University of Groningen Improving treatment and imaging in ADPKD van Gastel, Maatje Dirkje Adriana

Improving treatment and imaging in ADPKD

van Gastel, Maatje Dirkje Adriana

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2019

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

van Gastel, M. D. A. (2019). Improving treatment and imaging in ADPKD. Rijksuniversiteit Groningen.

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

Maatje D.A. van Gastel

Bart J. Kramers

Arlene B. Chapman

Vicente E. Torres

Esther Meijer

Ron T. Gansevoort

ABSTRACT

Background: In autosomal dominant polycystic kidney disease (ADPKD) vasopressin is causally involved with disease progression. We investigated in a cross-sectional study which modifiable lifestyle factors are associated with copeptin concentration (a surro-gate of vasopressin), because intervening on these factors could potentially influence the rate of disease progression.

Methods: Data were used of ADPKD patients participating in two observational cohort studies (the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease (CRISP) cohort, n=241, and the University Medical Center Groningen (UMCG) cohort, n=133). Copeptin concentration, which was measured by an immunoluminometric assay, glomerular filtration rate (GFR) by iothalamate clearance and total kidney volume (TKV) by magnetic resonance imaging.

Results: Of the participants (43.2% male, mean age 35.0±10.1 year) GFR was 99.7±30.9 ml/min and TKV 950.2 (IQR 650.9-1602.0) mL. Median copeptin concentration was 3.54 (IQR 2.17-7.29) pmol/L. Higher copeptin concentrations associated with higher albuminu-ria and TKV, and with lower GFR univaalbuminu-riately, and when adjusted for cohort, age, gender, race and height (both p<0.001). In a stepwise backward multivariable regression model, adjusted for non-modifiable factors cohort, age, gender, race, height, GFR and TKV,

(ad-justed R2_{=0.48) modifiable factors associated with higher copeptin concentration were:}

lower 24-hour urine volume (surrogate for fluid intake, p=0.001), higher sodium excre-tion (surrogate for sodium intake, p=0.004) and higher systolic blood pressure (p=0.01). No associations were found for urea excretion (surrogate for protein intake) and coffee consumption. When analyzing both cohorts separately similar results were obtained. Conclusions: In patients with ADPKD modifiable factors that are associated with higher copeptin concentration are increased systolic blood pressure, and especially low fluid and high sodium intake.

68

4

INTRODUCTION

The antidiuretic hormone vasopressin is critical for extracellular tonicity and homeosta-sis in humans. Recently, it has become clear that vasopressin can also have a detrimental role in patients with autosomal dominant polycystic kidney disease (ADPKD).

Vasopres-sin regulates cyst growth1_{, and vasopressin V2 receptor antagonists can prevent}

cys-togenesis and protects against kidney function decline in PKD rodent models2-4_.

In patients with ADPKD, copeptin - as surrogate marker for vasopressin – associates with

disease severity in cross-sectional studies5-7_{, and increase in total kidney volume (TKV)}8

as well as decline in estimated glomerular filtration rate (eGFR), both measures of

dis-ease progression, in longitudinal studies9-10_{. Most importantly, in two large randomized}

controlled clinical trials treatment with the vasopressin V2 receptor antagonist (V2RA) tolvaptan slowed the rates of increase in TKV11 and decline in eGFR when compared to

placebo11,12_.

Given the association between vasopressin and disease progression in ADPKD, it has become of interest to identify lifestyle factors that are associated with vasopressin con-centration, because interventions that target these factors could potentially slow the rate of disease progression.

Vasopressin is secreted directly in response to increased plasma osmolality and to

de-creased intravascular volume, both dependent on fluid and osmolar intake13_{. In turn,}

osmolar intake is determined by the salt and protein content of food. It is well estab-lished that shortly after salt or protein intake, vasopressin increases. However, whether relatively stable long-term ad libitum high intake of salt and protein influences systemic vasopressin concentration is not known.

Besides these dietary factors, there also are other lifestyle factors that have been

sug-gested to influence vasopressin concentration, like smoking14_{, alcohol consumption}15-17_,

and obesity18_{. However, the studies that have been performed did not investigate the}

overall effect of various lifestyle factors on vasopressin concentration in an integrated manner to see which ones are effectively influencing vasopressin concentration. Moreo-ver, these results were obtained in non-ADPKD populations. Given these considerations we investigated in a cross-sectional study which modifiable lifestyle factors are associ-ated with copeptin concentration in ADPKD patients.

69

MATERIALS AND METHODS

Included in this study were data obtained from two cohort studies. The CRISP Study is a multicenter observational US based cohort study that was created to analyze the rate of disease progression in patients with ADPKD and to find predictors of rapid disease progression. Inclusion criteria were age 15-46 years and creatinine clearance > 70 mL/min with enrichment of clinical risk factors for progression to ESRD in two thirds of patients (early onset hypertension before the age of 35 or presence of proteinuria). Detailed

de-scriptions of the study protocol have been published elsewhere19_{. The UMCG cohort is a}

single center observational cohort study of patients with ADPKD in the Netherlands. All patients with ADPKD visiting the outpatient clinic were asked to participate. No inclusion criteria were set to age or renal function, but subjects receiving renal replacement were excluded. In both studies the diagnosis of ADPKD was made based on the Ravine

crite-ria20_{, and patients were deemed ineligible if they had undergone kidney surgery or cyst}

drainage procedures, when they were unable to undergo magnetic resonance imaging (MRI), or had other medical conditions besides hypertension that potentially could af-fect kidney function (e.g., diabetes mellitus). Both studies were performed in adherence to the Declaration of Helsinki and all participants gave written informed consent allow-ing to use data for the present analyses.

