Clinical Kidney Journal, 2020, vol. 13, no. 6, 1017–1024 doi: 10.1093/ckj/sfaa014 Advance Access Publication Date: 14 March 2020 Original Article ORIGINAL ARTICLE Association of a single nucleotide polymorphism combination pattern of the Klotho gene with non-cardiovascular death in patients with chronic kidney disease Serafi Cambray, Marcelino Bermudez-Lopez, Milica Bozic and Jose M. Valdivielso ; on behalf of the NEFRONA investigators* Vascular and Renal Translational Research Group, Institute for Biomedical Research Dr. Pifarré Foundation, IRBLleida and RedinRen RETIC, ISCIII, Lleida, Spain *The NEFRONA investigators are listed in the Acknowledgements section. Correspondence to: Serafi Cambray; E-mail: scambray@irblleida.cat, José M. Valdivielso; E-mail: valdivielso@irblleida.cat ABSTRACT Background. Chronic kidney disease (CKD) is associated with an elevated risk of all-cause mortality, with cardiovascular death being extensively investigated. However, non-cardiovascular mortality represents the biggest percentage, showing an evident increase in recent years. Klotho is a gene highly expressed in the kidney, with a clear influence on lifespan. Low levels of Klotho have been linked to CKD progression and adverse outcomes. Single nucleotide polymorphisms (SNPs) of the Klotho gene have been associated with several diseases, but studies investigating the association of Klotho SNPs with non- cardiovascular death in CKD populations are lacking. Methods. The main aim of this study was to assess whether 11 Klotho SNPs were associated with non-cardiovascular death in a subpopulation of the National Observatory of Atherosclerosis in Nephrology (NEFRONA) study (n ¼ 2185 CKD patients). Results. After 48 months of follow-up, 62 cardiovascular deaths and 108 non-cardiovascular deaths were recorded. We identified a high non-cardiovascular death risk combination of SNPs corresponding to individuals carrying the most frequent allele (G) at rs562020, the rare allele (C) at rs2283368 and homozygotes for the rare allele (G) at rs2320762 (rs562020 GG/AG þ rs2283368 CC/CT þ rs2320762 GG). Among the patients with the three SNPs genotyped (n ¼ 1016), 75 (7.4%) showed this combination. Furthermore, 95 (9.3%) patients showed a low-risk combination carrying all the opposite genotypes (rs562020 AA þ rs2283368 TT þ rs2320762 GT/TT). All the other combinations [n ¼ 846 (83.3%)] were considered as normal risk. Using competing risk regression analysis, we confirmed that the proposed combinations are independently associated with a higher fhazard ratio [HR] 3.28 [confidence interval (CI) 1.51–7.12]g and lower [HR 6  106 (95% CI 3.3  107–1.1  105)] risk of suffering a non-cardiovascular death in the CKD population of the NEFRONA cohort compared with patients with the normal-risk combination. Received: 28.10.2019; Editorial decision: 13.1.2020 C The Author(s) 2020. Published by Oxford University Press on behalf of ERA-EDTA. V This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 1017 1018 | S. Cambray et al. Conclusions. Determination of three SNPs of the Klotho gene could help in the prediction of non-cardiovascular death in CKD. Keywords: chronic kidney disease, Klotho, non-cardiovascular death, polymorphisms INTRODUCTION related to diabetic nephropathy [37], type 2 diabetes [38, 39], ath- eromatous progression [40], onset of stroke [41], chronic haemo- Chronic kidney disease (CKD) is a worldwide health problem, dialysis mortality [42] and the incidence of CKD [43]. the incidence and prevalence of which for 2017 were estimated Considering the abovementioned data, we decided to assess at 19.7  106 and 697  106, respectively [1]. It is predicted that how 11 Klotho SNPs were related to all-cause death, cardiovascular CKD will become the fifth leading cause of years of life lost by death and non-cardiovascular death in the National Observatory of 2040, showing a 100% increase compared with 2016 rates [2]. It Atherosclerosis in Nephrology (NEFRONA) study [44]. has been widely documented that CKD patients show an in- creased risk of presenting with cardiovascular disease [3, 4], which could explain the high death rates [5, 6]. Consequently, MATERIALS AND METHODS most of the genetic studies aimed at identifying new single nu- Study population cleotide polymorphisms (SNPs) associated with an increased risk of death in CKD patients are focused on cardiovascular risk The NEFRONA study is a prospective, multicentre cohort study [7–10]. Nevertheless, despite an evident increase in non-cardio- to analyse the morbidity and mortality of CKD patients in Spain vascular mortality in CKD patients [5, 11], no studies have fo- [45]. From 2009 to 2012, we recruited patients between 18 and cused on polymorphisms related to this type of death. 75 years old with different degrees of CKD (950 Stage 3, 807 Klotho, a co-receptor of fibroblast growth factor 23 (FGF23), is Stages 4 and 5 and 688 on dialysis) and 559 non-CKD volunteers a single-pass encoded protein that has 1012 amino acids and is (eGFR >60 mL/min/1.73 m2) and followed them for 4 years, col- subject to cleavage to produce soluble Klotho (sKlotho) [12]. In lecting data every 6 months on fatal and non-fatal cardiovascu- humans, this protein is expressed in a variety of tissues, showing lar events (CVEs), non-cardiovascular death or initiation of particularly strong expression in kidney tubules [13]. Klotho renal replacement therapy. Physicians responsible for patient gained relevance when it was described as an anti-ageing pro- recruitment recorded the incidence of CVEs following the tein, as Klotho knockout mice showed a phenotype resembling International Classification of Diseases, Ninth Revision, Clinical human ageing, including shorter lifespan [14]. Furthermore, the Modification. For out-of-hospital deaths, the cause of death was main