- Email: [email protected]
Soluble urokinase receptor is a biomarker of cardiovascular disease in chronic kidney disease
clinical investigation
http://www.kidney-international.org & 2014 International Society of Nephrology
Soluble urokinase receptor is a biomarker of cardiovascular disease in chronic kidney disease Bjo¨rn Meijers1,2, Ruben Poesen2, Kathleen Claes1,2, Ruth Dietrich3, Bert Bammens1,2, Ben Sprangers1,2, Maarten Naesens1,2, Markus Storr3, Dirk Kuypers1,2 and Pieter Evenepoel1,2 1
UZ Leuven, Department of Nephrology, Leuven, Belgium; 2KU Leuven, Department of Microbiology and Immunology, Leuven, Belgium and 3Research and Development, Gambro Dialysatoren GmbH, Hechingen, Germany
Soluble urokinase-type plasminogen activator receptor (suPAR) accumulates in patients with chronic kidney disease (CKD). In various non-CKD populations, suPAR has been proposed as a prognostic marker for mortality and cardiovascular disease. However, it is not known whether suPAR holds prognostic information in patients with mild-tomoderate CKD. In a prospective observational study of 476 patients with mild-to-moderate kidney disease, we examined multivariate associations between suPAR, overall mortality, and cardiovascular events. After a mean follow-up of 57 months, 52 patients died and 76 patients had at least one fatal or nonfatal cardiovascular event. Higher suPAR was directly and significantly associated with both overall mortality (univariate hazard ratio 5.35) and cardiovascular events (univariate hazard ratio 5.06). In multivariate analysis, suPAR remained significantly associated with cardiovascular events (full model, hazard ratio 3.05). Thus, in patients with mild-to-moderate CKD, suPAR concentrations show a clear, direct, and graded association with a higher risk for new-onset cardiovascular disease. Kidney International (2015) 87, 210–216; doi:10.1038/ki.2014.197; published online 4 June 2014 KEYWORDS: cardiovascular events; chronic kidney disease; urokinase
Correspondence: Bjo¨rn Meijers, UZ Leuven, Department of Nephrology, Herestraat 49, Leuven 3000, Belgium. E-mail: [email protected] Received 16 December 2013; revised 26 March 2014; accepted 3 April 2014; published online 4 June 2014 210
Soluble urokinase-type plasminogen activator receptor (suPAR) originates from proteolytic cleavage of the urokinasetype plasminogen activator receptor (uPAR) at its glycosyl phosphatidyl inosytol anchor site.1,2 uPAR elicits a plethora of cellular responses that include cellular adhesion, differentiation, proliferation, and migration.1 Given the role of uPAR as a versatile cellular signaling orchestrator, it is not surprising that both uPAR and suPAR have been implicated in various pathologies. suPAR is present in low concentrations in healthy individuals, where it has a role in neutrophil trafficking and stem cell mobilization.1 Serum concentrations are elevated in infectious diseases induced by various pathogenic species, including infection with viruses (HIV, Hanta),3,4 mycobacteria,5 and malaria.6 suPAR levels are also elevated in patients with inflammatory disorders, including arthritis7 and inflammatory bowel disease.8 On the basis of these data, suPAR has been considered a marker of (low-grade) activation of the immune system.9 The observed variation in suPAR concentrations holds prognostic information. In the general population, individuals with high suPAR concentrations are at an increased risk for cardiovascular events, independent from Framingham risk factors.10,11 In patients with non-ST elevation acute coronary syndrome, as well as in patients with ST-elevation myocardial infarction (MI), suPAR predicts all-cause mortality and recurrent MI.12,13 A substantial number of studies reported on suPAR in critically ill patients. In these patients, suPAR correlates with markers of organ dysfunction and is a predictor of both ICU mortality and long-term mortality beyond and above severity-of-disease classification systems such as APACHE-II or SOFA.9,14 In the kidneys, suPAR regulates the permeability of the glomerular filtration barrier, as it is involved in a signaling pathway leading to podocyte injury, podocyte effacement, and proteinuria.2,15 We recently demonstrated that in patients with chronic kidney disease (CKD), but without focal segmental glomerulosclerosis (FSGS), the glomerular filtration rate is a strong independent determinant of suPAR.16,17 Not unexpectedly, we and others noted very high concentrations of suPAR in patients with advanced CKD.16–18 Kidney International (2015) 87, 210–216
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Given that suPAR is a prognostic marker in various nonCKD populations and given the significant accumulation in patients at reduced glomerular filtration rate, we questioned whether suPAR is of clinical relevance in CKD. We therefore studied associations between suPAR and clinical outcomes in the Leuven mild-to-moderate CKD cohort.
clinical investigation
cholesterol (P ¼ 0.0004). There was no relationship with current smoking. Patients with a past history of cardiovascular event had significantly higher suPAR concentrations (Po0.0001) as compared with those without a previous cardiovascular event. suPAR and mortality
RESULTS Patients and demographics
We measured suPAR in patients at different stages of CKD included in the Leuven mild-to-moderate CKD cohort. As there is still some controversy over whether patients with FSGS have higher concentrations of suPAR, we excluded patients (n ¼ 13) with biopsy-proven FSGS. The study cohort thus consisted of 476 patients (Figure 1), with a median age of 64 years. There were slightly more men (n ¼ 260, 54.6%) than women (n ¼ 216, 45.4%). Baseline demographics and biochemical measurements are summarized in Table 1. When looking for determinants of serum suPAR concentrations using Spearman rank analyses, we found known positive associations with age, female sex, presence of diabetes, and the c-reactive protein (CRP), and known negative associations with serum albumin and the estimated glomerular filtration rate (eGFR) (Table 2). When looking at the conventional cardiovascular risk factors, higher suPAR was associated with higher age, higher blood pressure, and diabetes mellitus (Po0.0001 for all), but with lower serum
Assessed for eligibility (n =548)
We analyzed associations between suPAR and overall mortality, with a mean follow-up of 57 (standard deviation 2.7) months. During the study period, 52 patients died, 17 (32.7%) because of cardiovascular death (cause of death, Supplementary Table S1 online). The survival curve (Figure 2) for tertiles of suPAR demonstrated a clear association between suPAR and overall mortality (Po0.0001). In univariate Cox proportional hazards analysis, high suPAR was directly associated with mortality, as were higher age, a past history of cardiovascular disease, diabetes, higher CRP, hyperparathyroidism, and hyperphosphatemia. Anemia, hypoalbuminemia, hypocalcemia, and low eGFR were also associated with mortality (Supplementary Table S2 online). In multivariate analysis, suPAR was associated with mortality after correction for eGFR and the Framingham risk factors (age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol). (Table 3 and Supplementary Table S3 online). In the final model, suPAR was, however, not independently associated with overall mortality (P ¼ 0.08), presumably owing to lack of power. When looking at death due to cardiovascular cause only, high suPAR was significantly associated with a hazard ratio of 6.856 (2.880–16.320; P ¼ 0.00001). Owing to the low number of events (n ¼ 17), multivariate models for death due to cardiovascular cause have not been constructed. suPAR and cardiovascular disease
Excluded (n =49) Refused to participate (n =32) Prior renal replacement therapy (n =2) Enrolled in intervention study (n =4) Other (n =11)
