C, Liu LG, Liu G. Association of Serum 25-Hydroxyvitamin D Concentrations With All-Cause and Cause-Specific Mortality Among Individuals With Diabetes. Diabetes Care. 2021;44(2):350–7.
31. Kulkarni AS, Gubbi S, Barzilai N. Benefits of Metformin in Attenuating the Hallmarks of Aging. Cell Metab. 2020;32(1):15–30.
32. Campbell JM, Bellman SM, Stephenson MD, Lisy K. Metformin reduces all-cause mortality and diseases of ageing independent of its effect on diabetes control: A systematic review and meta-analysis. Ageing Res Rev. 2017;40:31–44.
33. Barzilai N, Crandall JP, Kritchevsky SB, Espeland MA. Metformin as a Tool to Target Aging. Cell Metab. 2016;23(6):1060–5.
34. Cabreiro F, Au C, Leung KY, Vergara-Irigaray N, Cocheme HM, Noori T, et al. Metformin Retards Aging in C. elegans by Altering Microbial Folate and Methionine Metabolism. Cell. 2013;153(1):228–39.
35. Yang CB, Zhang W, Dong XD, Fu CJ, Yuan JM, Xu ML, et al. A natural product solution to aging and aging-associated diseases. Pharmacol Ther. 2020;216:107673.
36. Novais EJ, Tran VA, Johnston SN, Darris KR, Roupas AJ, Sessions GA, et al. Long-term treatment with senolytic drugs Dasatinib and Quercetin ameliorates age-dependent intervertebral disc degeneration in mice. Nat Commun. 2021;12(1):5213.
37. Justice J, Nambiar A, Tchkonia T, LeBrasseur N, Pascual R, Hashmi S, et al. Senolytics in idiopathic pulmonary fibrosis: Results from a first-in-human, open-label, pilot study. EBioMedicine. 2019;40:554–63.