anti-ageing effects of Klotho were attributable to the pro- established upon interviewing family members. All included tein produced in the kidney [15]. Currently there are many stud- volunteers signed an informed consent provided by their re- ies investigating Klotho effects on ageing in both humans and spective local ethics committee that previously approved the mice and many clinical trials involving the analysis of circulating study. Volunteers with a history of CVEs, significant carotid ste- sKlotho levels are being performed [16]. It has been demonstrated nosis, active infections (human immunodeficiency virus, tuber- that the levels of both Klotho messenger RNA and protein, are de- culosis), pregnancy, life expectancy <12 months and having creased in CKD [17]. Furthermore, low levels of sKlotho have been received any organ transplant or carotid artery surgery were not linked to CKD progression [18–20] and adverse outcomes [19, 20]. allowed to participate in the NEFRONA study. However, some studies did not find a relationship between For this study, controls and non-Caucasian volunteers were sKlotho levels and adverse outcomes in CKD patients [21, 22] af- excluded. A flow diagram of the experiment design is shown in ter considering FGF23 levels. Indeed, Klotho effects do not have Supplementary data, Figure S1. to be strictly limited to FGF23 signalling. Hu et al. showed that FGF23 was not indispensable for Klotho-induced phosphaturia SNP selection and genotyping [23] and that Klotho has the capacity to interfere with insulin, DNA was extracted from blood samples of the NEFRONA Wnt and transforming growth factor-b1 pathways in an FGF23- study gathered in the REDinREN Biobank [46]. The 11 Klotho independent manner [24]. Altogether, it is clear that Klotho SNPs selected (rs9536254, rs567170, rs577912, rs580332, affects multiple age-related pathologies in FGF23-dependent and rs495392, rs2320762, rs562020, rs576404, rs385564, rs9536282 independent manners [25] and that premature death is present and rs2283368) were included in a previous study in which SNPs in the majority of patients with CKD, while the main causes are related to bone mineral disorders and vascular calcification not restricted to greater cardiovascular disease incidence [4]. were selected [43]. Moreover, SNPs should not have been previ- The Klotho gene (KL) is located on chromosome XIII and is ously related to cardiovascular or non-cardiovascular death. composed of five exons that expand through >50 000 nucleoti- Genotyping was performed as previously described [40]. Hardy– des [26]. According to the National Centre for Biotechnology Weinberg equilibrium (HWE) was performed by means of a chi- Information, Klotho contains >11 000 SNPs [27]. Most of the squared test and genotyping quality control was done by in- genetic studies on Klotho have focused on the KL-VS polymor- cluding Coriell Institute Biorepository replicates in each plate phisms located in exon 2 (name due to the F352V change of (samples NA07348, NA07349 and NA07351). rs9536314 and C370S from rs9527025), which are related to hu- man longevity [28], cognition [29], neuroprotection [30] and oc- cult coronary artery disease [31]. The rs1207568 (G395A) SNP Clinical data and biochemical variables located at the promoter region has also been widely studied and An itinerant team including a nurse and two technicians has been associated with coronary artery disease [32], metabolic obtained anthropometric data, clinical data and blood samples. syndrome [33], kidney stones [34], mortality in dialysis [35] and Family history of cardiovascular death and cardiovascular risk immunoglobulin A nephropathy [36]. Other Klotho SNPs that factors was obtained on recruitment day. Atherosclerosis as- emerged from genome-wide association studies have been sessment was performed at baseline as previously described Polymorphisms of Klotho and mortality in CKD | 1019 [47, 48] and plaques were defined according to the American mainly due to the marked difference in non-cardiovascular Society of Echocardiography and the Mannheim Carotid Intima- death. Finally, the dominant model for the more frequent allele Media Thickness (cIMT) Consensus as cIMT lumen protrusion T of rs2320762 showed a greater incidence of non-cardiovascu- 1.5 mm [49, 50]. lar death and of any death for the GG genotype. Taking into account data from individual genetic models, we Statistical analysis postulated a high non-cardiovascular death genotype combina- tion. This combination corresponds to individuals carrying All statistical analyses were done using SPSS statistics software high-risk allele G of the rs562020 SNP, the high-risk allele C of version 24.0.0.0 (IBM, Armonk, NY, USA). Quantitative variables the rs2283368 SNP and the high-risk genotype GG of the are shown as average and standard deviation (SD), qualitative rs2320762 SNP (rs562020 GG/AG þ rs2283368 CC/CT þ rs2320762 variables as n and percent and quantitative variables not follow- GG); the three SNPs were not in linkage disequilibrium ing a normal distribution as median [quartile 1 (Q1)–quartile 3 (Supplementary data, Table S3). In our population, 75 (7.3%) (Q3)]. Comparisons between groups were done using Student’s t patients showed this genotype. Furthermore, we also postu- test, chi-squared test and Mann–Whitney U test, respectively. lated that carriers of the alternative genotypes would show For comparisons between more than two groups, we used an fewer non-cardiovascular deaths (rs562020 AA þ rs2283368 analysis of variance (ANOVA) and multiple comparisons be- TT þ rs2320762 GT/TT), with 95 (9.3%) patients showing this tween groups were performed with an honestly significant dif- low-risk combination of genotypes. All the other combinations ference Tukey’s test. Multivariable Fine and Gray competing [846 (83.2%)] were considered as normal risk. It should be noted risk regressions were performed for non-cardiovascular death, that all three SNPs could be correctly genotyped in 1016 patients choosing as competing risk cardiovascular death and receiving (271 CKD3, 223 CKD 4–5 and 522 on dialysis), 46.4% of the initial a kidney transplant. Data are presented as hazard ratios (HRs) study cohort. To test if our hypothesis was certain, the cohort and 95% confidence intervals (CIs). P-values <0.05 were consid- was divided according to the three proposed risk combinations ered statistically significant. of genotypes and bivariate analysis was performed. As shown The linkage disequilibrium of selected SNPs was calculated in Table 3, the three groups only showed significant differences with CubeX [51] . in non-cardiovascular death and any death incidence, and the only parameters that were close to reaching significance were RESULTS the presence of basal atheroma plaques and glucose levels. Finally, in order to determine the importance of the pro- The analysis encompassed only patients with a Caucasian ge- posed genotypes in predicting non-cardiovascular death in our netic background and CKD [2906 (96.73% of the NEFRONA cohort, we performed a competing risk regression analysis study)] and subsequently only the ones with data on at least adjusting for all the variables in the NEFRONA study that could one of the Klotho SNPs analysed [2185 (72.7% of the NEFRONA contribute to non-cardiovascular death. As shown in Table 4, af- study); Supplementary data, Figure S1]. During the 48 months of ter adjusting for all the possible confounding variables, the pro- follow-up, we recorded 62 cardiovascular deaths (2.8%) and 108 posed combinations of genotypes still showed a significant non-cardiovascular deaths (4.9%), for 170 total deaths in our co- association with non-cardiovascular death. hort (7.7%). Table 1 shows that the variables related to both car- diovascular death and non-cardiovascular death were diabetes, advanced CKD stages and atheromatous plaque on the basal DISCUSSION evaluation. Higher levels of intact parathyroid hormone, C-reac- This study identifies a combination of SNPs of three Klotho poly- tive protein and phosphate were also associated with both types morphisms that are associated with a higher or lower risk of suf- of death; moreover, lower levels of albumin and serum calcium fering non-cardiovascular death in a CKD population. Patients were related to both cardiovascular and non-cardiovascular with CKD show an increased risk of death, and traditionally re- death. Consequently, all these variables were associated with search has focused on the high rates of cardiovascular death pre- any death, which also showed an association with male sex and sented by this population [3, 52]. However, CKD patients also diastolic blood pressure. Pulse pressure, high-density lipopro- show an increased risk of non-cardiovascular death [5, 11, 53], tein cholesterol, glucose and pathological ankle–brachial index and less attention has been paid to this group of patients. were associated with cardiovascular death, while hypertension, The main causes of non-cardiovascular death among CKD total cholesterol and low-density lipoprotein cholesterol were patients are infectious diseases and cancer. CKD patients show associated with non-cardiovascular death. increased risk of infection in both early CKD stages [54] and in After proving that all the SNPs satisfied the HWE dialysis [55], and it has been recently shown that mice with re- (Supplementary data, Table S1), we analysed the relationship of duced Klotho levels have lower survival rates after induced sep- all genotypes with cardiovascular death, non-cardiovascular sis [56]. Unfortunately, the low numbers of deaths due to death and any death (Supplementary data, Table S2) and elabo- infection in the high- and low-risk groups (two and one, respec- rated the models of dominance. Supplementary data, Table S3 tively) preclude further studies in this direction. Klotho also shows the allele frequencies in our cohort. As shown in Table 2, shows antitumor properties [57], and its levels are reduced in under dominant assumption of the most frequent allele G of certain tumours, such as pancreatic [58] and breast [59], while the rs562020 genotype, the carriers of the A allele in homozy- high levels are considered indicate a more favorable prognosis gosity showed a lower incidence of any death. When the domi- in many cancer types, such as hepatocellular [60] and oesopha- nant model for the less frequent C allele of rs2283368 was geal [61] carcinomas. In our cohort, deaths attributable to cancer tested, we noted that TT carriers showed a lower incidence of were three in the high-risk group and zero in the low-risk group, any death and non-cardiovascular death and CT þ TT carriers but again, the numbers are too low to perform further analyses. showed a higher incidence of both, as cardiovascular death was Despite this, we cannot discard the possible role of the de- equal in both groups. It is obvious that the differences observed scribed combination of genotypes in infections and/or cancer in any death under the dominant model of rs2283368 were deaths in CKD patients. 