Leuven mild-to-moderate CKD cohort (n = 499)
Excluded (n =23) FSGS patients (n =13) suPAR not measured (n =10)
Study cohort (n = 476)
Figure 1 | Patient enrollment during the study period. CKD, chronic kidney disease; FSGS, focal segmental glomerulosclerosis; suPAR, soluble urokinase-type plasminogen activator receptor. Kidney International (2015) 87, 210–216
Next, we explored the link between suPAR and cardiovascular disease in patients with mild-to-moderate CKD. During the study period, 76 patients had at least one cardiovascular event. We analyzed the time to first cardiovascular event (Supplementary Table S4 online for types of cardiovascular event). We observed a clear association between tertiles of suPAR and cardiovascular disease (Figure 3) (Po0.0001). In univariate Cox analyses, high suPAR was directly associated with cardiovascular events, as were higher age, a past history of cardiovascular disease, diabetes, higher blood pressure, higher CRP, hyperparathyroidism, hyperphosphatemia, and proteinuria. Anemia, hypoalbuminemia, hypocalcemia, and low eGFR were also associated with cardiovascular disease (Supplementary Table S5 online). In multivariate Cox analyses (Table 4 and Supplementary Table S6 online), the association between suPAR and cardiovascular events was independent after adjustment for kidney function and the Framingham risk factors. As we recently demonstrated that the eGFR is one of the strongest determinants of suPAR concentrations in patients with CKD,16 we performed additional analyses testing the nontransformed eGFR, the log-transformed eGFR, and 211
clinical investigation
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Table 1 | Baseline characteristics of the study population suPAR Variable
Overall (n ¼ 476)
Age (year) Gender: male/female (%) Prior CVD: yes/no (%) Diabetes: yes/no (%) Current smoker: yes/no (%) Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Hemoglobin (g/dl) Albumin (g/l) C-reactive protein (mg/l) Cholesterol (mg/dl) LDL (mg/dl) HDL (mg/dl) Calcium (mg/dl) Phosphate (mg/dl) Parathormone (ng/l) Creatinine (mg/dl) eGFR (ml/min per 1.73 m2) 24-hour proteinuria (g) Therapy with ACEI/ARB: yes/no (%) Therapy with statins: yes/no (%) Therapy with vit D: yes/no (%) Therapy with phosphate binder: yes/no (%) suPAR (pg/ml) suPAR cut-off (pg/ml)
64 260/216 132/344 86/390 86/390 25.69 135 80 13.3 44.9 2 178 85 57 9.6 3.3 24.6 1.79 34 0.27 333/143 224/252 86/390 130/346 3684.0
(51–75) (54.6/45.4) (27.7/72.3) (18.1/81.9) (18.1/81.9) (22.86–29.05) (120–150) (70–85) (1.8) (42.4–46.8) (1–6) (152–205) (67–112) (47–72) (9.2–9.9) (2.9–3.8) (12.8–53.7) (1.29–2.48) (23–54) (0.11–0.86) (70.0/30.0) (47.1/52.9) (18.1/91.9) (72.7/17.3) (2771.1–4994.5)
Tertile 1 (n ¼ 158) 51 96/62 19/139 10/148 31/127 24.91 130 80 14.1 46.2 1 184 92 61 9.6 3.0 14.4 1.23 62 0.16 107/51 56/102 15/143 21/137 2516.8
(39–62) (60.8/39.2) (12.0/88.0) (6.3/93.7) (19.6/80.4) (22.34–28.08) (120–150) (70–85) (1.8) (44.3–47.6) (1–3) (159–207) (71–118) (51–77) (9.3–9.9) (2.6–3.5) (5.5–24.5) (1.00–1.56) (46–78) (0.11–0.54) (67.7/32.3) (35.4/64.6) (9.5/90.5) (13.3/86.7) (2182.3–2770.1) o3093
Tertile 2 (n ¼ 159) 67 (56–75) 85/74 (53.5/46.5) 50/109 (31.4/68.6) 27/132 (17.0/83.0) 25/134 (15.7/84.3) 26.64 (23.39–30.12) 138 (120–150) 80 (70–85) 13.2 (1.7) 44.7 (42.8–46.8) 2 (1–5) 174 (151–207) 110 (85–136) 58 (46–71) 9.5 (9.3–9.9) 3.4 (3.0–3.7) 23.1 (13.5–48.0) 1.79 (1.45–2.36) 33 (26–41) 0.31 (0.10–0.90) 113/46 (71.1/28.9) 87/72 (54.7/45.3) 27/132 (17.0/83.0) 42/117 (26.4/73.6) 3677.8 (3333.2–3995.8) 3093–4333
Tertile 3 (n ¼ 159) 73 79/80 63/96 49/110 30/129 25.41 140 75 12.6 45.5 3 173 81 55 9.6 3.6 47.6 2.43 23 0.43 113/46 81/78 44/115 67/92 5219.4
(63–79) (49.7/50.3) (39.6/60.4) (30.8/69.2) (18.9/81.1) (22.99–29.02) (120–160) (70–80) (1.6) (43.5–47.0 ) (1–8) (144–192) (60–105) (45–70) (9.1–9.9) (3.2–4.1) (25–79.8) (1.94–3.14) (16–30) (0.12–1.29) (71.1/28.9) (50.94/46.06) (27.7/72.3) (42.1/57.9) (4690.4–6300.7) 44333
P o0.0001 0.05 o0.0001 o0.0001 0.86 0.03 0.004 0.01 o0.0001 o0.0001 o0.0001 0.007 0.006 0.03 0.39 o0.0001 o0.0001 o0.0001 o0.0001 0.003 0.52 0.006 o0.0001 o0.0001 o0.0001
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; suPAR, soluble urokinase-type plasminogen activator receptor; vit D, 25-hydroxy-vitamin D. Data are expressed as mean (s.d.) or median (IQR), as appropriate. Differences between tertiles were tested using parametric ANOVA, Kruskal–Wallis, or w2 test, as appropriate.
1/eGFR. The additional analyses did not materially alter the observed findings. In the final model, suPAR was significantly (HR 3.054 (1.690–5.517, P ¼ 0.0002)) associated with new-onset cardiovascular disease. Of note, analyses without exclusion of the patients with FSGS rendered identical results (data not shown). We analyzed the contribution of suPAR to risk estimation of cardiovascular disease using receiver operating characteristic statistics. As age is a strong determinant of cardiovascular outcome, we included age for all analyses. First, we compared the contribution of suPAR to the contribution of eGFR. In our population, suPAR is a stronger cardiovascular risk predictor than the eGFR (receiver operating characteristic AUC 0.7532). Second, we looked at the contribution of suPAR relative to the Framingham risk factors. Again, in these models, suPAR was an independent risk factor (receiver operating characteristic AUC 0.7817) (Supplementary Table S7 online). DISCUSSION
The suPAR is a versatile signaling orchestrator and has been considered a prognostic marker in various non-CKD populations. As suPAR accumulates in patients parallel to the loss of kidney function, we explored whether suPAR is associated with clinical outcomes in the Leuven mild-to212
moderate CKD cohort. High suPAR showed clear direct and graded associations with excess mortality and higher risk for new-onset cardiovascular disease. Mortality risk for patients with CKD is substantially greater than that for the general population, with cardiovascular disease as one of the prime contributors.19 Although part of this association can be explained by a higher prevalence and possibly larger impact of known risk factors, such as older age, hypertension, and diabetes,20 such ‘classical’ cardiovascular risk factors are insufficient to accurately predict cardiovascular risk in patients with CKD.21 We found strong graded associations with both crude mortality and cardiovascular disease. In the final model, the association with mortality was lost. The relationship with cardiovascular events persisted in the final model, raising questions on the role of suPAR in cardiovascular disease in patients with nondialysis-dependent CKD. We recently demonstrated that the eGFR is a strong determinant of suPAR concentrations.16 Given its molecular weight, between 20 and 50 kDa depending on proteolytic cleavage and degree of glycosylation, suPAR will be filtered by the glomeruli.15,22 Whether additional tubular handling is involved has not been studied to date. Urinary excretion of suPAR has been demonstrated.18,23 As kidney function in epidemiological studies is directly associated with cardiovascular outcomes,19 we also included eGFR in multivariate Kidney International (2015) 87, 210–216
clinical investigation
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Table 2 | Spearman’s rank correlation between suPAR serum concentration and baseline characteristics p
0.54 0.09 0.29 0.28 0.009 0.05 0.18 0.15 0.40 0.35 0.27 0.17 0.15 0.13 0.06 0.37 0.50 0.62 0.70 0.18 0.03 0.15 0.18 0.30
o0.0001 0.05 o0.0001 o0.0001 0.84 0.25 o0.0001 0.001 o0.0001 o0.0001 o0.0001 0.0004 0.001 0.006 0.21 o0.0001 o0.0001 o0.0001 o0.0001 0.003 0.57 0.001 o0.0001 o0.0001
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; 25-OH-vitamin D, 25-hydroxy-vitamin D; suPAR, soluble urokinase-type plasminogen activator receptor.