Measurements and definitions

In both studies patients collected a 24-hour urine sample the day before their visit to the study center. During the visit weight, height and blood pressure were measured (blood pressure using an automatic device) and patients provided information on smoking, cof-fee and alcohol consumption by questionnaire. Fasting blood was drawn, in which plas-ma glucose was measured using standard laboratory techniques. Plasplas-ma samples were stored at -80°C for later assessment of copeptin concentration. GFR was measured as

clearance of cold iothalamate in the CRISP cohort21 _{and as clearance of 125I-iothalamate}

in the UMCG cohort22_{, as described previously. In both cohorts TKV was measured}

simi-larly using MR images and classical volumetry (Analyze Direct software, AnalyzeDirect, Inc., Overland Park, KS, USA).

Copeptin was measured by sandwich immunoluminometric assay using a murine mon-oclonal antibody directed to amino acids 137-144 of pro-AVP (CT-proAVP LIA BRAHMS;

Thermo Scientific)23_{. The lower limit of detection was 0.4 pmol/L and the intra-assay}

coefficient of variation 15, 8, 4 and 3%, for the copeptin concentrations 3, 4, 15 and 50

pmol/L,respectively24_.

70

4

Patients with a missing copeptin value (n= 20), missing or assumed incorrect 24-hour urine collection (n= 20) were excluded, leaving 303 participants for analysis. Incorrect 24-hour urine collection was determined using the difference between expected and measured 24-hour urine volume. In case this was outside the 95% distribution range, the 24-hour urine collection was assumed to be incorrect. The expected 24-hour urine volume was calculated by comparing creatinine clearance estimated by the Cockcroft-Gault formula and actual creatinine clearance. In the UMCG cohort, three patients had a copeptin concentration over 10 times the interquartile range above the third quartile, even though their plasma osmolarity was within normal limits. These subjects were

con-sidered outliers25 _{and their data was excluded.}

Statistical analyses

Continuous data with normal distribution are reported as means ± standard deviation (SD), while data with a non-normal distribution are reported as median with interquar-tile range (IQR). Baseline characteristics of the study population are provided stratified to sex-specific quartiles of copeptin concentration, because copeptin concentration is

known to be higher in men than in women23,26_{. Differences between the four strata were}

tested by ANOVA (F-test) for continuous data, a Kruskal-Wallis test for non-normally dis-tributed variables and a Chi-square test for categorical variables. To test for significant trends across the quartiles, we used polynomial ANOVA trend analysis, or the Jonck-heere-Terpstra test in case of non-parametric variables and the Linear-by-Linear Associa-tions in case of categorical variables.

For linear regression analyses, non-normally distributed variables were logarithmically transformed to meet the assumptions for linear regression analyses. To corroborate ear-lier findings that copeptin is linked to chronic kidney disease, we first performed linear regression analyses testing univariable associations of copeptin concentration as inde-pendent variable with albuminuria, TKV and GFR as deinde-pendent variables. These analyses were repeated adjusting for cohort, age, sex, race and height. Second, we performed linear regression analyses, with copeptin concentration as dependent variable. In these analyses, we adjusted for non-modifiable factors (cohort, age, sex, race and height) as well as for GFR and TKV as markers of disease severity in ADPKD. The modifiable factors we investigated were weight, smoking status, use of alcohol and coffee, SBP, use of diu-retics and other antihypertensives, plasma glucose, 24-hour urinary volume (as surrogate for fluid intake), and excretion of sodium and urea (as surrogates for sodium and pro-tein intake, respectively). Lastly, we performed a stepwise backward multivariable linear regression analysis to investigate the association between copeptin concentration and the aforementioned modifiable lifestyle factors in an integrated manner. A p-value >0.05

71

was used to exclude factors from the model. Before factors were definitively excluded, we tested whether non-linear associations with copeptin concentration existed by add-ing their squared value to the model. The variables that were maintained in the final model were also tested for non-linear associations as well as for interactions. We re-peated the aforementioned analyses for both cohorts separately. All analyses were per-formed using SPSS Statistics, version 22.0. A p-value <0.05 was considered to indicate statistical significance.

RESULTS

In this study 303 ADPKD patients were included, of which 205 originated from the CRISP cohort and 98 from the UMC Groningen cohort. Of these patients, 43.2% was male, with a mean age of 35.0±10.1 years, mGFR of 99.7±30.9 ml/min, and median TKV of 950.2 (650.9-1602.0) mL. The median copeptin concentration was 3.54 (2.17-7.29) pmol/L with higher values in males than in females (5.47 (2.71-12.40) and 2.77 (1.85-5.02) pmol/L, respective-ly, p<0.001). Table 1 shows the characteristics of patients by sex-adjusted quartiles of copeptin concentration. Subjects with higher copeptin concentration were older, used more coffee and alcohol, had a higher systolic blood pressure, and lower urine volumes (P for trend 0.001, 0.02, 0.03, 0.02 and 0.02, respectively). Importantly, subjects with higher copeptin concentration also had higher albuminuria and TKV, and lower mGFR. After adjustment for age, sex and cohort, copeptin remained associated with albuminu-ria, TKV and mGFR (all p<0.001).

Linear regression models showed significant associations for all non-modifiable factors with copeptin concentrations, except for race and height (Table 2). Also some of the modifiable factors were significantly associated with copeptin concentration (Table 2, left columns). Thereafter, stepwise backward linear regression analysis was performed. The final model had an adjusted R2 of 0.48. Modifiable lifestyle factors that remained in the model, and explained a higher copeptin value, were a lower urine volume (as surro-gate for fluid intake, p=0.001), higher sodium excretion (as surrosurro-gate for sodium intake, p=0.004) and higher systolic blood pressure (p=0.01). The non-modifiable factors that were significantly associated with an increased copeptin concentration in this model were decreased mGFR (p<0.001), male sex (p=0.001) and younger age (p=0.02). No as-sociations with copeptin concentration were found for the lifestyle factors smoking, use of alcohol or coffee, or antihypertensive medication. Urea excretion – as surrogate for protein intake – was also not associated with copeptin concentration. In the stepwise backward model no significant interaction was found between any of these factors and