1020 | S. Cambray et al. Table 1. Epidemiological, clinical and biochemical parameters of cardiovascular, non-cardiovascular and any death of the cohort Cardiovascular Death Non-cardiovascular death Any death Yes No Yes No Yes No Variable (n ¼ 62) (n ¼ 2123) P-value (n ¼ 108) (n ¼ 2077) P-value (n ¼ 170) (n ¼ 2015) P- value Age (years) 63.02 (9.39) 58.21 (12.66) 0.000 64.78 (9.65) 58.01 (12.6) 0.000 64.1 (9.5) 57.86 (12.7) 0.000 Sex (female), n (%) 19 (30.6) 805 (37.9) 0.244 34 (31.5) 790 (38) 0.171 53 (31.2) 771 (38.3) 0,067 Smoking (yes), n (%) 38 (61.3) 1213 (57.1) 0.515 70 (64.8) 1181 (56.9) 0.103 108 (63.5) 1143 (56.7) 0.085 Diabetes (yes), n (%) 32 (51.6) 523 (24.6) 0.000 38 (35.2) 517 (24.9) 0.017 70 (41.2) 485 (24.1) 0.000 Hypertension (yes), n (%) 57 (91.9) 1933 (91.1) 0.810 91 (84.3) 1899 (91.4) 0.011 148 (87.1) 1842 (91.4) 0.056 Dyslipidaemia (yes), 39 (62.9) 1426 (67.2) 0.481 64 (59.3) 1401 (67.5) 0.077 103 (60.6) 1362 (67.6) 0.062 CKD stage, n (%) Stage 3 16 (25.8) 851 (40.1) 0.001 26 (24.1) 841 (40.5) 0.001 42 (24.7) 825 (40.9) 0.000 Stages 4 and 5 16 (25.8) 708 (33.3) 38 (35.2) 686 (33) 54 (31.8) 670 (33.3) Dialysis 30 (48.4) 564 (26.6) 44 (40.7) 550 (26.5) 74 (43.5) 520 (25.8) Body mass index (kg/m2) 29.01 (5.89) 28.25 (5.15) 0.261 28.21 (5) 28.28 (5.18) 0.884 28.5 (5.36) 28.26 (5.16) 0.561 Systolic blood 147.9 (25.8) 142.9 (21.8) 0.074 142.5 (24.3) 143 (21.8) 0.791 144.5 (24.9) 142.9 (21.6) 0.429 pressure (mmHg) Diastolic blood 78.7 (13.7) 81.4 (11.6) 0.138 79.27 (11.5) 81.45(11.7) 0.06 79.08 (12.35) 81.53 (11.65) 0.009 pressure (mmHg) Pulse pressure (mmHg) 69.1 (20.2) 61.5 (18.2) 0.001 63.23 (18.7) 61.63 (18.36) 0.377 65.41 (19.46) 61.39 (18.2) 0.006 Haematocrit (%) 37.5 (5.5) 38.5 (5) 0.117 36.47 (5) 38.66 (5) 0.000 36.86 (5.22) 38.69 (5.02) 0.000 Albumin (g/dL) 3.83 (0.46) 4.1 (0.46) 0.000 3.93 (0.5) 4.1 (0.46) 0.001 3.9 (0.49) 4.1 (0.45) 0.000 iPTH (pg/mL), median (Q1-Q3) 166 (105–264) 120 (68–220) 0.006 157 (76.6–263) 120 (68–220) 0.048 161.8 (86–263) 118 (68–217) 0.001 C-reactive protein, median (Q1-Q3) 4 (1.4–8.1) 1.96 (0.94–4.32) 0.002 4.6 (1.68–10.2) 1.92 (0.92–4.2) 0.000 4.24 (1.66–9.7) 1.9 (0.9–4.1) 0.000 Total cholesterol (mg/dL) 173.32 (51) 178.7 (38.9) 0.29 169.9 (39.7) 179.03 (39.2) 0.02 171.15 (44) 179.2 (38.84) 0.011 High-density lipoprotein 45.07 (15.34) 49.55 (15.39) 0.032 49.9 (17.67) 49.3 (15.28) 0.713 48.15 (17) 49.53 (15.2) 0.295 cholesterol (mg/dL) Low-density lipoprotein 98.13 (36.32) 102.14 (33.3) 0.381 93.25 (32.5) 102.47 (33.4) 0.01 95.08 (34) 102.6 (33.1) 0.009 cholesterol (mg/dL) Triglycerides (mg/dL) 149.76 (79) 144.4 (81.2) 0.623 140.3 (81) 144.77 (81.21) 0.588 143.6 (80) 144.6 (81,2) 0.888 Glucose (mg/dL) 122 (52.93) 107.2 (39.15) 0.035 104.2 (41.7) 107.8 (39) 0.358 110.6 (46.6) 107.4 (39) 0.385 Calcium (mg/dL) 9.11 (0.58) 9.32 (0.59) 0.007 9.15 (07) 9.32 (0.58) 0.018 9.14 (0.66) 9.33 (0.58) 0.000 Phosphate (mg/dL) 4.62 (1.23) 4.03 (1.09) 0.000 4.19 (1.14) 4.04 (1.09) 0.167 4.35 (1.19) 4.02 (1.08) 0.000 Sodium (mEq/L) 139.67 (3.26) 140.31 (3) 0.107 139.55 (4.2) 140.33 (2.99) 0.06 139.6 (3.87) 140.35 (2.98) 0.014 Potassium (mEq/L) 5.02 (0.73) 4.8 (0.6) 0.025 4.9 (0.7) 4.8 (0.59) 0.152 4.95 (0.71) 4.8 (0.59) 0.009 Any plaque basal (yes), n (%) 60 (96.8) 1489 (70.1) 0.000 98 (88.9) 1453 (70) 0.000 156 (91.8) 1393 (69.1) 0.000 Pathology ankle–brachial 33 (53.2) 587 (27.9) 0.000 38 (36.2) 582 (28.2) 0.079 71 (42.5) 549 (27.5) 0.000 index (yes), n (%) Values are expressed as average (SD) unless stated otherwise. Comparisons between groups performed with Student’s t test for quantitative normally distributed varia- bles, Mann–Whitney U test for non-normally distributed quantitative variables and Chi-squared test for categorical data. Statistically significant values are in bold. iPTH: intact parathyroid hormone. Table 2. Proposed SNPs and models with effects on cardiovascular, non-cardiovascular and any death rs562020_DOM G rs2283368_DOM C rs2320762_DOM T (n ¼ 2185) (n ¼ 1017) (n ¼ 2184) GG þ AG AA CC þ CT TT GG GT þ TT Type of death n ¼ 1944 n ¼ 241 P-value n ¼ 216 n ¼ 801 P-value n ¼ 318 n ¼ 1866 P–value Whole cohort (n ¼ 2185) Cardiovascular death, n (%) 59 (3) 3 (1.2) 0.114 7 (3.2) 27 (3.4) 0.925 11 (3.5) 51 (2.7) 0.471 62 (2.8) Non-cardiovascular death, n (%) 101 (5.2) 7 (2.9) 0.122 22 (10.2) 33 (4.1) 0.000 26 (8.2) 82 (4.4) 0.004 108 (4.9) All death, n (%) 160 (8.2) 10 (4.1) 0.026 29 (13.4) 60 (7.5) 0.006 37 (11.6) 133 (7.1) 0.006 170 (7.8) Comparisons between groups were performed with the chi-squared test. Statistically significant values are in bold. The three polymorphisms identified in this work were previ- background in their work, the different composition of the co- ously analysed in a cohort of CKD patients on haemodialysis hort (the Friedman et al. study was exclusively haemodialysis [42] and none of them was related to any type of death. patients) and/or the fact that they did not test combinations of Differences between our study and the study by Friedman et al. different polymorphisms. Nevertheless, it is worth mentioning [42] could be attributed to the 11% non-Caucasian genetic that in our population the rs577912 SNP that Friedman et al. Polymorphisms of Klotho and mortality in CKD | 1021 Table 3. Epidemiological, clinical, biochemical parameters and cause of death according to the three proposed genotypes High-risk phenotype All others Low-risk phenotype (n ¼ 75) (n ¼ 846) (n ¼ 95) rs562020 GG/AG þ rs2320762 rs562020 AA þ rs2320762 Variable GG þ rs2283368 CC/CT GT/TT þ rs2283368 TT P-value Age (years) 57.43 (12.32) 56.98 (13.28) 57.45 (11.91) 0.957 Sex (female), n (%) 30 (40) 325 (38.4) 41 (43.2) 0.656 Smoking (yes), n (%) 47 (62.7) 484 (57.2) 46 (48.4) 0.147 Diabetes (yes), n (%) 17 (22.7) 191 (22.6) 22 (23.2) 0.992 Hypertension (yes), n (%) 63 (84) 762 (90.1) 88 (92.6) 0.160 Dyslipidaemia (yes), n (%) 46 (61.3) 543 (64.2) 68 (71.6) 0.296 Stage 3 20 (26.7) 233 (27.5) 18 (18.9) 0.429 Stages 4 and 5 14 (18.7) 185 (21.9) 24 (25.3) Dialysis 41 (54.7) 428 (50.6) 53 (55.8) Months in dialysis 26.49 (25.96) 27.59 (33.96) 28.02 (28.18) 0.961 Type of dialysis (HD), n (%) 28 (68.3) 276 (64.5) 33 (62.3) 0.830 Body mass index (kg/m2) 27.45 (5.29) 27.77 (5.29) 27.95 (5.64) 0.751 Systolic blood pressure (mmHg) 140.4 (19.62) 142.3 (23.43) 140.05 (20.94) 0.719 Diastolic blood pressure (mmHg) 82.07(10.25) 80.84 (12.75) 82.09 (12.09) 0.717 Pulse pressure (mmHg) 58.36 (15.7) 61.52 (19.1) 58 (17.76) 0.308 Haematocrit (%) 38.27 (4.58) 37.59 (5) 37.68 (3.9) 0.538 Albumin (g/dL) 3.95 (0.5) 4.02 (0.44) 4.06 (0.4) 0.165 iPTH (pg/mL) 238.97 (232.18) 211.71 (204.13) 211.5 (182.86) 0.590 C-reactive protein 4.53 (6.4) 5 (9.35) 3 (4.11) 0.214 Total cholesterol (mg/dL) 175.16 (44.94) 173.26 (39.83) 174.46 (39.49) 0.931 High-density lipoprotein cholesterol (mg/dL) 49.19(18) 49.09 (16.1) 48.65 (12.93) 0.968 Low-density lipoprotein cholesterol (mg/dL) 95.96 (40.5) 97.74 (33.6) 98.07 (33.47) 0.898 Triglycerides (mg/dL) 157.63 (107.56) 139.26 (73.77) 148 (71) 0.162 Glucose (mg/dL) 101.46 (32.65) 105.68 (41.52) 113.04 (49.18) 0.083 Calcium (mg/dL) 9.19 (0.75) 9.23 (0.62) 9.21 (0.57) 0.9 Phosphate (mg/dL) 4.2 (1.34) 4.34 (1.26) 4.47 (1.32) 0.250 Sodium (mEq/L) 139.43 (3.27) 139.9 (3.29) 139.72 (2.92) 0.516 Potassium (mEq/L) 4.93 (0.65) 4.83 (0.65) 4.85 (0.62) 0.496 Any plaque basal (yes), n (%) 47 (62.7) 598 (70.7) 73 (76.8) 0.131 Pathological ankle–brachial index (yes), n (%) 19 (25.7) 256 (30.6) 36 (38.3) 0.184 Cardiovascular death (yes), n (%) 3 (4) 30 (3.5) 1 (1.1) 0.417 Non-cardiovascular death (yes), n (%) 10 (13.3) 43 (5.1) 2 (2.1) 0.003 Any death (yes), n (%) 13 (17.3) 73 (8.6) 3 (3.2) 0.005 Values are presented asaverage and SD unless stated otherwise. Comparisons between groups performed with ANOVA for quantitative variables and chi-squared test for categorical data. Statistically significant values are in bold. iPTH: intact parathyroid hormone. Table 4. Coefficient of competing risk (n ¼ 704) related to any death in their study is close to reaching a signifi- Variable HR (95% CI) P-value cant association in our study (P ¼ 0.062; Supplementary data, Table S2). The rs562020 SNP has been related to better cognitive Age 1.046 (0.994–1.101) 0.09 function in people 92–100 years old and, in the same work, Sex (women) 1.109 (0.474–2.598) 0.810 rs2283368 was significantly associated with cognitive decline Smoking (yes) 1.544 (0.587–4.056) 0.380 Diabetes (yes) 1.209 (0.460–3.175) 0.700 >7 years [62]. In the Framingham cohort, the relationship be- Hypertension (yes) 0.709 (0.286–1.757) 0.460 tween rs562020 and the presence of valvar or vascular calcifica- Dyslipidaemia (yes) 0.495 (0.243–1.010) 0.054 tion was analysed and no relationship was found [63], data that CKD Stages 4 and 5 1.275 (0.365–4.453) 0.700 agree with our results showing no relationship between the Dialysis 2.22 (0.744–6.630) 0.150 polymorphism and cardiovascular death. Diastolic blood pressure, mmHg 0.991 (0.960–1.023) 0.550 This work has certain limitations. The first limitation is that Pulse pressure, mmHg 0.990 (0.968–1.011) 0.350 we did not analyse Klotho plasma levels, so we cannot correlate Haematocrit (%) 0.925 (0.845–1.012) 0.085 the proposed genotypes with protein levels. The second limita- C-reactive protein 1.015 (0.993–1.037) 0.150 tion is that we do not have a validation cohort to reassess the Total cholesterol, mg/dL 1.009 (0.993–1.025) 0.270 importance of the proposed combination of SNPs. Despite evi- Low-density lipoprotein 0.993 (0.976–1.011) 0.440 dent limitations, our work has certain strengths that we would cholesterol, mg/dL Any plaque basal 2.574 (0.631–10.489) 0.190 like to highlight. First is that we focused on Klotho SNPs that Sodium, mEq/L 0.957 (0.833–1.100) 0.530 have not been extensively studied, so we pave the way for fur- Potassium, mEq/L 1.410 (0.814–2.440) 0.220 ther studies using the proposed genotypes in various diseases Klotho high-risk genotype 3.286 (1.516–7.129) 0.003 in which Klotho could be involved. Second is that the NEFRONA Klotho low-risk genotype 6  106 (3.3  107–1.1  105) 0.000 Statistically significant values are in bold. 1022 | S. Cambray et al. study is very comprehensive, allowing us to correct for the ef- Universitario Insular de Gran Canaria (Las Palmas); Garcı́a fect of the proposed combination of genotypes by multiple pos- Herrera, Antonio L. Hospital Universitario Puerto Real sible confounding factors. Third is that our cohort is relatively (Cádiz); Garcı́a Mena, Mercedes. Hospital San Juan de Dios big, with a prospective design and a relatively long follow-up (Zaragoza); Gil Sacaluga, Luis; Aguilar, Maria. Hospital time. Fourth is that we have considered a combination of three Virgen del Rocı́o (Sevilla); Górriz, José Luis. Hospital SNPs to check the relationship with all-cause death, an ap- Universitario Doctor Peset (Valencia); Huarte Loza, Emma. proach that could help to unravel synergistic effects of different Hospital San Pedro (Logron ~ o); Lerma, José Luis. Hospital polymorphisms that are missed by individual SNP analysis. Universitario Salamanca (Salamanca); Liebana Can ~ ada, In summary, we have identified a combination of SNPs of Antonio. Hospital de Jaén (Jaén); Marı́n Álvarez, Jesús Pedro. the Klotho gene independently associated with higher and Hospital San Pedro de Alcántara (Cáceres); Martı́n Alemany, lower risks for non-cardiovascular death in CKD. Validation of Nàdia. Hospital Jose p Trueta (Girona); Martı́n Garcı́a, Jesús. these findings could be useful in the stratification of non-car- Hospital Nuestra Sen ~ ora de Sonsoles (Ávila); Martı́nez diovascular death risk in CKD patients. Castelao, Alberto. Hospital Universitari de Bellvitge (Barcelona); Martı́nez Villaescusa, Marı́a. Complejo SUPPLEMENTARY DATA Hospitalario Universitario de Albacete (Albacete); Martı́nez, Isabel. Hospital Galdakao (Bilbao); Moina Eguren, In ~ igo. Supplementary data are available at ckj online. Hospital Basurto (Bilbao); Moreno Los Huertos, Silvia. Hospital Santa Bárbara (Soria); Mouzo Mirco, Ricardo. ACKNOWLEDGEMENTS Hospital El Bierzo, Ponferrada (León); Munar Vila, Antonia. The authors would like to thank the NEFRONA team (Eva Hospital Universitari Son Espases (Palma de Mallorca); Castro, Virtudes Marı́a, Teresa