Abstract
The formation of 24,25-dihydroxyvitamin D (24,25(OH)2D) from 25-hydroxyvitamin D (25(OH)D) is the primary mechanism for the metabolic clearance of 25(OH)D, and is regulated by tissue-level vitamin D activity. The ratio of 24,25(OH)2D3 to 25(OH)D3 in blood (vitamin D metabolite ratio, VDMR) is postulated to be a marker of 25(OH)D3 clearance, however this has never been tested. We measured baseline 24,25(OH)2D3 and 25(OH)D3 concentrations in 87 participants by liquid chromatography-tandem mass spectrometry. Following an infusion of deuterated 25(OH)D3, blood samples for each participant were collected over 56 days and analyzed for deuterated vitamin D metabolites. 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio (ratio of the AUC of deuterated 24,25(OH)2D3 to that of deuterated 25(OH)D3) were calculated. We compared the VDMR with these two measures using correlation coefficients and linear regression. Participants had a mean age of 64 ± 11years, 41% were female, 30% were self-described Black, 28% had non-dialysis chronic kidney disease (CKD) and 23% had kidney failure treated with hemodialysis. The VDMR was strongly correlated with 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio (r = 0.51 and 0.76, respectively). Adjusting for 25(OH)D3 clearance or the deuterated metabolite-to-parent AUC ratio in addition to clinical covariates, lower VDMR was observed in participants with CKD and kidney failure than in healthy controls; in Black than White participants; and in those with lower serum albumin. Our findings validate the VDMR as a measure of 25(OH)D3 clearance. This relationship was biased by characteristics including race and kidney disease, which warrant consideration in studies assessing the VDMR.
Keywords: Vitamin D, Vitamin D Metabolite Ratio, Mineral Metabolism
1. INTRODUCTION
Vitamin D and its metabolites are vital for human health. Vitamin D insufficiency has been linked with significantly increased risks of cardiovascular disease, bone disease and death.1,2 Serum or plasma concentration of 25-hydroxyvitamin D (25(OH)D) is the marker of vitamin D status most widely used to guide clinical decision-making and treatment.3 However, 25(OH)D is a relatively inactive metabolite, is inconsistently associated with adverse outcomes in all populations, and more accurately reflects long-term vitamin D exposure from sunlight and diet than tissue-level vitamin D activity.1,2,4 As such, novel biomarkers of vitamin D status remain an area of active investigation.
The metabolic clearance of 25(OH)D may reflect tissue-level vitamin D receptor activation. The binding of 1,25-dihydroxyvitamin D (1,25(OH)2D, the biologically active metabolite of vitamin D) to the vitamin D receptor potently induces the expression of mitochondrial CYP24A1, the principal enzyme that catalyzes the metabolic clearance of 25(OH)D into 24,25-dihydroxyvitamin D (24,25(OH)2D).5-7 The time- and labor-intensive pharmacokinetic methods required to precisely quantify 25(OH)D clearance are impractical for routine epidemiological or clinical use. As such, researchers have used the vitamin D metabolite ratio (VDMR), calculated as the ratio of 24,25(OH)2D to 25(OH)D, as an estimate of 25(OH)D3 clearance. They have proposed the VDMR as a superior biomarker of vitamin D status compared to 25(OH)D8-12 and used it to explain population-level differences in vitamin D metabolism,4,13,14 while clinicians have used the VDMR to diagnose idiopathic infantile hypercalcemia and other related genetic disorders.15-17 Whether the VDMR reflects 25(OH)D3 clearance, however, has never been tested.
In a prior pharmacokinetic study of 87 Black or White adults with a wide range of kidney function, we measured 25(OH)D3 clearance using an infusion of deuterated 25(OH)D3 (d-25(OH)D3), and showed it varied by race and kidney disease.18 Here, we used the same cohort to evaluate the VDMR’s accuracy in predicting 25(OH)D3 clearance measured using gold-standard pharmacokinetic methods, and whether this relationship is biased by characteristics including race and kidney disease.
2. METHODS
2.1. Study Design and Population
This was a cross-sectional study of participants from the Clearance of 25-Hydroxyvitamin D in Chronic Kidney Disease (CLEAR) study (ClinicalTrials.gov identifier: NCT02937350), a pharmacokinetic study which examined the metabolic clearance of 25(OH)D3 by race and kidney disease.18 Briefly, the CLEAR study analyzed 87 Black or White adults with a wide range of kidney function, including kidney failure on hemodialysis, recruited from the greater Seattle metropolitan area. Participants were excluded if they used any vitamin D2 supplements, 1,25(OH)2D3 or an analogue, or vitamin D3 supplements >400 IU/day, among other exclusion criteria previously described.18 At the baseline study visit, urine and blood samples were collected for analysis, and then participants were administered a single intravenous dose of d-25(OH)D3. Blood was drawn at 5 minutes, 4 hours, and 1, 4, 7, 14, 21, 28, 42, and 56 days post-administration, and analyzed for endogenous and deuterated metabolites as described below. The CLEAR study was approved by the University of Washington Institutional Review Board.