analyses. The observed graded association remained significant, independent of the eGFR. Moreover, renal handling is not the sole determinant of suPAR concentrations, as we observed a wide dispersion in serum concentrations, especially in patients at low eGFR.16 Apart from CKD-specific abnormalities, e.g., anemia, disturbances of bone and mineral metabolism, and retention of organic solutes,24 chronic inflammation has been implicated as a key player in the CKD-associated cardiovascular burden and excess mortality. Although the role of inflammation as an independent risk factor is better characterized in advanced stages of CKD,25 emerging data in CKD patients identify several inflammatory mediators as cardiovascular risk factors.26,27 As suPAR is proposed to be a marker of subclinical immunological activity, our findings may point to suPAR as just another biomarker of the malnutrition– inflammation–atherosclerosis complex. Although previous studies showed certain correlations between suPAR and other markers of inflammation,28 kinetic profiling of biomarkers after ST-elevation MI showed clear dissociation between changes in high-sensitivity CRP and suPAR concentrations.12 In our cohort, multivariate models adjusted for both albumin and CRP did not materially alter our finding of a graded association between suPAR and cardiovascular events (Table 4). Although this does not rule out the possibility of suPAR as a relevant biomarker of subclinical Kidney International (2015) 87, 210–216
90 Survival (percent)
Age Gender (female vs. male) Prior CVD Diabetes mellitus Current smoker Body mass index Systolic blood pressure Diastolic blood pressure Hemoglobin Albumin C-reactive protein Cholesterol LDL HDL Calcium Phosphate Parathormone Creatinine eGFR 24-h proteinuria Therapy with ACEI/ARB Therapy with statin Therapy with 25-OH-vitamin D Therapy with phosphate binder
r
80
70
60 P<0.0001 50 0
20
40 Time (months)
60
25
* Hazard ratio (95% confidence interval)
Variable
100
Tertile 1 Tertile 2 Tertile 3
*
20 15 10
*
*
*
5 R 0
Unadjusted
R Renal function
R
R
Framingham Full model risk factors
Figure 2 | Kaplan–Meier estimate of the fraction of patients with mild-to-moderate chronic kidney disease (CKD) who are alive as a function of tertiles of suPAR concentrations. There is a highly significant (Po0.0001) graded direct relationship between suPAR and the crude mortality risk. suPAR, soluble urokinase-type plasminogen activator receptor. R, reference. *Po0.05.
activation of the immune system, it appears that suPAR holds prognostic information beyond widely used markers of the malnutrition–inflammation–atherosclerosis complex. Finally, suPAR may reflect the (subclinical) atherosclerotic burden. Apart from tumor cells and monocytes, endothelial cells are a well-recognized source of circulating suPAR. In vitro, endothelial cells show clear polarization with apical expression of uPAR and basolateral release of suPAR.29 In vivo, suPAR is associated with inflammation in the vulnerable atherosclerotic plaque and is higher in patients with a symptomatic carotid lesion as compared with asymptomatic individuals.30 In a different study, suPAR was an independent determinant of the carotid intima-media thickness.31 suPAR may thus be a marker of established cardiovascular disease. In favor of this hypothesis, suPAR concentrations were significantly (Po0.0001) higher in those with a past medical history of cardiovascular disease. To take this into account, in multivariate analyses, we corrected the observed association between suPAR and outcomes for past history of 213
clinical investigation
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Table 3 | Cox proportional hazard survival analysis (multivariate models) Hazard ratio (95% confidence interval)
Variable 1. Unadjusted: suPAR (Ln) 2. Renal function: eGFR (Ln) 3. Framingham risk factors: age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol 4. Full model: creatinine (Ln), age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol, calcium, phosphate, parathyroid hormone (Ln), c-reactive protein (Ln), albumin
5.354 5.450 4.101 2.169
(3.222–8.896) (3.263–9.102) (1.957–8.594) (0.992–5.105)
P o0.0001 o0.0001 0.0002 0.08
Abbreviations: eGFR, estimated glomerular filtration rate; Ln, natural logarithm; suPAR, soluble urokinase-type plasminogen activator receptor.
Free of cardiovascular disease (percent)
100 90 80 70 60 P<0.0001 50 0
20
40 Time (months)
60
Hazard ratio (95% confidence interval)
20
Tertile 1 Tertile 2 Tertile 3
*
15
* *
10
* *
*
*
*
5 R
R
R
R
0 Unadjusted
Renal function
Framingham Full model risk factors
Figure 3 | Kaplan–Meier estimate of time to first cardiovascular event in patients with mild-to-moderate chronic kidney disease (CKD) as a function of tertiles of soluble urokinase-type plasminogen activator receptor (suPAR) concentrations. R, reference. *Po0.05.
cardiovascular disease (Table 4). In this analysis, suPAR remained an independent determinant of cardiovascular events. A limitation of our study is that we were unable to identify a definite mechanistic explanation for the observed strong association between suPAR and cardiovascular events. There are several plausible mechanisms that may explain our findings, including chronic inflammation and/or overproduction of suPAR by diseased vasculature. Intriguingly, the observed association is independent both from known 214
classical and from CKD-specific cardiovascular risk factors. Moreover, the observed association between suPAR and cardiovascular disease has repeatedly been demonstrated in different patient populations, including in the general population. Additional studies, however, are required to conclude whether the observed associations point to a direct causal relationship or that suPAR is a prognostic biomarker for risk stratification in CKD. Our study for the first time demonstrates a role for suPAR measurements in patients with kidney disease beyond FSGS. This observation warrants further investigations, especially as our understanding of the excess cardiovascular mortality and morbidity in CKD is far from complete. In conclusion, the suPAR is directly associated with cardiovascular disease in patients with mild-to-moderate CKD. Whether this association reflects a causal relationship and what are the mechanisms involved remain to be explored. PATIENTS AND METHODS Patients This was an ancillary study of the Leuven mild-to-moderate CKD study, a prospective cohort to investigate the role of uremic retention solutes in patients with CKD who are not yet on dialysis (clinicaltrials.gov NCT00441623). Prevalent CKD patients, followed up at the nephrology outpatient clinic of the University Hospitals Leuven, 18 years of age or older were eligible for inclusion. Data on baseline demographics and cause of kidney disease were collected at the time of informed consent. The study was conducted according to the Declaration of Helsinki and approved by the ethics committee of the University Hospitals Leuven. Informed consent was obtained from all patients. In this cohort, we have demonstrated a graded association between p-cresyl sulfate and cardiovascular disease.32 As follow-up of the patients in the current study has been extended as compared with the original publication, we reanalyzed the original data taking into account this extended follow-up. In univariate analysis, we observed a strong significant association of free p-cresol with cardiovascular disease (HR 1.85, Po0.0001) similar to the original publication (HR 1.79, Po0.0001). This association persisted after multivariate adjustment (HR 1.42, Po0.003), in the same range as our original report (HR 1.39, Po0.04). Measurements At inclusion, blood was taken by venous puncture for baseline measurement of biochemistry using standard laboratory techniques. Creatinine was measured using an IDMS-traceable method. eGFR Kidney International (2015) 87, 210–216
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
clinical investigation
Table 4 | Cox proportional hazard analysis of time to first cardiovascular event (multivariate models) Hazard ratio (95% confidence interval)
Variable 1. Unadjusted: suPAR (Ln) 2. Renal function: eGFR (Ln) 3. Framingham risk factors: age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterola 4. Full model: creatinine (Ln), age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol, calcium, phosphate, parathyroid hormone (Ln), c-reactive protein (Ln), albumina
5.064 5.011 4.086 3.054
(3.287–7.802) (3.248–7.730) (2.387–6.993) (1.690–5.517)
P o0.0001 o0.0001 o0.0001 0.0002
Abbreviations: eGFR, estimated glomerular filtration rate; Ln, natural logarithm; suPAR, soluble urokinase-type plasminogen activator receptor. a Age and cholesterol were entered as time-dependent covariates.