Molı́, Teresa Vidal and Mun ~ oz Dı́az, Ana Beatriz. Hospital Virgen del Consuelo Meritxell Soria) and the Biobank of RedInRen for their in- (Valencia); Navarro González, Juan F. Hospital Universitario valuable support. The NEFRONA study is funded by a re- Nuestra Sen ~ ora de Candelaria (Santa Cruz de Tenerife); search grant from AbbVie and the Spanish government Nieto, Javier; Carren ~ o, Agustı́n. Hospital General RETIC (RD16/0009/0011) and FIS PI16/01354. The NEFRONA Universitario de Ciudad Real (Ciudad Real); Novoa study investigator group is composed by the following: Fernández, Enrique. Complexo Hospitalario de Ourense Aladrén Regidor, Ma José. Hospital Comarcal Ernest Lluch (Ourense); Ortiz, Alberto; Fernandez, Beatriz. IIS-Fundación (Calatayud); Almirall, Jaume; Ponz, Esther. Corporació Parc Jiménez Dı́az (Madrid); Paraı́so, Vicente. Hospital Taulı́ (Barcelona); Arteaga Coloma, Jesús. Hospital de Universitario del Henares (Madrid); Pérez Fontán, Miguel. Navarra (Pamplona); Bajo Rubio, Ma Auxiliadora; Dı́az, Complejo Hospitalario Universitario A Corun ~ a (A Corun ~ a); Raquel Raquel Hospital La Paz (Madrid); Belart Rodrı́guez, Peris Domingo, Ana. Hospital Francesc de Borja (Valencia); Montserrat. Sistemes Renals (Lleida); Gascón, Antonio, ~ era Haces, Celestino. Hospital Universitario Marqués de Pin Hospital Obispo Polanco (Teruel); Bover Sanjuan, Jordi. Valdecilla (Santander); Prados Garrido, Ma Dolores. Hospital Fundació Puigvert. IIB Sant Pau (Barcelona); Bronsoms Universitario San Cecilio (Granada); Prieto Velasco, Mario. Artero, Josep. Clı́nica Girona (Girona); Cabezuelo Romero, Hospital de León (León); Puig Marı́, Carmina. Hospital Juan B; Muray Cases, Salomé. Hospital Reina Sofı́a (Murcia); d’Igualada (Barcelona); Rivera Gorrı́n, Maite. Hospital Calvin~ o Varela, Jesús. Hospital Universitario Lugus Augusti Universitario Ramón y Cajal (Madrid); Rubio, Esther. (Lugo); Caro Acevedo, Pilar. Clı́nica Ruber (Madrid); Carreras Hospital Puerta del Hierro (Madrid); Ruiz, Pilar. Hospital Sant Bassa, Jordi. Diaverum Baix Llobregat (Barcelona); Cases Joan Despı́ Moisès Broggi (Barcelona); Salgueira Lazo, Amenós, Aleix; Massó Jiménez, Elisabet. Hospital Clı́nic Mercedes; Martı́nez Puerto, Ana Isabel. Hospital Virgen (Barcelona); Moreno López, Rosario. Hospital de la Defensa Macarena (Sevilla); Sánchez Tomero, José Antonio. Hospital (Zaragoza); Cigarrán Guldris, Secundino; López Prieto, Saray. Universitario de la Princesa (Madrid); Sánchez, José Emilio. Hospital Da Costa (Lugo); Comas Mongay, Lourdes. Hospital Hospital Universitario Central de Asturias (Oviedo); Sans General de Vic (Barcelona); Comerma, Isabel. Hospital Lorman, Ramon. Hospital de Figueres (Girona); Saracho, General de Manresa (Barcelona); Compte Jové, Ma Teresa, Ramon. Hospital de Santiago (Vitoria); Sarrias, Maria; Serón, Hospital Santa Creu Jesús (Tarragona); Cuberes Izquierdo, Daniel. Hospital Universitari Vall d’Hebron (Barcelona); Marta. Hospital Reina Sofı́a (Navarra); de Álvaro, Fernando; Soler, Marı́a José; Barrios, Clara. Hospital del Mar Hevia Ojanguren, Covadonga. Hospital Infanta Sofı́a (Barcelona); Sousa, Fernando. Hospital Rio Carrión (Madrid); de Arriba de la Fuente, Gabriel. Hospital (Palencia); Toran, Daniel. Hospital General de Jerez (Cadiz); Universitario Guadalajara (Guadalajara); del Pino y Pino, Ma Tornero Molina, Fernando. Hospital de Sureste (Arganda del Dolores. Complejo Hospitalario Universitario Torrecardenas Rey); Usón Carrasco, José Javier. Hospital Virgen de la Luz (Almerı́a); Diaz-Tejeiro Izquierdo, Rafael; Ahijado Hormigos, (Cuenca); Valera Cortes, Ildefonso. Hospital Virgen de la Francisco Hospital Virgen de la Salud (Toledo); Dotori, Victoria (Málaga); Vilaprinyo del Perugia, Ma Merce. Institut Marta. USP Marbella (Málaga); Duarte, Verónica. Hospital de Catala d’Urologia i Nefrologia (Barcelona); Virto Ruiz, Rafael Terrassa (Barcelona); Estupin ~ an Torres, Sara. Hospital C. Hospital San Jorge (Huesca); Vicente Pallarés Carratalá Universitario Canarias (Santa Cruz de Tenerife); Fernández Clinica MEDEFIS (Vila-real. Castellón), Carlos Santos Reyes, Ma José. Hospital de Segovia (Segovia); Fernández Altozano CS Azuqueca de Henares (Guadalajara); Miguel Rodrı́guez, Ma Loreto. Hospital Prı́ncipe de Asturias (Madrid); Artigao Ródenas CS Zona III (Albacete); Inés Gil Gil Área Fernández, Guillermina. Clı́nica Santa Isabel (Sevilla); Galán Básica Sanitaria de Arán. CAP Viella (Lleida); Francisco Adan Serrano, Antonio. Hospital General Universitario de Gil CS Alfaro (La Rioja); Emilio Garcı́a Criado Centro de Salud Valencia (Valencia); Garcı́a Cantón, Cesar. Hospital del Carpio (Córdoba.); Rafael Durá Belinchón CS Godella Polymorphisms of Klotho and mortality in CKD | 1023 (Valencia); Jose Ma Fernández Toro CS Zona Centro cardiovascular events in patients with CKD. Clin J Am Soc (Cáceres); Juan Antonio Divisón Garrote Centro de Salud de Nephrol 2015; 10: 232–240 Casas Ibán~ ez. Consultorio de Fuentealbilla (Albacete). 