72

4

Sex-specific quartiles of plasma copeptin concentration (pmol/L) M ≤ 2.71 F ≤ 1.85 n: 7 M 2.71-5.47 F 1.85-2.77 n: 74 M 5.47-12.40 F 2.77-5.02 n: 76 M ≥ 12.40 F ≥ 5.02 n: 75 p-value p for trend Non-modifiable factors Age (years) 32.5 ± 8.5 35.4 ± 8.5 33.4 ± 11.0 39.0 ± 11.1 <0.001 0.001 Sex (n (% male)) 33 (42.3) 33 (44.6) 33 (43.4) 32 (42.7) 1.0 1.0 Caucasian (n (%)) 71 (91.0) 67 (90.5) 67 (88.2) 66 (88.0) 0.9 0.5 Height (cm) 173.0 ± 10.2 175.3 ± 11.5 173.5 ± 11.8 175.9 ± 11.3 0.3 0.2 mGFR (mL/min) 112.3 ± 23.5 103.7 ± 29.0 106.0 ± 26.2 76.2 ± 31.8 <0.001 <0.001 TKV (mL) 804 [544-1143] 905 [705-1442] 965 [598-1602] 1477 [851-2160] <0.001 <0.001 Albuminuria (mg/24h) 16.8 [10.0-29.5] 26.5 [13.5-51.5] 37.8 [13.4-59.0] 54.9 [25.2-166.2] <0.001 <0.001

Modifiable factors Weight (kg)

77.3 ± 18.6 75.8 ± 15.9 80.4 ± 17.3 80.6 ± 20.1 0.3 0.1 Smoking (n (%)) 13 (16.9) 16 (21.6) 13 (17.8) 19 (26.8) 0.4 0.2 Alcohol use (n (%)) 32 (45.7) 45 (66.2) 39 (61.9) 47 (68.1) 0.03 0.02 Coffee use (n (%)) 37 (52.1) 44 (62.9) 44 (66.7) 48 (69.6) 0.2 0.03 SBP (mmHg) 122.0 ± 19.4 127.8 ± 11.9 124.3 ± 11.8 129.3 ± 14.1 0.009 0.02 Diuretics (n (%)) 11 (14.1) 7 (9.5) 9 (11.8) 16 (21.3) 0.2 0.2 Other AHT (n (%)) 31 (39.7) 39 (52.7) 33 (43.4) 44 (58.7) 0.08 0.06

Serum glucose (mmol/L)

4.9 ± 0.6 4.6 ± 2.9 5.2 ± 0.8 4.3 ± 3.7 0.2 0.3 Urine volume (L/24h) 2.6 ± 1.1 2.6 ± 1.2 2.3 ± 0.9 2.3 ± 0.9 0.05 0.02 Table 1.

Characteristics of the study population according to sex-specific quartiles of plasma copeptin concentration

73

Sex-specific quartiles of plasma copeptin concentration (pmol/L) M ≤ 2.71 F ≤ 1.85 n: 7 M 2.71-5.47 F 1.85-2.77 n: 74 M 5.47-12.40 F 2.77-5.02 n: 76 M ≥ 12.40 F ≥ 5.02 n: 75 p-value p for trend Modifiable factors

Urea excretion (mmol/24h)

346.9 ± 123.3 339.2 ± 146.4 341.0 ± 102.1 358.0 ± 126.9 0.8 0.6

Sodium excretion (mmol/24h)

184.8 ± 87.9 186.0 ± 75.9 197.0 ± 80.2 176.5 ± 81.3 0.5 0.7 Table 1. Continued Variables are expressed as mean ± SD or median with their interquartile range. The p-value for diff erences between strata is obtained using an ANOV A, a Kruskal-W allis test in case of non-normal distribution, or a Chi-square test in case of a categorical variable. The p-value for trend for continuous variables is obtained using linear regression analysis, Jonckheere-Terpstra in case of a non-normal distribution, or Linear-by-Linear associations in case of categorical variables. Ab

-breviations are: M: male, F: f

emale, mGFR: measured glomerular filtration rate, TKV

: total kidney volume, SBP: systolic blood pressure, AHT

: antihypertensives.

74

4

mGFR in their relation with copeptin concentration.

Figure 1 shows the adjusted associations of all variables that remained significant in the final stepwise backward model, with variables stratified according to quartiles or to clini-cal class. The modifiable factor most strongly associated with copeptin concentration was 24-hour urine volume, as surrogate for fluid intake (St. β -0.167, P=0.001), followed by sodium excretion, as surrogate for salt intake (St. β 0.151, P=0.004) and SBP (St. β 0.109, P=0.01). For urinary sodium excretion, there appeared to be a J-shaped associa-tion with copeptin concentraassocia-tion, with especially high copeptin values in the quartile of subjects with the highest sodium excretion.

Additionally, we performed the aforementioned analyses in both cohorts separately. Co-hort was significantly associated with copeptin concentration in the stepwise backward model (p<0.001), with the UMCG cohort having higher copeptin values. When interac-tions with cohort were tested, we observed that this could be explained by the fact that in the UMCG cohort a higher percentage of patients was male (55.1% vs. 37.6% in CRISP), and that UMCG cohort patients had worse kidney functions, which reflects the differ-ence in GFR inclusion criteria between both studies. Both male sex and lower kidney function are known to be associated with higher copeptin concentrations. The results obtained for the CRISP cohort (adjusted R2=0.17) were essentially similar to the overall analysis, with no differences in the final stepwise backward model (Supplementary Ta-bles 1 and 3, and Supplementary Figure 1). Separate analysis of the UMCG cohort (adjust-ed R2=0.56) reveal(adjust-ed an additional significant association with urea excretion, whereas sodium excretion did not reach formal statistical significance (Supplementary Tables 2 and 4 and Supplementary Figure 2).