was calculated using the CKD-EPI equation.33 Measurement of suPAR was performed using the Human uPAR Quantikine ELISA kit (DUP00, R&D systems GmbH, Wiesbaden-Nordenstadt, Germany). For determination of suPAR, all samples were measured twice and mean suPAR concentrations were used. Reproducibility of the assay was excellent, with a mean coefficient of variation of 2.43% in individual patients. End-point evaluation Patients were followed up at the nephrology CKD outpatient clinic at 3- to 6-month intervals. Patients were prospectively followed up until December 31, 2010. If information could not be obtained, the patient was assumed to be lost to follow-up starting from the date of the last actual visit. Cardiovascular end points were prospectively recorded and coded, blinded from clinical and biochemical data. The primary end points were overall mortality and time to first cardiovascular event. The latter was a composite of death from cardiac causes, nonlethal MI, myocardial ischemia, coronary intervention, ischemic stroke, or new-onset peripheral vascular disease, whichever occurred first. Only one event per subject was included in the analysis. After review of available information, cause of death was classified as either cardiovascular, infectious, malignancy, or other. Cardiovascular deaths included fatal MI, sudden death, and death due to congestive heart failure. Cases of unobserved sudden death were considered cardiovascular death only when other potential causes could be excluded. Otherwise, they were classified as ‘other cause of death’. Out-of-hospital deaths were coded after consultation of the general practitioner. Nonlethal cardiovascular events included MI, diagnosed based on elevated levels of cardiac enzymes and/or typical electrocardiography changes, myocardial ischemia with typical electrocardiography changes without elevated cardiac enzymes, and coronary intervention (thrombolysis, percutaneous coronary intervention, or coronary artery bypass grafting). Ischemic stroke was defined as a neurological deficit lasting more than 24 h. Hemorrhagic stroke was excluded from the primary end point. Peripheral vascular disease included new-onset ischemic pain in the lower limbs, with abnormal ankle brachial pressure index or radiological evidence of peripheral vascular disease, new-onset ischemic necrotic lesions, or surgical arterial intervention. Statistics Data are expressed as mean (standard deviation) for normally distributed variables or median (interquartile range) for nonnormally distributed variables. Differences between baseline variables according to tertiles of suPAR levels were tested using analysis of variance, Kruskal–Wallis test, or Chi-square test, as appropriate. Kidney International (2015) 87, 210–216
Correlations between levels of suPAR and other variables were calculated by Spearman’s rank correlation coefficients. Variables following a non-normal distribution were log-transformed using the natural logarithm. Time to mortality and time to first cardiovascular event analysis were performed using Cox proportional hazards analysis. For Cox proportional hazard analysis of first cardiovascular event, data were censored at the start of renal replacement therapy, death other than cardiovascular, loss to follow-up, or at the end of the study observation period. A two-sided Po0.05 was considered statistical significant. For multivariate analysis, we used a double backward elimination approach, with inclusion of all variables at Po0.20 for secondary backward elimination at Po0.05. In case the proportionality assumption was violated, follow-up timeadjusted variables were entered as covariates. All statistical analyses were performed using SAS (version 9.2, the SAS Institute, Cary, NC). DISCLOSURE
We report as potential conflict of interest that RD and MS are employees of Gambro Dialysatoren GmbH, 72379 Hechingen, Germany. ACKNOWLEDGMENTS
RP is the recipient of a Ph.D. fellowship of the Research Foundation— Flanders (FWO grant 11E9813N). These results were presented in part at the American Society of Nephrology Renal Week, Atlanta, 2013. SUPPLEMENTARY MATERIAL Table S1. Cause of death. Table S2. Cox proportional hazard survival analysis (univariate). Table S3. Cox proportional hazard survival analysis (multivariate models). Table S4. Cardiovascular events. Table S5. Cox proportional hazard analysis of time to first cardiovascular event (univariate). Table S6. Cox proportional hazard analysis of time to first cardiovascular event (multivariate models). Table S7. receiver operating characteristic (ROC) statistics of new cardiovascular events. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki REFERENCES 1. 2. 3.
Blasi F, Carmeliet P. uPAR: a versatile signalling orchestrator. Nat Rev Mol Cell Biol 2002; 3: 932–943. Wei C, Moller CC, Altintas MM et al. Modification of kidney barrier function by the urokinase receptor. Nat Med 2008; 14: 55–63. Nebuloni M, Zawada L, Ferri A et al. HIV-1 infected lymphoid organs upregulate expression and release of the cleaved form of uPAR that modulates chemotaxis and virus expression. PLoS One 2013; 8: e70606.
215
clinical investigation
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15. 16.
17.
216
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Outinen TK, Tervo L, Makela S et al. Plasma levels of soluble urokinasetype plasminogen activator receptor associate with the clinical severity of acute Puumala hantavirus infection. PLoS One 2013; 8: e71335. Rabna P, Andersen A, Wejse C et al. Utility of the plasma level of suPAR in monitoring risk of mortality during TB treatment. PLoS One 2012; 7: e43933. Ostrowski SR, Ullum H, Goka BQ et al. Plasma concentrations of soluble urokinase-type plasminogen activator receptor are increased in patients with malaria and are associated with a poor clinical or a fatal outcome. J Infect Dis 2005; 191: 1331–1341. Slot O, Brunner N, Locht H et al. Soluble urokinase plasminogen activator receptor in plasma of patients with inflammatory rheumatic disorders: increased concentrations in rheumatoid arthritis. Ann Rheum Dis 1999; 58: 488–492. Lonnkvist MH, Theodorsson E, Holst M et al. Blood chemistry markers for evaluation of inflammatory activity in Crohn’s disease during infliximab therapy. Scand J Gastroenterol 2011; 46: 420–427. Backes Y, van der Sluijs KF, Mackie DP et al. Usefulness of suPAR as a biological marker in patients with systemic inflammation or infection: a systematic review. Intensive Care Med 2012; 38: 1418–1428. Lyngbaek S, Marott JL, Sehestedt T et al. Cardiovascular risk prediction in the general population with use of suPAR, CRP, and Framingham Risk Score. Int J Cardiol 2013; 167: 2904–2911. Persson M, Engstrom G, Bjorkbacka H et al. Soluble urokinase plasminogen activator receptor in plasma is associated with incidence of CVD. Results from the Malmo Diet and Cancer Study. Atherosclerosis 2012; 220: 502–505. Lyngbaek S, Marott JL, Moller DV et al. Usefulness of soluble urokinase plasminogen activator receptor to predict repeat myocardial infarction and mortality in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous intervention. Am J Cardiol 2012; 110: 1756–1763. Lyngbaek S, Andersson C, Marott JL et al. Soluble urokinase plasminogen activator receptor for risk prediction in patients admitted with acute chest pain. Clin Chem 2013; 59: 1621–1629. Koch A, Voigt S, Kruschinski C et al. Circulating soluble urokinase plasminogen activator receptor is stably elevated during the first week of treatment in the intensive care unit and predicts mortality in critically ill patients. Crit Care 2011; 15: R63. Wei C, El HS, Li J et al. Circulating urokinase receptor as a cause of focal segmental glomerulosclerosis. Nat Med 2011; 17: 952–960. Meijers B, Maas RJ, Sprangers B et al. The soluble urokinase receptor is not a clinical marker for focal segmental glomerulosclerosis. Kidney Int 2014; 85: 636–640. Bock ME, Price HE, Gallon L et al. Serum soluble urokinase-type plasminogen activator receptor levels and idiopathic FSGS in children: a single-center report. Clin J Am Soc Nephrol 2013; 8: 1304–1311.
18.
19.
20. 21.
22. 23. 24. 25.
26.
27.
28.
29. 30.
31.
32.
33.
Franco Palacios CR, Lieske JC, Wadei HM et al. Urine but not serum soluble urokinase receptor (suPAR) may identify cases of recurrent FSGS in kidney transplant candidates. Transplantation 2013; 96: 394–399. Go AS, Chertow GM, Fan D et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: 1296–1305. Foley R, Parfrey P, Sarnak M. The clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998; 32: S112–S119. Sarnak MJ, Levey AS, Schoolwerth AC et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the american heart association councils on kidney in cardiovascular disease, high blood pressure research, clinical cardiology, and epidemiology and prevention. Circulation 2003; 108: 2154–2169. Maas RJ, Deegens JK, Wetzels JF. Serum suPAR in patients with FSGS: trash or treasure? Pediatr Nephrol 2013; 28: 1041–1048. Naesens M, Meijers B, Sprangers B. suPAR and FSGS: the gap between bench and bedside. Transplantation 2013; 96: 368–369. Meyer TW, Hostetter TH. Uremia. N Engl J Med 2007; 357: 1316–1325. Stinghen AE, Bucharles S, Riella MC et al. Immune mechanisms involved in cardiovascular complications of chronic kidney disease. Blood Purif 2010; 29: 114–120. Carrero JJ, Ortiz A, Qureshi AR et al. Additive effects of soluble TWEAK and inflammation on mortality in hemodialysis patients. Clin J Am Soc Nephrol 2009; 4: 110–118. Liabeuf S, Barreto DV, Barreto FC et al. The circulating soluble TRAIL is a negative marker for inflammation inversely associated with the mortality risk in chronic kidney disease patients. Nephrol Dial Transplant 2010; 25: 2596–2602. Almroth G, Lonn J, Uhlin F et al. Fibroblast growth factor 23, hepatocyte growth factor, interleukin-6, high-sensitivity C-reactive protein and soluble urokinase plasminogen activator receptor. Inflammation markers in chronic haemodialysis patients? Scand J Immunol 2013; 78: 285–290. Chavakis T, Willuweit AK, Lupu F et al. Release of soluble urokinase receptor from vascular cells. Thromb Haemost 2001; 86: 686–693. Edsfeldt A, Nitulescu M, Grufman H et al. Soluble urokinase plasminogen activator receptor is associated with inflammation in the vulnerable human atherosclerotic plaque. Stroke 2012; 43: 3305–3312. Pawlak K, Mysliwiec M, Pawlak D. The urokinase-type plasminogen activator/its soluble receptor system is independently related to carotid atherosclerosis and associated with CC-chemokines in uraemic patients. Thromb Res 2008; 122: 328–335. Meijers BK, Claes K, Bammens B et al. p-Cresol and cardiovascular risk in mild-to-moderate kidney disease. Clin J Am Soc Nephrol 2010; 5: 1182–1189. Levey AS, Stevens LA, Schmid CH et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150: 604–612.