2.2. Vitamin D Measurements
Plasma concentrations of endogenous (non-deuterated) 24,25(OH)2D3, 25(OH)D2, and 25(OH)D3 were determined using immunoaffinity extraction and liquid chromatography-tandem mass spectrometry (LC-MS/MS).19,20 Total 25(OH)D was calculated as the sum of 25(OH)D2 and 25(OH)D3. As there was no spectrometric evidence of 24,25(OH)2D2, the VDMR was calculated by dividing the baseline concentrations of 24,25(OH)2D3 (ng/mL) by 25(OH)D3 (ng/mL), and then multiplying by 1000 such that its units are in pg/ng.4 For each participant at each time point, serum concentrations of d-25(OH)D3 and deuterated 24,25(OH)2D3 (d-24,25(OH)2D3) were measured using liquid-liquid extraction and LC-MS/MS methods that used 4-phenyl-1,2,4-triazline-3,5-dione (PTAD) as previously described,13,19,21 except that tri-deuterated 25(OH)D3 (Medical Isotopes Inc, Pelham, NH) was used as the internal standard. The between-batch analytical variability for each analyte is reported in Supplemental Table 1. All vitamin D measurements were made by the University of Washington Nutrition Obesity Research Center.
2.3. Pharmacokinetic Modeling
For each participant, non-compartmental analyses of d-25(OH)D3 and d-24,25(OH)2D3 concentrations vs. time were performed using Phoenix WinNonlin (version 8.2, Certara, Princeton, NJ) as previously described.18 Briefly, a terminal rate constant (k) was estimated from the terminal log-linear d-25(OH)D3 concentration (C56, the concentration at day 56) vs. time points for each participant. The linear trapezoidal rule was used to calculate the area under the serum concentration-time curve from time = 0 to 56 days (AUC0-56) and the area under the serum concentration-time curve from time = 0 was extrapolated to infinity (AUC0-∞) by using AUC0-56 + C56/k. Given observed variability in d-24,25(OH)2D3 concentrations, a terminal rate constant for d-24,25(OH)2D3 vs. time could not be estimated, thus only AUC0-56 was determined. 25(OH)D3 clearance, the pharmacokinetic measurement of the volume of blood from which 25(OH)D3 is completely removed per unit time, was calculated by dividing the administered d-25(OH)D3 dose by the AUC0-∞. The molar ratio of the AUC0-56 for d-24,25(OH)2D3 to the AUC0-56 of d-25(OH)D3, referred to hereafter as the deuterated metabolite-to-parent AUC ratio, was calculated to approximate CYP24A1-mediated 25(OH)D3 clearance.
2.4. Covariates
Baseline demographics (including race), smoking status and comorbidities were ascertained by self-report. Three kidney disease groups were defined: healthy controls, as participants with estimated glomerular filtration rate (eGFR) ≥60 mL/min/1.73 m2; chronic kidney disease (CKD), as participants with eGFR <60 mL/min/1.73 m2 not treated with dialysis; and kidney failure, as participants on chronic maintenance hemodialysis. Blood pressure was calculated as the average of three measurements performed 5 minutes apart on an automated sphygmomanometer. Hypertension was defined by self-report, systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or use of antihypertensive medications. Diabetes was defined by self-report, fasting blood glucose >126 mg/dL, non-fasting blood glucose >200 mg/dL, hemoglobin A1c ≥6.5% or use of glucose-lowering medications. Current medication and supplement use were ascertained from pill bottles and computerized medication lists. Weight and height were measured and used to calculate body mass index (BMI) in units of kg/m2. Estimated blood volume (EBV) was calculated using the Nadler equation.22
Laboratory measurements were obtained from serum and urine samples that were stored at −80 °C, with the exception of intact parathyroid hormone (PTH) and urine albumin to creatinine (UACR), which were assayed immediately after the baseline study visit. General chemistries, urine albumin and creatinine were measured on a Beckman-Coulter DxC autoanalyzer (Beckman-Coulter Inc, Brea, CA). PTH was measured with the Beckman-Coulter DxI automated immunoassay, intact FGF-23 using the Kainos immunoassay (Kainos Laboratories, Tokyo, Japan), and vitamin D binding protein (VDBP) by LC-MS/MS.23-25 UACR was quantified from a single-voided urine sample. eGFR was calculated using the creatinine-only CKD-EPI equation.26
2.5. Statistical Analysis