DISCUSSION

In this study, we found several modifiable lifestyle factors that were independently associated with the surrogate measure of plasma vasopressin, copeptin. These factors were lower 24-hour urine volume (a surrogate of fluid intake), higher 24-hour sodium excretion (a surrogate of sodium intake) and higher systolic blood pressure (SBP). No associations were found with smoking, alcohol use, coffee use and 24-hour urinary urea excretion (a surrogate of protein intake). In addition, we confirm that copeptin is as-sociated with albuminuria, total kidney volume and measured GFR in ADPKD, also after adjustment for age and sex (p for all <0.001), in support of the notion that vasopressin is associated with disease progression in patients with ADPKD. Vasopressin is known to

75

Table 2.

Multivariable linear regression analyses with copeptin concentration as dependent variable

Adjusted for non-modifiable factors

Stepwise backward model

eβ * e95% CI * St. β p-value eβ * e95% CI * St. β p-value Non-modifiable factors Age (years) 0.987 0.977-0.996 - 0.144 0.007 0.988 0.979-0.998 - 0.124 0.02 Sex (F/M) 0.635 0.489-0.825 - 0.237 0.001 0.642 0.500-0.824 - 0.232 0.001 Race (Caucasian/non-Caucasian) 0.778 0.593-1.020 - 0.081 0.07 0.791 0.611-1.024 - 0.076 0.08 Height (cm) 1.012 1.000-1.025 0.143 0.06 1.009 0.997-1.021 0.105 0.2 mGFR (mL/min) 0.988 0.984-0.991 - 0.402 <0.001 0.965 0.952-0.976 - 1.171 <0.001 mGFR 2 1.000 1.000-1.000 0.727 <0.001 TKV (mL) 1.000 1.000-1.000 0.125 0.01 1.000 1.000-1.000 0.074 0.1 Modifiable factors W eight (kg) 1.007 1.001-1.013 0.138 0.01 Smoking (Y/N) 1.027 0.834-1.266 0.012 0.8

Alcohol use (Y/N)

1.034

0.854-1.251

0.016

0.7

Coffee use (Y/N)

1.039 0.858-1.256 0.018 0.7 SBP (mmHg) 1.007 1.001-1.013 0.110 0.02 1.007 1.001-1.012 0.109 0.01 Diuretics (Y/N) 1.004 0.787-1.280 0.002 0.9

Other antihypertensives (Y/N)

1.105 0.932-1.311 0.053 0.3 1.066 1.034 1.099 <0.001

Serum glucose (mmol/L)

1.015 0.980-1.051 0.038 0.4 Urine volume (L/24h) 0.895 0.825-0.970 - 0.120 0.007 0.857 0.787-0.934 - 0.167 0.001

Urea excretion (mmol/24h)

1.000

0.999-1.000

- 0.047

0.4

Sodium excretion (mmol/24h)

1.001 1.000-1.002 0.069 0.2 1.002 1.001-1.003 0.151 0.004 76

4

Table 2. Continued Note: For continuous variables, the eff ect per 1–standard deviation increment is used. Non-modifiable factors are adjusted for the other non-modifiable factors. * Values we re back ln-transf ormed for ease of comparison. Abbreviations are : M: male, F: female, mGFR: measured glomerular filtration rate, TKV : total kidney

volume, SBP: systolic blood pressure. Figure

1. Modifiable subject cha racterist ics independently associated with copeptin concentration. Variables are shown adjusted for all other variables that remained in the final stepwise backward regression model, thus allowing a comparison of the strength of the various associations. Variables are shown in order of decreasing strength of their association with copeptin concentration. Cont inuous variables are subdivided in quar tiles (numbered 1 to 4 represen ting the lowest to highest quartile, respectively), whereas cate gorical variables are subdivided in clinical classes. Abbreviations: mGFR , measured glomerular filtration rate; Na+ excr ., 24-hour sodium excre -tion; SBP

, systolic blood pressure; M, male (n=111); F

, f

emale (n=162).

77

be very sensitive to changes in plasma osmolality, reflected by the strongest associa-tion we found with 24-hour urine volume. Sodium excreassocia-tion was positively associated with copeptin concentration, which is in line with observational findings in healthy

sub-jects27_{. Of note, this is one of the first studies to show that vasopressin concentration is}

directly associated with ad libitum sodium intake, and the first to show this association in ADPKD patients. In line, a recent small scale phase 2 study in which a low osmolar diet (low in salt as well as in protein) and an increase in water intake was prescribed to 34 ADPKD patients indeed showed that this dietary intervention lowered copeptin

concentration28_{. Which of the dietary changes in this study actually caused the decrease}

in copeptin concentration could not be investigated, because the dietary interventions were prescribed in conjunction.

Systolic blood pressure was the third modifiable factor that was found to be associat-ed with copeptin concentration. Vasopressin is known to increase in response to lower

blood volume and blood pressure13,29_{. In contrast, we found a positive association. It is}

therefore more likely that higher vasopressin was a cause of increased systolic blood pressure, or alternatively, that an independent factor caused both to increase.