Kidney International (2015) 87, 210–216
http://www.kidney-international.org & 2014 International Society of Nephrology
Soluble urokinase receptor is a biomarker of cardiovascular disease in chronic kidney disease Bjo¨rn Meijers1,2, Ruben Poesen2, Kathleen Claes1,2, Ruth Dietrich3, Bert Bammens1,2, Ben Sprangers1,2, Maarten Naesens1,2, Markus Storr3, Dirk Kuypers1,2 and Pieter Evenepoel1,2 1
UZ Leuven, Department of Nephrology, Leuven, Belgium; 2KU Leuven, Department of Microbiology and Immunology, Leuven, Belgium and 3Research and Development, Gambro Dialysatoren GmbH, Hechingen, Germany
Soluble urokinase-type plasminogen activator receptor (suPAR) accumulates in patients with chronic kidney disease (CKD). In various non-CKD populations, suPAR has been proposed as a prognostic marker for mortality and cardiovascular disease. However, it is not known whether suPAR holds prognostic information in patients with mild-tomoderate CKD. In a prospective observational study of 476 patients with mild-to-moderate kidney disease, we examined multivariate associations between suPAR, overall mortality, and cardiovascular events. After a mean follow-up of 57 months, 52 patients died and 76 patients had at least one fatal or nonfatal cardiovascular event. Higher suPAR was directly and significantly associated with both overall mortality (univariate hazard ratio 5.35) and cardiovascular events (univariate hazard ratio 5.06). In multivariate analysis, suPAR remained significantly associated with cardiovascular events (full model, hazard ratio 3.05). Thus, in patients with mild-to-moderate CKD, suPAR concentrations show a clear, direct, and graded association with a higher risk for new-onset cardiovascular disease. Kidney International (2015) 87, 210–216; doi:10.1038/ki.2014.197; published online 4 June 2014 KEYWORDS: cardiovascular events; chronic kidney disease; urokinase
Correspondence: Bjo¨rn Meijers, UZ Leuven, Department of Nephrology, Herestraat 49, Leuven 3000, Belgium. E-mail: [email protected] Received 16 December 2013; revised 26 March 2014; accepted 3 April 2014; published online 4 June 2014 210
Soluble urokinase-type plasminogen activator receptor (suPAR) originates from proteolytic cleavage of the urokinasetype plasminogen activator receptor (uPAR) at its glycosyl phosphatidyl inosytol anchor site.1,2 uPAR elicits a plethora of cellular responses that include cellular adhesion, differentiation, proliferation, and migration.1 Given the role of uPAR as a versatile cellular signaling orchestrator, it is not surprising that both uPAR and suPAR have been implicated in various pathologies. suPAR is present in low concentrations in healthy individuals, where it has a role in neutrophil trafficking and stem cell mobilization.1 Serum concentrations are elevated in infectious diseases induced by various pathogenic species, including infection with viruses (HIV, Hanta),3,4 mycobacteria,5 and malaria.6 suPAR levels are also elevated in patients with inflammatory disorders, including arthritis7 and inflammatory bowel disease.8 On the basis of these data, suPAR has been considered a marker of (low-grade) activation of the immune system.9 The observed variation in suPAR concentrations holds prognostic information. In the general population, individuals with high suPAR concentrations are at an increased risk for cardiovascular events, independent from Framingham risk factors.10,11 In patients with non-ST elevation acute coronary syndrome, as well as in patients with ST-elevation myocardial infarction (MI), suPAR predicts all-cause mortality and recurrent MI.12,13 A substantial number of studies reported on suPAR in critically ill patients. In these patients, suPAR correlates with markers of organ dysfunction and is a predictor of both ICU mortality and long-term mortality beyond and above severity-of-disease classification systems such as APACHE-II or SOFA.9,14 In the kidneys, suPAR regulates the permeability of the glomerular filtration barrier, as it is involved in a signaling pathway leading to podocyte injury, podocyte effacement, and proteinuria.2,15 We recently demonstrated that in patients with chronic kidney disease (CKD), but without focal segmental glomerulosclerosis (FSGS), the glomerular filtration rate is a strong independent determinant of suPAR.16,17 Not unexpectedly, we and others noted very high concentrations of suPAR in patients with advanced CKD.16–18 Kidney International (2015) 87, 210–216
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Given that suPAR is a prognostic marker in various nonCKD populations and given the significant accumulation in patients at reduced glomerular filtration rate, we questioned whether suPAR is of clinical relevance in CKD. We therefore studied associations between suPAR and clinical outcomes in the Leuven mild-to-moderate CKD cohort.
clinical investigation
cholesterol (P ¼ 0.0004). There was no relationship with current smoking. Patients with a past history of cardiovascular event had significantly higher suPAR concentrations (Po0.0001) as compared with those without a previous cardiovascular event. suPAR and mortality
RESULTS Patients and demographics
We measured suPAR in patients at different stages of CKD included in the Leuven mild-to-moderate CKD cohort. As there is still some controversy over whether patients with FSGS have higher concentrations of suPAR, we excluded patients (n ¼ 13) with biopsy-proven FSGS. The study cohort thus consisted of 476 patients (Figure 1), with a median age of 64 years. There were slightly more men (n ¼ 260, 54.6%) than women (n ¼ 216, 45.4%). Baseline demographics and biochemical measurements are summarized in Table 1. When looking for determinants of serum suPAR concentrations using Spearman rank analyses, we found known positive associations with age, female sex, presence of diabetes, and the c-reactive protein (CRP), and known negative associations with serum albumin and the estimated glomerular filtration rate (eGFR) (Table 2). When looking at the conventional cardiovascular risk factors, higher suPAR was associated with higher age, higher blood pressure, and diabetes mellitus (Po0.0001 for all), but with lower serum
Assessed for eligibility (n =548)
We analyzed associations between suPAR and overall mortality, with a mean follow-up of 57 (standard deviation 2.7) months. During the study period, 52 patients died, 17 (32.7%) because of cardiovascular death (cause of death, Supplementary Table S1 online). The survival curve (Figure 2) for tertiles of suPAR demonstrated a clear association between suPAR and overall mortality (Po0.0001). In univariate Cox proportional hazards analysis, high suPAR was directly associated with mortality, as were higher age, a past history of cardiovascular disease, diabetes, higher CRP, hyperparathyroidism, and hyperphosphatemia. Anemia, hypoalbuminemia, hypocalcemia, and low eGFR were also associated with mortality (Supplementary Table S2 online). In multivariate analysis, suPAR was associated with mortality after correction for eGFR and the Framingham risk factors (age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol). (Table 3 and Supplementary Table S3 online). In the final model, suPAR was, however, not independently associated with overall mortality (P ¼ 0.08), presumably owing to lack of power. When looking at death due to cardiovascular cause only, high suPAR was significantly associated with a hazard ratio of 6.856 (2.880–16.320; P ¼ 0.00001). Owing to the low number of events (n ¼ 17), multivariate models for death due to cardiovascular cause have not been constructed. suPAR and cardiovascular disease
Excluded (n =49) Refused to participate (n =32) Prior renal replacement therapy (n =2) Enrolled in intervention study (n =4) Other (n =11)
Leuven mild-to-moderate CKD cohort (n = 499)
Excluded (n =23) FSGS patients (n =13) suPAR not measured (n =10)
Study cohort (n = 476)
Figure 1 | Patient enrollment during the study period. CKD, chronic kidney disease; FSGS, focal segmental glomerulosclerosis; suPAR, soluble urokinase-type plasminogen activator receptor. Kidney International (2015) 87, 210–216
Next, we explored the link between suPAR and cardiovascular disease in patients with mild-to-moderate CKD. During the study period, 76 patients had at least one cardiovascular event. We analyzed the time to first cardiovascular event (Supplementary Table S4 online for types of cardiovascular event). We observed a clear association between tertiles of suPAR and cardiovascular disease (Figure 3) (Po0.0001). In univariate Cox analyses, high suPAR was directly associated with cardiovascular events, as were higher age, a past history of cardiovascular disease, diabetes, higher blood pressure, higher CRP, hyperparathyroidism, hyperphosphatemia, and proteinuria. Anemia, hypoalbuminemia, hypocalcemia, and low eGFR were also associated with cardiovascular disease (Supplementary Table S5 online). In multivariate Cox analyses (Table 4 and Supplementary Table S6 online), the association between suPAR and cardiovascular events was independent after adjustment for kidney function and the Framingham risk factors. As we recently demonstrated that the eGFR is one of the strongest determinants of suPAR concentrations in patients with CKD,16 we performed additional analyses testing the nontransformed eGFR, the log-transformed eGFR, and 211