10. O’Seaghdha CM, Tin A, Yang Q et al. Association of a cystatin C gene variant with cystatin C levels, CKD, and risk of inci- dent cardiovascular disease and mortality. Am J Kidney Dis FUNDING 2014; 63: 16–22 This study was funded by the intramural programme of the 11. de Jager DJ, Vervloet MG, Dekker FW. Noncardiovascular IRBLleida, the Instituto de Salud Carlos III (RETIC RD16/0009, mortality in CKD: an epidemiological perspective. Nat Rev PI15/00960 and PI16/01354) and FEDER funds. Nephrol 2014; 10: 208–214 12. Xu Y, Sun Z. Molecular basis of Klotho: from gene to function in aging. Endocr Rev 2015; 36: 174–193 AUTHORS’ CONTRIBUTIONS 13. Lim K, Groen A, Molostvov G et al. a-Klotho expression in hu- V.J.M. and C.S. were involved in the study concept and design. man tissues. J Clin Endocrinol Metab 2015; 100: E1308–E1318 14. Kuro-O M, Matsumura Y, Aizawa H et al. Mutation of the Data acquisition was done by B.M. and data interpretation by mouse klotho gene leads to a syndrome resembling ageing. V.J.M., C.S. and B.L.M. C.S. and V.J.M. were involved in the sta- Nature 1997; 390: 45–51 tistical analysis. Drafting of the manuscript was done by C.S. 15. Lindberg K, Amin R, Moe OW et al. The kidney is the principal and V.J.M. All authors participated in critical revision of the organ mediating klotho effects. J Am Soc Nephrol 2014; 25: manuscript for important intellectual content. Study supervi- 2169–2175 sion was done by V.J.M. The NEFRONA investigators collected 16. Cheikhi A, Barchowsky A, Sahu A et al. Klotho: an elephant baseline and prospective follow-up data for the cohort. in aging research. J Gerontol A Biol Sci Med Sci 2019; 74: 1031–1042 CONFLICT OF INTEREST STATEMENT 17. Hu MC, Kuro-OM, Moe OW. Klotho and chronic kidney dis- ease. Contrib Nephrol 2013; 180: 47–63 None declared. 18. Drew DA, Katz R, Kritchevsky S et al. Association between soluble klotho and change in kidney function: the health ag- REFERENCES ing and body composition study. J Am Soc Nephrol 2017; 28: 1859–1866 1. James SL, Abate D, Abate KH et al. Global, regional, and na- 19. Kim HR, Nam BY, Kim DW et al. Circulating a-klotho levels in tional incidence, prevalence, and years lived with disability CKD and relationship to progression. Am J Kidney Dis 2013; for 354 diseases and injuries for 195 countries and territo- 61: 899–909 ries, 1990–2017: a systematic analysis for the Global Burden 20. Liu QF, Ye JM, Yu LX et al. Plasma s-Klotho is related to kid- of Disease Study 2017. Lancet 2018; 392: 1789–1858 ney function and predicts adverse renal outcomes in 2. Foreman KJ, Marquez N, Dolgert A et al. Forecasting life ex- patients with advanced chronic kidney disease. J Investig pectancy, years of life lost, and all-cause and cause-specific Med 2018; 66: 669–675 mortality for 250 causes of death: reference and alternative 21. Nowak A, Friedrich B, Artunc F et al. Prognostic value and scenarios for 2016–40 for 195 countries and territories. Lancet link to atrial fibrillation of soluble Klotho and FGF23 in he- 2018; 392: 2052–2090 modialysis patients. PLoS One 2014; 9: e100688 3. Sarnak MJ, Levey AS, Schoolwerth AC et al. Kidney disease as 22. Seiler S, Rogacev KS, Roth HJ et al. Associations of FGF- a risk factor for development of cardiovascular disease: a 23 and sKlotho with cardiovascular outcomes among statement from the American Heart Association Councils on patients with CKD stages 24. Clin J Am Soc Nephrol 2014; 9: Kidney in Cardiovascular Disease, High Blood Pressure 1049–1058 Research, Clinical Cardiology, and Epidemiology and 23. Hu MC, Shi M, Zhang, J et al. Klotho: a novel phosphaturic Prevention. Circulation. 2003; 108: 2154–2169 substance acting as an autocrine enzyme in the renal proxi- 4. Thompson S, James M, Wiebe N et al. Cause of death in mal tubule. FASEB J 2010; 24: 3438–3450 patients with reduced kidney function. J Am Soc Nephrol 24. Hu MC, Shiizaki K, Kuro-O M et al. Fibroblast growth factor 23 2015; 26: 2504–2511 and Klotho: physiology and pathophysiology of an endo- 5. de Jager DJ, Grootendorst DC, Jager KJ et al. Cardiovascular crine network of mineral metabolism. Annu Rev Physiol 2013; and noncardiovascular mortality among patients starting 75: 503–533 dialysis. JAMA 2009; 302: 1782–1789 25. Kuro-O M. The Klotho proteins in health and disease. Nat 6. Raymond NT, Zehnder D, Smith SC et al. Elevated relative Rev Nephrol 2019; 15: 27–44 mortality risk with mild-to-moderate chronic kidney disease 26. Matsumura Y, Aizawa H, Shiraki-Iida T et al. Identification of decreases with age. Nephrol Dial Transplant 2007; 22: the human klotho gene and its two transcripts encoding 3214–3220 membrane and secreted klotho protein. Biochem Biophys Res 7. Brancaccio D, Biondi ML, Gallieni M et al. Matrix GLA protein Commun 1998; 242: 626–630 gene polymorphisms: clinical correlates and cardiovascular 27. National Center for Biotechnology Information. https:// mortality in chronic kidney disease patients. Am J Nephrol www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId¼9365&choose 2005; 25: 548–552 Rs¼all (25 May 2019, date last accessed) 8. Rodrigo E, Pich S, Subirana I et al. A clinical-genetic approach 28. Arking DE, Krebsova A, Macek M et al. Association of human to assessing cardiovascular risk in patients with CKD. Clin aging with a functional variant of klotho. Proc Natl Acad Sci Kidney J 2017; 10: 672–678 USA 2002; 99: 856–861 9. Spoto B, Mattace-Raso F, Sijbrands E et al. Association of IL-6 29. Dubal DB, Yokoyama JS, Zhu L et al. Life extension factor klo- and a functional polymorphism in the IL-6 gene with tho enhances cognition. Cell Rep 2014; 7: 1065–1076 1024 | S. Cambray et al. 30. Erickson CM, Schultz SA, Oh JM et al. KLOTHO heterozygosity 48. Betriu A, Martinez-Alonso M, Arcidiacono MV et al.; on behalf attenuates APOE4-related amyloid burden in preclinical AD. of the investigators from the NEFRONA study. Prevalence of Neurology 2019; 92: e1878–e1889 subclinical atheromatosis and associated risk factors in 31. Arking DE, Becker DM, Yanek LR et al. KLOTHO allele status chronic kidney disease: the NEFRONA study. Nephrol Dial and the risk of early-onset occult coronary artery disease. Transplant 2014; 29: 1415–1422 Am J Hum Genet 2003; 72: 1154–1161 49. Stein JH, Korcarz CE, Hurst RT et al.; American Society of 32. Jo SH, Kim SG, Choi YJ et al. KLOTHO gene polymorphism is as- Echocardiography Carotid Intima-Media Thickness Task sociated with coronary artery stenosis but not with coronary Force. Use of carotid ultrasound to identify subclinical vas- calcification in a Korean population. Int Heart J 2009; 50: 23–32 cular disease and evaluate cardiovascular disease risk: a 33. Luo L, Hao