The influence of vitamin D analogs on calcification modulators, N-terminal pro-B-type natriuretic peptide and inflammatory markers in hemodialysis patients: a randomized crossover study

Subjects in this study were a subset of participants (57 of 86 enrolled patients), in a multicenter, open-label, 1:1 randomized, crossover study comparing alfacalcidol and paricalcitol (SHPT-study). These subjects and the primary results of the study have been described elsewhere [15, 16]; in brief, adults receiving chronic hemodialysis therapy with well-controlled calcium and phosphate levels, and secondary hyperparathyroidism (p-iPTH >350 pg/ml) received alfacalcidol or paricalcitol for 16 weeks in order to control the secondary hyperparathyroidism.

Only participants from whom blood had been collected during the entire trial period were included. The entire study population did not have blood sample collection because of: 1) improper collection at study sites; 2) participants refusing to participate in Biobank collection; and 3) withdrawal from the original study.

The study was in compliance with the Helsinki Declaration of 1975, revised in 2000, and was approved by the Danish National Committee on Biomedical Research Ethics (SJ-27). All participants had received written and oral information prior to the study and had given written informed consent.

The design has been described previously and discussed in detail [15]. The study took place in 10 public Danish dialysis departments, all being a part of hospital nephrology departments. The randomization was performed after a 6-week washout period. The first arm received alfacalcidol for 16 weeks and then went through another 6-week washout period before they received paricalcitol for 16 weeks. The second arm received paricalcitol for 16 weeks followed by a 6-week washout period before they received alfacalcidol for 16 weeks. The starting dose of alfacalcidol (Etalpha®Injection) was 3 μg per week and the starting dose of paricalcitol (Zemplar®Injection) was 9 μg per week. Every second week the dose was titrated 50% according to p-phosphate, p-calcium and p-iPTH. As long as p-phosphate was <1.8 mmol/l, ionized p-calcium was <1.30 mmol/l and p-iPTH was >150 pg/ml, the dose was increased. When p-iPTH was ≤150 pg/ml, p-phosphate was <1.8 mmol/l and ionized p-calcium was <1.35 mmol/l, the dose was maintained. If at any time p-phosphate was >1.8 mmol/l or ionized p-calcium was >1.35 mmol/l in two repeated measurements, the dose was reduced. The minimal dose of alfacalcidol was 1.5 μg/week and that of paricalcitol was 4.5 μg/week. If further reduction was needed, treatment was paused.

After inclusion, the dose of calcium-containing phosphate-binders could only be reduced or left unchanged. Elevated p-phosphate levels were treated thoroughly with calcium-free phosphate-binders, dietary intervention and re-evaluation of the dialysis dose. Elevated p-calcium would lead to dietary intervention and reduction of calcium containing phosphate binders. The calcium concentration of the dialysate was fixed at 1.25 mmol/l.

Blood samples were collected at the beginning (weeks 6 and 22) and at the end (weeks 28 and 44) of each treatment period, separated by a 6-week washout period. Blood samples were drawn from blood lines of the dialyzer before the start of dialysis. Samples were frozen and collected at Roskilde County Hospital. At the end of the study, samples were aliquoted and cooled, and sent to Odense University Hospital or Rigshospitalet for analysis.

Plasma OPG was measured using a sandwich enzyme-linked immunoabsorbent assay (ELISA) with commercially available antibodies (R&D Systems, Minneapolis, MN) and modified antibodies as previously described [17, 18]. In summary, mouse anti-human OPG was used as the capture antibody, and a biotinylated goat anti-human OPG, in combination with an Eu-labeled streptavidin, was used for detection. Recombinant human OPG was used for calibration and the analytical range of the assay was 62.5–4,000 pg/ml. Washing steps were performed on an automated wash system (1296-026 Delfia Platewasher, Wallac, Waltham, MA, USA) and optical densities were determined at 450 nm on a VICTOR X5 Multimode Plate Reader (Perkin-Elmer, Waltham, MA, USA). Data were plotted with a four-parameter logistic (4PL) curve fit of OD readings using GraphPad Prism software (version 5; GraphPad Software, La Jolla, CA). Our intra-assay coefficient of variation was 3%, and the inter-assay variation was 8% in duplicate measurements.

A commercially available ELISA kit for human fetuin-A with monoclonal antibodies (R&D Quantikine, catalog no. DFTA00) was used. NS0-expressed recombinant human fetuin-A was used for calibration and plasma was measured in a 4,000-fold dilution to read within the linear range of the standard curve (75–1,000 pg/ml). Washing steps were performed on an automated wash system (1296-026 Delfia Platewasher) and optical densities were determined at 450 nm on a VICTOR X5 Multimode Plate Reader (Perkin-Elmer). Data were plotted with a four-parameter logistic (4PL) curve fit of OD readings using GraphPad Prism software (version 5; GraphPad Software). Our intra-assay coefficient of variation was 5%, and the inter-assay variation was 9% in duplicate measurements.

Measurement of NT-proBNP was done with a Cobas e411 instrument (Roche). Imprecision of the analysis was below 10%.

Hs-Crp was measured on an Architect C8000 (Abbott Laboratories, Chicago, USA) using the latex immunoassay Crp Vario (Sentinel Diagnostics, Milan, Italy). Reaction parameters were applied to the Architect C8000 system as recommended by the supplier. Within-run/between-run coefficients of variations for the Crp Vario assay on the C8000 were <2.3% and <4.3%, respectively.

An electro-chemiluminescence multiplex system was used on a Sector 2400 Imager from Meso Scale Discovery (Gaithersburg, MD, USA) according to the manufacturer’s instructions. IL-6 and TNF-α were measured using a multiplex system. All samples were run as duplicates.

FGF23 was measured using a sandwich enzyme-linked immunosorbent assay (Kainos Laboratories Inc., Tokyo, Japan), which detects only the biologically active intact FGF23. The intra- and inter-assay coefficients of variation were less than 5.0%.

P-iPTH, ionized p-calcium and p-phosphate were measured every second week during treatment periods in order to guide dose adjustments. These were then analyzed at participating department local laboratories.

The distribution of variables and changes in the variables from the start until the end was described. Continuous data were described as mean (standard deviation or SEM) if normally distributed or median (range) if not. Categorical data were described as numbers and percentages. Comparison of changes between groups were performed according to Altman et al. [19]. Parametric tests were used if the distributions were normal (paired and unpaired t-tests) and non-parametric tests (Fischer’s Exact test, Mann-Whitney U and Wilcoxon signed-rank test) were used for ordered or continuous data. NT-proBNP and hs-Crp were logarithmic transformed in order to reach normality. Explanatory factors for changes in OPG, fetuin-A and NT-proBNP were found using multivariate analysis of repeated measurements with backward selection. In order to explore whether the presence of infection influenced the inflammatory markers, patients experiencing at least one adverse event of infection during the study were censored, and the data were analyzed in remaining patients. All tests were two-sided tests (α = 0.05). Analysis was performed using the SAS 9.1 software package (SAS Institute Inc., Cary, NC, USA).

In 57 patients, blood samples from all four visits during the randomized controlled SHPT study were available (Figure 1). There was no difference between patient characteristics of the present study and enrolled patients in the SHPT study [16]. There was no difference in patient characteristics between treatment groups (Table 1).

Changes in. iPTH, ionized calcium and phosphate (FGF23 has been reported previously [16, 20]).

Changes in NT-proBNP, OPG, fetuin-A, IL-6, TNF-α and hs-Crp are shown in Table 2.