clinical investigation
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Table 1 | Baseline characteristics of the study population suPAR Variable
Overall (n ¼ 476)
Age (year) Gender: male/female (%) Prior CVD: yes/no (%) Diabetes: yes/no (%) Current smoker: yes/no (%) Body mass index (kg/m2) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) Hemoglobin (g/dl) Albumin (g/l) C-reactive protein (mg/l) Cholesterol (mg/dl) LDL (mg/dl) HDL (mg/dl) Calcium (mg/dl) Phosphate (mg/dl) Parathormone (ng/l) Creatinine (mg/dl) eGFR (ml/min per 1.73 m2) 24-hour proteinuria (g) Therapy with ACEI/ARB: yes/no (%) Therapy with statins: yes/no (%) Therapy with vit D: yes/no (%) Therapy with phosphate binder: yes/no (%) suPAR (pg/ml) suPAR cut-off (pg/ml)
64 260/216 132/344 86/390 86/390 25.69 135 80 13.3 44.9 2 178 85 57 9.6 3.3 24.6 1.79 34 0.27 333/143 224/252 86/390 130/346 3684.0
(51–75) (54.6/45.4) (27.7/72.3) (18.1/81.9) (18.1/81.9) (22.86–29.05) (120–150) (70–85) (1.8) (42.4–46.8) (1–6) (152–205) (67–112) (47–72) (9.2–9.9) (2.9–3.8) (12.8–53.7) (1.29–2.48) (23–54) (0.11–0.86) (70.0/30.0) (47.1/52.9) (18.1/91.9) (72.7/17.3) (2771.1–4994.5)
Tertile 1 (n ¼ 158) 51 96/62 19/139 10/148 31/127 24.91 130 80 14.1 46.2 1 184 92 61 9.6 3.0 14.4 1.23 62 0.16 107/51 56/102 15/143 21/137 2516.8
(39–62) (60.8/39.2) (12.0/88.0) (6.3/93.7) (19.6/80.4) (22.34–28.08) (120–150) (70–85) (1.8) (44.3–47.6) (1–3) (159–207) (71–118) (51–77) (9.3–9.9) (2.6–3.5) (5.5–24.5) (1.00–1.56) (46–78) (0.11–0.54) (67.7/32.3) (35.4/64.6) (9.5/90.5) (13.3/86.7) (2182.3–2770.1) o3093
Tertile 2 (n ¼ 159) 67 (56–75) 85/74 (53.5/46.5) 50/109 (31.4/68.6) 27/132 (17.0/83.0) 25/134 (15.7/84.3) 26.64 (23.39–30.12) 138 (120–150) 80 (70–85) 13.2 (1.7) 44.7 (42.8–46.8) 2 (1–5) 174 (151–207) 110 (85–136) 58 (46–71) 9.5 (9.3–9.9) 3.4 (3.0–3.7) 23.1 (13.5–48.0) 1.79 (1.45–2.36) 33 (26–41) 0.31 (0.10–0.90) 113/46 (71.1/28.9) 87/72 (54.7/45.3) 27/132 (17.0/83.0) 42/117 (26.4/73.6) 3677.8 (3333.2–3995.8) 3093–4333
Tertile 3 (n ¼ 159) 73 79/80 63/96 49/110 30/129 25.41 140 75 12.6 45.5 3 173 81 55 9.6 3.6 47.6 2.43 23 0.43 113/46 81/78 44/115 67/92 5219.4
(63–79) (49.7/50.3) (39.6/60.4) (30.8/69.2) (18.9/81.1) (22.99–29.02) (120–160) (70–80) (1.6) (43.5–47.0 ) (1–8) (144–192) (60–105) (45–70) (9.1–9.9) (3.2–4.1) (25–79.8) (1.94–3.14) (16–30) (0.12–1.29) (71.1/28.9) (50.94/46.06) (27.7/72.3) (42.1/57.9) (4690.4–6300.7) 44333
P o0.0001 0.05 o0.0001 o0.0001 0.86 0.03 0.004 0.01 o0.0001 o0.0001 o0.0001 0.007 0.006 0.03 0.39 o0.0001 o0.0001 o0.0001 o0.0001 0.003 0.52 0.006 o0.0001 o0.0001 o0.0001
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; suPAR, soluble urokinase-type plasminogen activator receptor; vit D, 25-hydroxy-vitamin D. Data are expressed as mean (s.d.) or median (IQR), as appropriate. Differences between tertiles were tested using parametric ANOVA, Kruskal–Wallis, or w2 test, as appropriate.
1/eGFR. The additional analyses did not materially alter the observed findings. In the final model, suPAR was significantly (HR 3.054 (1.690–5.517, P ¼ 0.0002)) associated with new-onset cardiovascular disease. Of note, analyses without exclusion of the patients with FSGS rendered identical results (data not shown). We analyzed the contribution of suPAR to risk estimation of cardiovascular disease using receiver operating characteristic statistics. As age is a strong determinant of cardiovascular outcome, we included age for all analyses. First, we compared the contribution of suPAR to the contribution of eGFR. In our population, suPAR is a stronger cardiovascular risk predictor than the eGFR (receiver operating characteristic AUC 0.7532). Second, we looked at the contribution of suPAR relative to the Framingham risk factors. Again, in these models, suPAR was an independent risk factor (receiver operating characteristic AUC 0.7817) (Supplementary Table S7 online). DISCUSSION
The suPAR is a versatile signaling orchestrator and has been considered a prognostic marker in various non-CKD populations. As suPAR accumulates in patients parallel to the loss of kidney function, we explored whether suPAR is associated with clinical outcomes in the Leuven mild-to212
moderate CKD cohort. High suPAR showed clear direct and graded associations with excess mortality and higher risk for new-onset cardiovascular disease. Mortality risk for patients with CKD is substantially greater than that for the general population, with cardiovascular disease as one of the prime contributors.19 Although part of this association can be explained by a higher prevalence and possibly larger impact of known risk factors, such as older age, hypertension, and diabetes,20 such ‘classical’ cardiovascular risk factors are insufficient to accurately predict cardiovascular risk in patients with CKD.21 We found strong graded associations with both crude mortality and cardiovascular disease. In the final model, the association with mortality was lost. The relationship with cardiovascular events persisted in the final model, raising questions on the role of suPAR in cardiovascular disease in patients with nondialysis-dependent CKD. We recently demonstrated that the eGFR is a strong determinant of suPAR concentrations.16 Given its molecular weight, between 20 and 50 kDa depending on proteolytic cleavage and degree of glycosylation, suPAR will be filtered by the glomeruli.15,22 Whether additional tubular handling is involved has not been studied to date. Urinary excretion of suPAR has been demonstrated.18,23 As kidney function in epidemiological studies is directly associated with cardiovascular outcomes,19 we also included eGFR in multivariate Kidney International (2015) 87, 210–216
clinical investigation
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Table 2 | Spearman’s rank correlation between suPAR serum concentration and baseline characteristics p
0.54 0.09 0.29 0.28 0.009 0.05 0.18 0.15 0.40 0.35 0.27 0.17 0.15 0.13 0.06 0.37 0.50 0.62 0.70 0.18 0.03 0.15 0.18 0.30
o0.0001 0.05 o0.0001 o0.0001 0.84 0.25 o0.0001 0.001 o0.0001 o0.0001 o0.0001 0.0004 0.001 0.006 0.21 o0.0001 o0.0001 o0.0001 o0.0001 0.003 0.57 0.001 o0.0001 o0.0001
Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; 25-OH-vitamin D, 25-hydroxy-vitamin D; suPAR, soluble urokinase-type plasminogen activator receptor.
analyses. The observed graded association remained significant, independent of the eGFR. Moreover, renal handling is not the sole determinant of suPAR concentrations, as we observed a wide dispersion in serum concentrations, especially in patients at low eGFR.16 Apart from CKD-specific abnormalities, e.g., anemia, disturbances of bone and mineral metabolism, and retention of organic solutes,24 chronic inflammation has been implicated as a key player in the CKD-associated cardiovascular burden and excess mortality. Although the role of inflammation as an independent risk factor is better characterized in advanced stages of CKD,25 emerging data in CKD patients identify several inflammatory mediators as cardiovascular risk factors.26,27 As suPAR is proposed to be a marker of subclinical immunological activity, our findings may point to suPAR as just another biomarker of the malnutrition– inflammation–atherosclerosis complex. Although previous studies showed certain correlations between suPAR and other markers of inflammation,28 kinetic profiling of biomarkers after ST-elevation MI showed clear dissociation between changes in high-sensitivity CRP and suPAR concentrations.12 In our cohort, multivariate models adjusted for both albumin and CRP did not materially alter our finding of a graded association between suPAR and cardiovascular events (Table 4). Although this does not rule out the possibility of suPAR as a relevant biomarker of subclinical Kidney International (2015) 87, 210–216
90 Survival (percent)
Age Gender (female vs. male) Prior CVD Diabetes mellitus Current smoker Body mass index Systolic blood pressure Diastolic blood pressure Hemoglobin Albumin C-reactive protein Cholesterol LDL HDL Calcium Phosphate Parathormone Creatinine eGFR 24-h proteinuria Therapy with ACEI/ARB Therapy with statin Therapy with 25-OH-vitamin D Therapy with phosphate binder
r
80
70
60 P<0.0001 50 0
20
40 Time (months)
60
25
* Hazard ratio (95% confidence interval)
Variable
100
Tertile 1 Tertile 2 Tertile 3
*
20 15 10
*
*
*
5 R 0
Unadjusted
R Renal function
R
R
Framingham Full model risk factors
Figure 2 | Kaplan–Meier estimate of the fraction of patients with mild-to-moderate chronic kidney disease (CKD) who are alive as a function of tertiles of suPAR concentrations. There is a highly significant (Po0.0001) graded direct relationship between suPAR and the crude mortality risk. suPAR, soluble urokinase-type plasminogen activator receptor. R, reference. *Po0.05.