Q, Dong B et al. The Klotho gene G-395A polymor- consensus statement from the American Society of phism and metabolic syndrome in very elderly people. BMC Echocardiography Carotid Intima-Media Thickness Task Geriatr 2016; 16: 46 Force. Endorsed by the Society for Vascular Medicine. J Am 34. Telci D, Dogan AU, Ozbek E et al. KLOTHO gene polymor- Soc Echocardiogr 2008; 21: 93–111; quiz 189–190. phism of G395A is associated with kidney stones. Am J 50. Touboul PJ, Hennerici MG, Meairs S et al. Mannheim carotid Nephrol 2011; 33: 337–343 intima-media thickness and plaque consensus (2004–2006– 35. Ko GJ, Lee YM, Lee EA et al. The association of Klotho gene 2011). An update on behalf of the advisory board of the 3rd, polymorphism with the mortality of patients on mainte- 4th and 5th watching the risk symposia, at the 13th, 15th nance dialysis. Clin Nephrol 2013; 80: 263–269 and 20th European Stroke Conferences, Mannheim, 36. Ko GJ, Lee EA, Jeon US et al. The association of Klotho poly- Germany, 2004, Brussels, Belgium, 2006, and Hamburg, morphism with disease progression and mortality in IgA ne- Germany, 2011. Cerebrovasc Dis 2012; 34: 290–296 phropathy. Kidney Blood Press Res 2012; 36: 191–199 51. Gaunt TR, Rodrı́guez S, Day IN.Cubic exact solutions for the 37. Iyengar SK, Sedor JR, Freedman BI et al.; Family Investigation estimation of pairwise haplotype frequencies: implications of Nephropathy and Diabetes. Genome-wide association for linkage disequilibrium analyses and a web tool ’CubeX’. and trans-ethnic meta-analysis for advanced diabetic kid- BMC Bioinformatics 2007; 8: 428. ney disease: Family Investigation of Nephropathy and 52. Gansevoort RT, Correa-Rotter R, Hemmelgarn BR et al. Diabetes (FIND). PLoS Genet 2015; 11: e1005352 Chronic kidney disease and cardiovascular risk: epidemiol- 38. Zhao W, Rasheed A, Tikkanen E et al. Identification of new sus- ogy, mechanisms, and prevention. Lancet 2013; 382: 339–352 ceptibility loci for type 2 diabetes and shared etiological path- 53. Runesson B, Qureshi AR, Xu H et al. Causes of death across ways with coronary heart disease. Nat Genet 2017; 49: 1450–1457 categories of estimated glomerular filtration rate: the 39. Xue A, Wu Y, Zhu Z et al. Genome-wide association analyses Stockholm CREAtinine Measurements (SCREAM) project. identify 143 risk variants and putative regulatory mecha- PLoS One 2019; 14: e0209440 nisms for type 2 diabetes. Nat Commun 2018; 9: 2941 54. Dalrymple LS, Katz R, Kestenbaum B et al. The risk of 40. Valdivielso JM, Bozic M, Galimudi RK et al. Association of the infection-related hospitalization with decreased kidney rs495392 Klotho polymorphism with atheromatosis progres- function. Am J Kidney Dis 2012; 59: 356–363 sion in patients with chronic kidney disease. Nephrol Dial 55. Berman SJ, Johnson EW, Nakatsu C et al. Burden of infection Transplant 2018; 34: 2079–2088 in patients with end-stage renal disease requiring long-term 41. Yokoyama S, Oguro R, Yamamoto K et al. A klotho gene single dialysis. Clin Infect Dis 2004; 39: 1747–1753 nucleotide polymorphism is associated with the onset of stroke 56. Jorge LB, Coelho FO, Sanches TR et al. Klotho deficiency and plasma klotho concentration. Aging 2018; 11: 104–114 aggravates sepsis-related multiple organ dysfunction. Am J 42. Friedman DJ, Afkarian M, Tamez H et al. Klotho variants and chronic hemodialysis mortality. J Bone Miner Res 2009; 24: Physiol Renal Physiol 2019; 316: F438–F448 1847–1855 57. Rubinek T, Wolf I. The role of alpha-klotho as a universal tu- 43. Valls J, Cambray S, Pérez-Guallar C et al. Association of can- mor suppressor. Vitam Horm 2016; 101: 197–214 didate gene polymorphisms with chronic kidney disease: 58. Abramovitz L, Rubinek T, Ligumsky H et al. KL1 internal re- results of a case-control analysis in the nefrona cohort. Front peat mediates klotho tumor suppressor activities and inhib- Genet 2019; 10: 118 its bFGF and IGF-I signaling in pancreatic cancer. Clin Cancer 44. Arroyo D, Betriu A, Martinez-Alonso M et al. Observational Res 2011; 17: 4254–4266 multicenter study to evaluate the prevalence and prognosis 59. Rubinek T, Shulman M, Israeli S et al. Epigenetic silencing of of subclinical atheromatosis in a Spanish chronic kidney the tumor suppressor klotho in human breast cancer. Breast disease cohort: baseline data from the NEFRONA study. BMC Cancer Res Treat 2012; 133: 649–657 Nephrol 2014; 15: 168 60. Xie B, Zhou J, Yuan L et al. Epigenetic silencing of Klotho ex- 45. Junyent M, Martı́nez M, Borràs M et al. Predicting cardiovas- pression correlates with poor prognosis of human hepato- cular disease morbidity and mortality in chronic kidney dis- cellular carcinoma. Hum Pathol 2013; 44: 795–801 ease in Spain. The rationale and design of NEFRONA: a 61. Tang X, Fan Z, Wang Y et al. Expression of klotho and b-cate- prospective, multicenter, observational cohort study. BMC nin in esophageal squamous cell carcinoma, and their clini- Nephrol 2010; 11: 14 copathological and prognostic significance. Dis Esophagus 46. Calleros-Basilio L, Cortés MA, Garcı́a-Jerez A et al. Quality as- 2016; 29: 207–214 surance of samples and processes in the Spanish Renal 62. Mengel-From J, Soerensen M, Nygaard M et al. Genetic Research Network (REDinREN) Biobank. Biopreserv Biobank vriants in KLOTHO associate with cognitive function in the 2016; 14: 499–510 oldest old group. J Gerotol A2016; 71: 1151–1159 47. Gracia M, Betriu À, Martı́nez-Alonso M et al. Predictors of 63. Tangri N, Alam A, Wooten EC et al. Lack of association of subclinical atheromatosis progression over 2 years in Klotho gene variants with valvular and vascular calcification patients with different stages of CKD. Clin J Am Soc Nephrol in Caucasians: a candidate gene study of the Framingham 2016; 11: 287–296 Offspring Cohort. Nephrol Dial Transplant 2011; 26: 3998–4002