Lifestyle factors that did not show significant associations with copeptin concentra-tion were urea excreconcentra-tion (a surrogate for protein intake), smoking, coffee use and al-cohol use. Not finding an association between urea excretion and copeptin concen-tration was unexpected. In steady state, urea excretion can be used as surrogate for protein intake, as it is one of the main components of protein. Bankir et al. describe that protein intake is known to directly increase plasma vasopressin

concentra-tion30_{. It is possible that we did not find an association because urea is an ineffective}

osmol in plasma, as it freely crosses the cell membrane31_{. Another explanation might}

be that sodium has a larger influence on vasopressin secretion than urea32_{, and}

add-ing both sodium as well as urea excretion to our multivariable model might blunt the association with urea due to co-linearity. Indeed, sodium excretion was strongly as-sociated to urea excretion in the overall cohort(r=0.49, p<0.001), as well as the CRISP (r=0.55, p<0.001) and the UMCG (r=0.56, p<0.001) cohorts when analyzed separately. We previously reported in a general population cohort that smoking and alcohol use

were associated with copeptin concentration27_{. Nicotine is known to acutely increase}

vasopressin concentration in healthy subjects33_{, and in ADPKD patients a history of}

smok-ing and pack-years was found to be significantly associated with a more rapid disease

progression34_{. Not finding an association in the multivariable analysis of our data might}

be explained by the low number of smokers. We also did not observe a significant associ-ation between coffee consumption and copeptin concentrassoci-ation. However, this does not deny a possible role of coffee intake in disease progression, because a disease specific

78

4

pathway has been suggested in ADPKD that is vasopressin independent35_.

Our study has limitations inherent to the cross-sectional study design. Most importantly, no firm conclusions can be drawn on the cause-effect relationship of the associations we found. The same caution should be kept in mind when suggesting that the aforemen-tioned lifestyle interventions could improve prognosis in ADPKD patients. However, our finding that copeptin concentration was associated with albuminuria, TKV and eGFR is compatible with a causal role of vasopressin in the progression of this disease. Strengths of this study are that we were able to combine data of two well-phenotyped, with among others gold standard measurement of TKV (by MRI volumetry) and GFR (by iothalamate infusion). Moreover, essentially similar results were obtained when both cohorts were analyzed separately, which makes our findings robust.

In conclusion we found that three modifiable lifestyle factors were associated with an increased copeptin concentration: higher systolic blood pressure, lower 24-hour urine volume and higher urinary sodium excretion (as surrogate for fluid and sodium intake, respectively). No significant associations were found for other lifestyle factors, such as urinary urea excretion (as surrogate for protein intake), smoking, alcohol use and coffee use. Combined, these findings suggest that a high water and low salt intake could poten-tially decrease copeptin concentration, and strengthen the rationale to start adequately powered randomized controlled trials of sufficient duration to elucidate whether an in-crease in water intake and a dein-crease in sodium intake is feasible in the long-term and will decrease copeptin, as surrogate for vasopressin concentration. Furthermore, the question remains whether this lowered vasopressin concentration will beneficially affect the rate of disease progression in ADPKD.

ACKNOWLEDGEMENTS

The CRISP Study is supported by cooperative agreements from the National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases (DK056943, DK056956, DK056957, and DK056961) and by the National Center for Research Re-sources (NCRR) GCRCs at each institution (RR000039 Emory, RR00585 Mayo, RR23940 Kansas, and RR000052 UAB) and the NCRR CTSAs at each institution (RR025008 Emory, RR024150 Mayo, RR033179 Kansas, RR025777 UAB, and RR024153 Pittsburgh).

We would like to thank dr. J. Struck from Adrenomed AG, Hennigsdorf, Germany, who donated the assays for the copeptin measurements.

79

REFERENCES

1. Wang X, Wu Y, Ward CJ, et al. Vasopressin directly regulates cyst growth in polycystic kidney disease. J Am Soc Nephrol. 2008;19(1):102-108.

2. Gattone VH,2nd, Wang X, Harris PC, et al. Inhibition of renal cystic disease development and progression by a vasopressin V2 receptor antagonist. Nat Med. 2003;9(10):1323-1326.

3. Torres VE, Wang X, Qian Q, et al. Effective treatment of an orthologous model of autosomal dominant polycystic kidney disease. Nat Med. 2004;10(4):363-364.

4. Meijer E, Gansevoort RT, de Jong PE, et al. Therapeutic potential of vasopressin V2 receptor antagonist in a mouse model for autosomal dominant polycystic kidney disease: Optimal timing and dosing of the drug. Nephrol Dial Transplant. 2011;26(8):2445-2453.

5. Meijer E, Boertien W, Zietse R, et al. Potential deleterious effects of vasopressin in chronic kidney disease and par-ticularly autosomal dominant polycystic kidney disease. Kidney Blood Press Res. 2011;34(4):235-244.

6. Meijer E, Bakker SJ, van der Jagt EJ, et al. Copeptin, a surrogate marker of vasopressin, is associated with disease severity in autosomal dominant polycystic kidney disease. Clin J Am Soc Nephrol. 2011;6(2):361-368.

7. Zittema D, van den Berg E, Meijer E, et al. Kidney function and plasma copeptin levels in healthy kidney donors and autosomal dominant polycystic kidney disease patients. Clin J Am Soc Nephrol. 2014;9(9):1553-1562. 8. Boertien W, Riphagen I, Drion I, et al. Copeptin, a surrogate marker for arginine vasopressin, is associated with

de-clining glomerular filtration in patients with diabetes mellitus (ZODIAC-33). Diabetologia. 2013;56(8):1680-1688. 9. Lacquaniti A, Chirico V, Lupica R, et al. Apelin and copeptin: Two opposite biomarkers associated with kidney

func-tion decline and cyst growth in autosomal dominant polycystic kidney disease. Peptides. 2013;49:1-8.

10. Boertien W, Meijer E, Zittema D, et al. Copeptin, a surrogate marker for vasopressin, is associated with kid-ney function decline in subjects with autosomal dominant polycystic kidkid-ney disease. Nephrol Dial Transplant. 2012;27(11):4131-4137.

11. Torres VE, Chapman AB, Devuyst O, et al. Tolvaptan in patients with autosomal dominant polycystic kidney disease. N Engl J Med. 2012;367(25):2407-2418.

12. Torres VE, Chapman AB, Devuyst O, et al. Tolvaptan in later-stage autosomal dominant polycystic kidney disease. N Engl J Med. 2017;377(20):1930-1942.