activation of the immune system, it appears that suPAR holds prognostic information beyond widely used markers of the malnutrition–inflammation–atherosclerosis complex. Finally, suPAR may reflect the (subclinical) atherosclerotic burden. Apart from tumor cells and monocytes, endothelial cells are a well-recognized source of circulating suPAR. In vitro, endothelial cells show clear polarization with apical expression of uPAR and basolateral release of suPAR.29 In vivo, suPAR is associated with inflammation in the vulnerable atherosclerotic plaque and is higher in patients with a symptomatic carotid lesion as compared with asymptomatic individuals.30 In a different study, suPAR was an independent determinant of the carotid intima-media thickness.31 suPAR may thus be a marker of established cardiovascular disease. In favor of this hypothesis, suPAR concentrations were significantly (Po0.0001) higher in those with a past medical history of cardiovascular disease. To take this into account, in multivariate analyses, we corrected the observed association between suPAR and outcomes for past history of 213
clinical investigation
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Table 3 | Cox proportional hazard survival analysis (multivariate models) Hazard ratio (95% confidence interval)
Variable 1. Unadjusted: suPAR (Ln) 2. Renal function: eGFR (Ln) 3. Framingham risk factors: age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol 4. Full model: creatinine (Ln), age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol, calcium, phosphate, parathyroid hormone (Ln), c-reactive protein (Ln), albumin
5.354 5.450 4.101 2.169
(3.222–8.896) (3.263–9.102) (1.957–8.594) (0.992–5.105)
P o0.0001 o0.0001 0.0002 0.08
Abbreviations: eGFR, estimated glomerular filtration rate; Ln, natural logarithm; suPAR, soluble urokinase-type plasminogen activator receptor.
Free of cardiovascular disease (percent)
100 90 80 70 60 P<0.0001 50 0
20
40 Time (months)
60
Hazard ratio (95% confidence interval)
20
Tertile 1 Tertile 2 Tertile 3
*
15
* *
10
* *
*
*
*
5 R
R
R
R
0 Unadjusted
Renal function
Framingham Full model risk factors
Figure 3 | Kaplan–Meier estimate of time to first cardiovascular event in patients with mild-to-moderate chronic kidney disease (CKD) as a function of tertiles of soluble urokinase-type plasminogen activator receptor (suPAR) concentrations. R, reference. *Po0.05.
cardiovascular disease (Table 4). In this analysis, suPAR remained an independent determinant of cardiovascular events. A limitation of our study is that we were unable to identify a definite mechanistic explanation for the observed strong association between suPAR and cardiovascular events. There are several plausible mechanisms that may explain our findings, including chronic inflammation and/or overproduction of suPAR by diseased vasculature. Intriguingly, the observed association is independent both from known 214
classical and from CKD-specific cardiovascular risk factors. Moreover, the observed association between suPAR and cardiovascular disease has repeatedly been demonstrated in different patient populations, including in the general population. Additional studies, however, are required to conclude whether the observed associations point to a direct causal relationship or that suPAR is a prognostic biomarker for risk stratification in CKD. Our study for the first time demonstrates a role for suPAR measurements in patients with kidney disease beyond FSGS. This observation warrants further investigations, especially as our understanding of the excess cardiovascular mortality and morbidity in CKD is far from complete. In conclusion, the suPAR is directly associated with cardiovascular disease in patients with mild-to-moderate CKD. Whether this association reflects a causal relationship and what are the mechanisms involved remain to be explored. PATIENTS AND METHODS Patients This was an ancillary study of the Leuven mild-to-moderate CKD study, a prospective cohort to investigate the role of uremic retention solutes in patients with CKD who are not yet on dialysis (clinicaltrials.gov NCT00441623). Prevalent CKD patients, followed up at the nephrology outpatient clinic of the University Hospitals Leuven, 18 years of age or older were eligible for inclusion. Data on baseline demographics and cause of kidney disease were collected at the time of informed consent. The study was conducted according to the Declaration of Helsinki and approved by the ethics committee of the University Hospitals Leuven. Informed consent was obtained from all patients. In this cohort, we have demonstrated a graded association between p-cresyl sulfate and cardiovascular disease.32 As follow-up of the patients in the current study has been extended as compared with the original publication, we reanalyzed the original data taking into account this extended follow-up. In univariate analysis, we observed a strong significant association of free p-cresol with cardiovascular disease (HR 1.85, Po0.0001) similar to the original publication (HR 1.79, Po0.0001). This association persisted after multivariate adjustment (HR 1.42, Po0.003), in the same range as our original report (HR 1.39, Po0.04). Measurements At inclusion, blood was taken by venous puncture for baseline measurement of biochemistry using standard laboratory techniques. Creatinine was measured using an IDMS-traceable method. eGFR Kidney International (2015) 87, 210–216
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
clinical investigation
Table 4 | Cox proportional hazard analysis of time to first cardiovascular event (multivariate models) Hazard ratio (95% confidence interval)
Variable 1. Unadjusted: suPAR (Ln) 2. Renal function: eGFR (Ln) 3. Framingham risk factors: age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterola 4. Full model: creatinine (Ln), age, gender, systolic blood pressure, current smoker, diabetes mellitus, cholesterol, calcium, phosphate, parathyroid hormone (Ln), c-reactive protein (Ln), albumina
5.064 5.011 4.086 3.054
(3.287–7.802) (3.248–7.730) (2.387–6.993) (1.690–5.517)
P o0.0001 o0.0001 o0.0001 0.0002
Abbreviations: eGFR, estimated glomerular filtration rate; Ln, natural logarithm; suPAR, soluble urokinase-type plasminogen activator receptor. a Age and cholesterol were entered as time-dependent covariates.