13. Bankir L. Antidiuretic action of vasopressin: Quantitative aspects and interaction between V1a and V2 receptor-mediated effects. Cardiovasc Res. 2001;51(3):372-390.

14. Aubert JF, Burnier M, Waeber B, et al. Nicotine-induced release of vasopressin in the conscious rat: Role of opioid peptides and hemodynamic effects. J Pharmacol Exp Ther. 1987;243(2):681-685.

15. Linkola J, Ylikahri R, Fyhrquist F, et al. Plasma vasopressin in ethanol intoxication and hangover. Acta Physiol Scand. 1978(104):180-187.

16. Madeira MD, Paula-Barbosa MM. Effects of alcohol on the synthesis and expression of hypothalamic peptides. Brain Res Bull. 1999;48(1):3-22.

17. Silva SM, Madeira MD, Ruela C, et al. Prolonged alcohol intake leads to irreversible loss of vasopressin and oxy-tocin neurons in the paraventricular nucleus of the hypothalamus. Brain Res. 2002;925(1):76-88.

18. Enhorning S, Bankir L, Bouby N, et al. Copeptin, a marker of vasopressin, in abdominal obesity, diabetes an microalbuminuria: The prospective malmo diet and cancer study cardiovascular cohort. Int J Obes (Lond). 80

18. 2013;37(4):598-603.

19. Chapman A, Guay-Woodford L, Grantham J, et al. Renal structure in early autosomal-dominant polycystic kidney disease (ADPKD): The consortium for radiologic imaging studies of polycystic kidney disease (CRISP) cohort. Kid-ney Int. 2003;64(3):1035-1045.

20. Ravine D, Gibson R, Walker R, et al. Evaluation of ultrasonographic diagnostic criteria for autosomal dominant polycystic kidney disease 1. Lancet. 1994;343(8901):824-827.

21. Rule A, Torres V, Chapman A, et al. Comparison of methods for determining renal function decline in early autoso-mal dominant polycystic kidney disease: The consortium of RadiologicImaging studies of polycystic kidney disease cohort. J Am Soc Nephrol. 2006;17(3):854-862.

22. Apperloo AJ, de Zeeuw D, Donker AJ, et al. Precision of glomerular filtration rate determinations for long-term slope calculations is improved by simultaneous infusion of 125I-iothalamate and 131I-hippuran. J Am Soc Nephrol. 1996;7(4):567-572.

23. Morgenthaler NG, Struck J, Alonso C, et al. Assay for the measurement of copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem. 2006(52):112-119.

24. B·R·A·H·M·S GmbH. Instruction for use - B·R·A·H·M·S CT-proAVP KRYPTOR. 2012;R03en.

25. Moore DS, McCabe GP. Introduction to the practice of statistics. 4th ed. New York: W.H. Freeman; 2002. 26. Bhandari S, Loke I, Davies J, et al. Gender and renal function influence plasma levels of copeptin in healthy

indi-viduals. Clin Sci (Lond). 2009(116):257-263.

27. van Gastel MD, Meijer E, Scheven LE, et al. Modifiable factors associated with copeptin concentration: A general population cohort. Am J Kidney Dis. 2015;65(5):719-727.

28. Amro OW, Paulus JK, Noubary F, et al. Low-osmolar diet and adjusted water intake for vasopressin reduction in autosomal dominant polycystic kidney disease: A pilot randomized controlled trial. Am J Kidney Dis. 2016. 29. Bichet DG. Central vasopressin: dendritic and axonal secretion and renal actions. Clin Kidney J.

2014;7(3):242-247.

30. Bankir L, Roussel R, Bouby N. Protein- and diabetes-induced glomerular hyperfiltration: Role of glucagon, vaso-pressin, and urea. Am J Physiol Renal Physiol. 2015;309(1):F2-23.

31. Rasouli M. Basic concepts and practical equations on osmolality: Biochemical approach. Clin Biochem. 2016;49(12):936-941.

32. Bankir L, Bichet DG, Morgenthaler NG. Vasopressin: Physiology, assessment and osmosensation. J Intern Med. 2017;282(4):284-297.

33. Chiodera P, Volpi R, Capretti L, et al. Gamma-aminobutyric acid mediation of the inhibitory effect of endoge-nous opioids on the arginine vasopressin and oxytocin responses to nicotine from cigarette smoking. Metabolism. 1993(42):762-765.

34. Ozkok A, Akpinar TS, Tufan F, et al. Clinical characteristics and predictors of progression of chronic kidney disease in autosomal dominant polycystic kidney disease: A single center experience. Clin Exp Nephrol. 2013;17(3):345-351.

35. 35. Belibi FA, Wallace DP, Yamaguchi T, et al. The effect of caffeine on renal epithelial cells from patients with autosomal dominant polycystic kidney disease. J Am Soc Nephrol. 2002;13(11):2723-2729.

81

82

4

SUPPLEMENT

AR

Y MA

TERIAL

Sex-specific quartiles of plasma copeptin concentration (pmol/L)

M ≤ 2.27 F ≤ 1.60 n: 51 M >2.27-3.45 F >1.60-2.42 n: 52 M >3.45-6.80 F >2.42-4.50 n: 51 M > 6.80 F > 4.50 n: 51 p-value p for trend Non-modifiable factors Age (years) 33.2 ± 8.7 31.3 ± 8.1 32.0 ± 9.1 33.3 ± 9.0 0.6 0.9 Sex (n (% male)) 19 (37.3) 20 (38.5) 19 (37.3) 19 (37.3) 0.9 0.9 Caucasian (Y/N) 46 (90.2) 47 (90.4) 40 (78.4) 44 (86.3) 0.3 0.3 Height (cm) 171.2 ± 10.1 172.6 ± 10.0 171.1 ± 11.2 172.9 ± 11.7 0.8 0.6 mGFR (mL/min) 110.8 ± 24.3 110.8 ± 29.3 102.6 ± 24.0 93.9 ± 25.3 0.003 <0.001 TKV (mL) 817 [571-1261] 774 [553-1212] 850 [587-1342] 1093 [650-1710] 0.2 0.03 Albuminuria (mg/24h) 19.0 [8.4-29.1] 22.5 [12.0-51.5] 29.1 [14.5-51.6] 43.8 [24.5-66.6] 0.001 <0.001 Modifiable factors W eight (kg) 74.6 ± 16.6 77.7 ± 19.3 73.6 ± 14.7 78.1 ± 19.8 0.5 0.6 Smoking (n (%)) 7 (13.7) 11 (21.2) 8 (15.7) 8 (15.7) 0.8 0.9 Alcohol use (n (%)) 18 (40.0) 21 (46.7) 25 (59.5) 23 (47.9) 0.3 0.3 Coffee use (n (%)) 25 (55.6) 23 (47.9) 23 (51.1) 25 (52.1) 0.9 0.8 SBP (mmHg) 120.5 ± 21.8 126.0 ± 13.1 123.8 ± 9.8 126.8 ± 15.8 0.2 0.09 Diuretics (n (%)) 7 (13.7) 5 (9.6) 3 (5.9) 7 (13.7) 0.5 0.8 Other AHT (n (%)) 20 (39.2) 24 (46.2) 25 (49.0) 24 (47.1) 0.8 0.4

83

4

Supplementary T

able 1.

Continued

Sex-specific quartiles of plasma copeptin concentration (pmol/L)

M ≤ 2.27 F ≤ 1.60 n: 51 M >2.27-3.45 F >1.60-2.42 n: 52 M >3.45-6.80 F >2.42-4.50 n: 51 M > 6.80 F > 4.50 n: 51 p-value p for trend Modifiable factors

Serum glucose (mmol/L)

4.89 ± 0.53 4.96 ± 0.64 5.06 ± 0.53 4.95 ± 0.59 0.6 0.5 Urine volume (L/24h) 2.57 ± 1.08 2.72 ± 1.26 2.28 ± 1.03 2.26 ± 1.03 0.1 0.05

Urea excretion (mmol/24h)

327.0 ± 125.4 331.2 ± 108.4 297.2 ± 119.3 294.5 ± 92.9 0.2 0.07

Sodium excretion (mmol/24h)

177.3 ± 82.8 203.9 ± 89.9 183.7 ± 72.3 192.6 ± 93.3 0.4 0.6 Variables are expre ssed as mean ± SD or median with their interquartile range. The p-value for diff erences between strata is obtained using an ANOV A, a Kruskal-W allis test in case of non-normal distribution, or a Chi-square test in case of a categorical variable. The p-value for trend for continuous variables is obtained using linear regression ana lysis, Jonckheere-Terpstra in case of a non-normal distribution, or Linear-by-Linear associations in case of categorical variables. Abbreviat ions are: M: male, F: female, mGFR: measured glomerular filtration rate, AHT : antihypertensives, TKV : total kidney volume, SBP: systolic blood pressure.

84

4

Sex stratified quartiles of plasma copeptin concentration (pmol/L)

M ≤ 5.26 F ≤ 2.40 n: 24 M > 5.26-12.40 F > 2.40-3.16 n: 25 M > 12.40-21.20 F > 3.16-8.94 n: 25 M > 21.20 F > 8.94 n: 24 p-value p for trend Non-modifiable factors Age (years) 37.2 ± 7.6 40.1 ± 12.2 43.4 ± 11.2 41.1 ± 11.3 0.2 0.1 Sex (n (% male)) 13 (54.2) 14 (56.0) 14 (56.0) 13 (54.2) 0.9 0.9 Caucasian (Y/N) 23 (95.8) 25 (100.0) 24 (96.0) 22 (91.7) 0.5 0.4 Height (cm) 179.4 ± 12.3 180.6 ± 10.0 178.4 ± 10.7 179.8 ± 9.4 0.9 0.9 mGFR (mL/min) 109.0 ± 23.0 109.1 ± 29.9 78.4 ± 28.4 61.1 ± 38.6 <0.001 <0.001 TKV (mL) 834 [679-1385] 1451 [799-1850] 1378 [981-2385] 1783 [1004-2332] 0.06 0.006 Albuminuria (mg/24h) 23.6 [9.8-44.7] 37.8 [11.8-64.6] 103.8 [23.4-181.0] 70.4 [22.2-183.3] 0.008 0.001 Modifiable factors W eight (kg) 79.8 ± 18.8 84.9 ± 12.9 82.8 ± 17.5 87.8 ± 21.3 0.5 0.2 Smoking (n (%)) 7 (30.4) 4 (18.2) 8 (33.3) 8 (38.1) 0.5 0.4 Alcohol use (n (%)) 22 (95.7) 18 (81.8) 20 (83.3) 16 (76.2) 0.3 0.1 Coffee use (n (%)) 19 (82.6) 21 (95.5) 19 (79.2) 18 (85.7) 0.4 0.8 SBP (mmHg) 129.1 ± 13.2 128.0 ± 11.8 128.2 ± 13.4 130.8 ± 10.9 0.9 0.6 Diuretics (n (%)) 4 (16.7) 4 (16.0) 5 (20.0) 8 (33.3) 0.4 0.2 Other AHT (n (%)) 12 (50.0) 11 (44.0) 16 (64.0) 15 (62.5) 0.4 0.2

85

4

Supplementary T

able 2.