was calculated using the CKD-EPI equation.33 Measurement of suPAR was performed using the Human uPAR Quantikine ELISA kit (DUP00, R&D systems GmbH, Wiesbaden-Nordenstadt, Germany). For determination of suPAR, all samples were measured twice and mean suPAR concentrations were used. Reproducibility of the assay was excellent, with a mean coefficient of variation of 2.43% in individual patients. End-point evaluation Patients were followed up at the nephrology CKD outpatient clinic at 3- to 6-month intervals. Patients were prospectively followed up until December 31, 2010. If information could not be obtained, the patient was assumed to be lost to follow-up starting from the date of the last actual visit. Cardiovascular end points were prospectively recorded and coded, blinded from clinical and biochemical data. The primary end points were overall mortality and time to first cardiovascular event. The latter was a composite of death from cardiac causes, nonlethal MI, myocardial ischemia, coronary intervention, ischemic stroke, or new-onset peripheral vascular disease, whichever occurred first. Only one event per subject was included in the analysis. After review of available information, cause of death was classified as either cardiovascular, infectious, malignancy, or other. Cardiovascular deaths included fatal MI, sudden death, and death due to congestive heart failure. Cases of unobserved sudden death were considered cardiovascular death only when other potential causes could be excluded. Otherwise, they were classified as ‘other cause of death’. Out-of-hospital deaths were coded after consultation of the general practitioner. Nonlethal cardiovascular events included MI, diagnosed based on elevated levels of cardiac enzymes and/or typical electrocardiography changes, myocardial ischemia with typical electrocardiography changes without elevated cardiac enzymes, and coronary intervention (thrombolysis, percutaneous coronary intervention, or coronary artery bypass grafting). Ischemic stroke was defined as a neurological deficit lasting more than 24 h. Hemorrhagic stroke was excluded from the primary end point. Peripheral vascular disease included new-onset ischemic pain in the lower limbs, with abnormal ankle brachial pressure index or radiological evidence of peripheral vascular disease, new-onset ischemic necrotic lesions, or surgical arterial intervention. Statistics Data are expressed as mean (standard deviation) for normally distributed variables or median (interquartile range) for nonnormally distributed variables. Differences between baseline variables according to tertiles of suPAR levels were tested using analysis of variance, Kruskal–Wallis test, or Chi-square test, as appropriate. Kidney International (2015) 87, 210–216
Correlations between levels of suPAR and other variables were calculated by Spearman’s rank correlation coefficients. Variables following a non-normal distribution were log-transformed using the natural logarithm. Time to mortality and time to first cardiovascular event analysis were performed using Cox proportional hazards analysis. For Cox proportional hazard analysis of first cardiovascular event, data were censored at the start of renal replacement therapy, death other than cardiovascular, loss to follow-up, or at the end of the study observation period. A two-sided Po0.05 was considered statistical significant. For multivariate analysis, we used a double backward elimination approach, with inclusion of all variables at Po0.20 for secondary backward elimination at Po0.05. In case the proportionality assumption was violated, follow-up timeadjusted variables were entered as covariates. All statistical analyses were performed using SAS (version 9.2, the SAS Institute, Cary, NC). DISCLOSURE
We report as potential conflict of interest that RD and MS are employees of Gambro Dialysatoren GmbH, 72379 Hechingen, Germany. ACKNOWLEDGMENTS
RP is the recipient of a Ph.D. fellowship of the Research Foundation— Flanders (FWO grant 11E9813N). These results were presented in part at the American Society of Nephrology Renal Week, Atlanta, 2013. SUPPLEMENTARY MATERIAL Table S1. Cause of death. Table S2. Cox proportional hazard survival analysis (univariate). Table S3. Cox proportional hazard survival analysis (multivariate models). Table S4. Cardiovascular events. Table S5. Cox proportional hazard analysis of time to first cardiovascular event (univariate). Table S6. Cox proportional hazard analysis of time to first cardiovascular event (multivariate models). Table S7. receiver operating characteristic (ROC) statistics of new cardiovascular events. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki REFERENCES 1. 2. 3.
Blasi F, Carmeliet P. uPAR: a versatile signalling orchestrator. Nat Rev Mol Cell Biol 2002; 3: 932–943. Wei C, Moller CC, Altintas MM et al. Modification of kidney barrier function by the urokinase receptor. Nat Med 2008; 14: 55–63. Nebuloni M, Zawada L, Ferri A et al. HIV-1 infected lymphoid organs upregulate expression and release of the cleaved form of uPAR that modulates chemotaxis and virus expression. PLoS One 2013; 8: e70606.
215
clinical investigation
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15. 16.
17.
216
B Meijers et al.: Soluble urokinase receptor is a biomarker of cardiovascular disease
Outinen TK, Tervo L, Makela S et al. Plasma levels of soluble urokinasetype plasminogen activator receptor associate with the clinical severity of acute Puumala hantavirus infection. PLoS One 2013; 8: e71335. Rabna P, Andersen A, Wejse C et al. Utility of the plasma level of suPAR in monitoring risk of mortality during TB treatment. PLoS One 2012; 7: e43933. Ostrowski SR, Ullum H, Goka BQ et al. Plasma concentrations of soluble urokinase-type plasminogen activator receptor are increased in patients with malaria and are associated with a poor clinical or a fatal outcome. J Infect Dis 2005; 191: 1331–1341. Slot O, Brunner N, Locht H et al. Soluble urokinase plasminogen activator receptor in plasma of patients with inflammatory rheumatic disorders: increased concentrations in rheumatoid arthritis. Ann Rheum Dis 1999; 58: 488–492. Lonnkvist MH, Theodorsson E, Holst M et al. Blood chemistry markers for evaluation of inflammatory activity in Crohn’s disease during infliximab therapy. Scand J Gastroenterol 2011; 46: 420–427. Backes Y, van der Sluijs KF, Mackie DP et al. Usefulness of suPAR as a biological marker in patients with systemic inflammation or infection: a systematic review. Intensive Care Med 2012; 38: 1418–1428. Lyngbaek S, Marott JL, Sehestedt T et al. Cardiovascular risk prediction in the general population with use of suPAR, CRP, and Framingham Risk Score. Int J Cardiol 2013; 167: 2904–2911. Persson M, Engstrom G, Bjorkbacka H et al. Soluble urokinase plasminogen activator receptor in plasma is associated with incidence of CVD. Results from the Malmo Diet and Cancer Study. Atherosclerosis 2012; 220: 502–505. Lyngbaek S, Marott JL, Moller DV et al. Usefulness of soluble urokinase plasminogen activator receptor to predict repeat myocardial infarction and mortality in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous intervention. Am J Cardiol 2012; 110: 1756–1763. Lyngbaek S, Andersson C, Marott JL et al. Soluble urokinase plasminogen activator receptor for risk prediction in patients admitted with acute chest pain. Clin Chem 2013; 59: 1621–1629. Koch A, Voigt S, Kruschinski C et al. Circulating soluble urokinase plasminogen activator receptor is stably elevated during the first week of treatment in the intensive care unit and predicts mortality in critically ill patients. Crit Care 2011; 15: R63. Wei C, El HS, Li J et al. Circulating urokinase receptor as a cause of focal segmental glomerulosclerosis. Nat Med 2011; 17: 952–960. Meijers B, Maas RJ, Sprangers B et al. The soluble urokinase receptor is not a clinical marker for focal segmental glomerulosclerosis. Kidney Int 2014; 85: 636–640. Bock ME, Price HE, Gallon L et al. Serum soluble urokinase-type plasminogen activator receptor levels and idiopathic FSGS in children: a single-center report. Clin J Am Soc Nephrol 2013; 8: 1304–1311.
18.
19.
20. 21.
22. 23. 24. 25.
26.
27.
28.
29. 30.
31.
32.
33.
Franco Palacios CR, Lieske JC, Wadei HM et al. Urine but not serum soluble urokinase receptor (suPAR) may identify cases of recurrent FSGS in kidney transplant candidates. Transplantation 2013; 96: 394–399. Go AS, Chertow GM, Fan D et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: 1296–1305. Foley R, Parfrey P, Sarnak M. The clinical epidemiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998; 32: S112–S119. Sarnak MJ, Levey AS, Schoolwerth AC et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the american heart association councils on kidney in cardiovascular disease, high blood pressure research, clinical cardiology, and epidemiology and prevention. Circulation 2003; 108: 2154–2169. Maas RJ, Deegens JK, Wetzels JF. Serum suPAR in patients with FSGS: trash or treasure? Pediatr Nephrol 2013; 28: 1041–1048. Naesens M, Meijers B, Sprangers B. suPAR and FSGS: the gap between bench and bedside. Transplantation 2013; 96: 368–369. Meyer TW, Hostetter TH. Uremia. N Engl J Med 2007; 357: 1316–1325. Stinghen AE, Bucharles S, Riella MC et al. Immune mechanisms involved in cardiovascular complications of chronic kidney disease. Blood Purif 2010; 29: 114–120. Carrero JJ, Ortiz A, Qureshi AR et al. Additive effects of soluble TWEAK and inflammation on mortality in hemodialysis patients. Clin J Am Soc Nephrol 2009; 4: 110–118. Liabeuf S, Barreto DV, Barreto FC et al. The circulating soluble TRAIL is a negative marker for inflammation inversely associated with the mortality risk in chronic kidney disease patients. Nephrol Dial Transplant 2010; 25: 2596–2602. Almroth G, Lonn J, Uhlin F et al. Fibroblast growth factor 23, hepatocyte growth factor, interleukin-6, high-sensitivity C-reactive protein and soluble urokinase plasminogen activator receptor. Inflammation markers in chronic haemodialysis patients? Scand J Immunol 2013; 78: 285–290. Chavakis T, Willuweit AK, Lupu F et al. Release of soluble urokinase receptor from vascular cells. Thromb Haemost 2001; 86: 686–693. Edsfeldt A, Nitulescu M, Grufman H et al. Soluble urokinase plasminogen activator receptor is associated with inflammation in the vulnerable human atherosclerotic plaque. Stroke 2012; 43: 3305–3312. Pawlak K, Mysliwiec M, Pawlak D. The urokinase-type plasminogen activator/its soluble receptor system is independently related to carotid atherosclerosis and associated with CC-chemokines in uraemic patients. Thromb Res 2008; 122: 328–335. Meijers BK, Claes K, Bammens B et al. p-Cresol and cardiovascular risk in mild-to-moderate kidney disease. Clin J Am Soc Nephrol 2010; 5: 1182–1189. Levey AS, Stevens LA, Schmid CH et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150: 604–612.
Kidney International (2015) 87, 210–216