Continued

Sex stratified quartiles of plasma copeptin concentration (pmol/L)

M ≤ 5.26 F ≤ 2.40 n: 24 M > 5.26-12.40 F > 2.40-3.16 n: 25 M > 12.40-21.20 F > 3.16-8.94 n: 25 M > 21.20 F > 8.94 n: 24 p-value p for trend Modifiable factors

Serum glucose (mmol/L)

4.31 ± 4.18 4.68 ± 2.99 4.63 ± 4.24 3.57 ± 5.07 0.8 0.5 Urine volume (L/24h) 2.59 ± 0.74 2.26 ± 1.10 2.37 ± 0.61 2.26 ± 0.81 0.5 0.3

Urea excretion (mmol/24h)

416.3 ± 138.5 426.0 ± 132.2 396.0 ± 109.3 405.0 ± 117.9 0.8 0.6

Sodium excretion (mmol/24h)

185.5 ± 68.1 186.0 ± 75.9 173.0 ± 68.6 172.3 ± 85.4 0.9 0.4 Variables are expressed as mean ± SD or median with their interquartile range. The p-value for diff erences between strata is obtained using an ANOV A, a Kruskal-W allis test in case of non-norm al distribution, or a Chi-square test in case of a categorical variable. The p-value for trend for continuous variables is obtained using linear regression analysis, Jonckheere-Terpstra in case of a non-normal distribution, or Linear-by-Linear associations in case of categorical variables. Abbreviations are: M: male, F: female, mGFR: measured glomerular filtration rate, TKV : total kidney volume, SBP: systolic blood pressure, AHT : antihypertensives.

86

4

Adjusted for non-modifiable factors

Stepwise backward model

eβ * e95% CI * St. β p-value eβ * e95% CI * St. β p-value Non-modifiable factors Age (years) 0.988 0.958 1.014 0.3 0.987 0.956 1.018 0.4 Sex (F/M) 0.735 0.560 0.617 <0.001 0.651 0.617 0.688 <0.001 Race (Caucasian/non-Caucasian) 0.820 0.874 0.926 <0.001 0.619 0.510 0.751 <0.001 Height (cm) 1.010 mGFR (mL/min) 0.992 mGFR 2 1.446 1.191 1.754 <0.001 TKV (mL) 1.000 0.830 0.870 <0.001 0.636 0.582 0.695 <0.001 Modifiable factors W eight (kg) 1.006 0.999-1.013 0.134 0.1 Smoking (Y/N) 0.982 0.757-1.274 - 0.009 0.9

Alcohol use (Y/N)

1.107

0.898-1.365

0.064

0.3

Coffee use (Y/N)

1.066 0.862-1.318 0.041 0.6 SBP (mmHg) 1.007 1.000-1.013 0.139 0.04 1.008 1.002-1.013 0.122 0.007 Diuretics (Y/N) 0.959 0.696-1.322 - 0.017 0.8

Other antihypertensives (Y/N)

1.060

0.867-1.296

0.039

0.6

Serum glucose (mmol/L)

1.111 0.932-1.322 0.080 0.2 Urine volume (L/24h) 0.933 0.855-1.020 - 0.102 0.1 0.839 0.768-0.919 - 0.190 <0.001

Urea excretion (mmol/24h)

0.999

0.998-1.000

- 0.105

0.2

Sodium excretion (mmol/24h)

1.001 1.000-1.002 0.097 0.2 1.002 1.000-1.003 0.136 0.01

87

4

blood pressure. Supplementary

Table 4. Multivariable linear regression analyses in the UMCG cohort with copeptin concentration as dep endent variable

Adjusted for non-modifiable factors

Stepwise backward model

eβ * e95% CI * St. β p-value eβ * e95% CI * St. β p-value Non-modifiable factors Age (years) 0.995 0.985-1.005 1.014 0.3 0.987 0.956 1.018 0.4 Sex (F/M) 0.728 0.554- 0.957 0.617 <0.001 0.651 0.617 0.688 <0.001 Race (Caucasian/non-Caucasian) 0.844 0.634-1.122 0.926 <0.001 0.619 0.510 0.751 <0.001 Height (cm) 1.025 1.012-1.039 mGFR (mL/min) 0.986 0.983-0.990 TKV (mL) 1.000 1.000-1.000 0.870 <0.001 0.636 0.582 0.695 <0.001 Modifiable factors W eight (kg) 1.007 1.001-1.013 0.128 0.03 1.061 1.000-1.125 1.025 <0.05 W eight 2 1.000 0.999-1.000 - 0.978 <0.05 Smoking (Y/N) 1.111 0.892-1.383 0.044 0.3

Alcohol use (Y/N)

1.182

0.972-1.438

0.082

0.09

Coffee use (Y/N)

1.217 1.004-1.402 0.096 0.05 SBP (mmHg) 1.008 1.002-1.014 0.131 0.006 1.016 1.002-1.030 0.191 0.02 Diuretics (Y/N) 1.021 0.790-1.322 0.008 0.9

88

4

Supplementary T

able 4.

Continued

Adjusted for non-modifiable factors

Stepwise backward model

eβ * e95% CI * St. β p-value eβ * e95% CI * St. β p-value Modifiable factors

Other antihypertensives (Y/N)

1.141

0.952-1.366

0.069

0.2

Serum glucose (mmol/L)

1.004 0.967-1.042 0.011 0.8 Urine volume (L/24h) 0.864 0.795-0.940 - 0.158 0.001 0.661 0.534-0.818 -0.336 <0.001

Urea excretion (mmol/24h)

1.000 0.999-1.001 0.039 0.5 1.002 1.000-1.003 0.208 0.04

Sodium excretion (mmol/24h)

1.000 0.999-1.002 0.030 0.6 For continuous variables, the eff ect per 1–standard deviation increment is used. Non-modifiable factors are adjusted for the other non-modifiable factors. * Values were back ln-transf ormed for ease of comparison. Abbreviations are: M: male, F: female, mGFR: measured glomerular filtration rate, TKV : total kidney volume,

89

4

Supplementary Figure 1. Modifiable subject characteristics independently associated with

copeptin concentration in the CRISP cohort.

Supplementary Figure 2. Modifiable subject characteristics independently associated with copeptin concentration in the UMCG cohort.