We used scatterplots and Pearson correlation coefficients to examine the relationships of the VDMR with 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio across kidney disease group and race. Best-fit lines were generated using linear regression. We also used linear regression with Huber-White robust SEM to compare the VDMR, 25(OH)D3 clearance, and deuterated metabolite-to-parent AUC ratio across kidney disease group and race. With VDMR as the sole predictor, we used linear regression to calculate the root mean square error (RMSE) and the percentage of predicted values within 30% of measured 25(OH)D3 clearance or the deuterated metabolite-to-parent AUC ratio (P30) as metrics of predictive precision. We obtained 95% confidence intervals using nonparametric bootstrap methods with 3000 replicates.27 Finally, we examined the covariate determinants of bias of the VDMR, defined as the difference in the VDMR comparing those with different values of the covariate but the same 25(OH)D3 clearance or deuterated metabolite-to-parent AUC ratio. To do this, we performed multivariable linear regression of the VDMR on the potential determinant of bias adjusted for 25(OH)D3 clearance or the deuterated metabolite-to-parent AUC ratio, age, sex, race, BMI, kidney disease group, serum albumin and VDBP. We examined these characteristics given known differences in 25(OH)D clearance by race and kidney disease,18 and the roles of adiposity and serum albumin and VDBP concentrations on vitamin D bioavailability.28,29 As with the primary analyses in the parent CLEAR study, all primary analyses in this study excluded two participants whose total 25(OH)D3 clearances deviated markedly from other observations (919 and 801 mL/day).18 These extreme values were felt to be from unknown and/or unusual factors unrelated to kidney function or race that justified their exclusion in studies assessing these characteristics. All analyses were conducted with R version 3.6.1 (R Foundation for Statistical Computing). A two-tailed P <0.05 was taken as evidence of statistical significance in all analyses.
3. RESULTS
3.1. Participant Characteristics
Among the 87 participants included in this study, the mean age was 64 ± 11 years, 41% were female, 30% were self-described Black, 28% had non-dialysis CKD and 23% had kidney failure. Participants with lower VDMR were more likely to have kidney disease, hypertension, and diabetes, and had lower 25(OH)D3 clearance and deuterated metabolite-to-parent AUC ratios (Table 1). All 20 participants with kidney failure were in the lowest tertile of VDMR. Racial distribution was similar across VDMR tertiles, with 28-35% Black participants in each tertile. Among the kidney disease groups, healthy controls had the highest VDMR, 25(OH)D3 clearance and deuterated metabolite-to-parent AUC ratio, while participants with kidney failure had the lowest of all of these measures (Table 2). There were no observed differences in the VDMR, 25(OH)D3 clearance, or deuterated metabolite-to-parent AUC ratio by race.
Table 1.
Baseline Characteristics of Participants in the Clearance of 25-hydroxyvitamin D in Chronic Kidney Disease Study by VDMR (24,25(OH)2D3/25(OH)D3) Tertile
| Tertile 1 (n=29) |
Tertile 2 (n=29) |
Tertile 3 (n=29) |
|
|---|---|---|---|
| Baseline VDMR range (pg/ng) | 3.7 – 26.1 | 26.1 – 50.2 | 51.7 – 98.0 |
| Age (yr), mean (SD) | 59 (11) | 66 (10) | 65 (9) |
| Female, n (%) | 8 (28) | 13 (45) | 15 (52) |
| Race, n (%) | |||
| Black | 10 (35) | 8 (28) | 8 (28) |
| White | 19 (66) | 21 (72) | 21 (72) |
| Kidney disease, n (%) | |||
| Healthy controls | 2 (7) | 15 (52) | 26 (90) |
| CKD | 7 (24) | 14 (48) | 3 (10) |
| Kidney failure | 20 (69) | 0 (0) | 0 (0) |
| Hypertension, n (%) | 18 (62) | 19 (66) | 9 (31) |
| Diabetes, n (%) | 12 (41) | 13 (45) | 4 (14) |
| Ever smoker, n (%) | 12 (41) | 12 (41) | 13 (45) |
| RAAS-I use, n (%) | 4 (14) | 17 (59) | 5 (17) |
| Statin use, n (%) | 9 (31) | 16 (55) | 7 (24) |
| Vitamin D3 supplement use (≤400 IU/day), n (%) | 1 (3) | 2 (7) | 5 (17) |
| Systolic BP (mmHg), mean (SD) | 124 (18) | 126 (31) | 126 (16) |
| BMI (kg/m2), mean (SD) | 31.2 (8.3) | 29.8 (7.1) | 27.5 (4.4) |
| EBV (L), mean (SD) | 5.5 (1.2) | 5.2 (0.9) | 4.8 (0.8) |
| eGFR (ml/Min per 1.73 m2), mean (SD) | 21 (28) | 61 (22) | 84 (19) |
| UACR (mg/g), mean (IQR) | 14 (0, 28) | 15 (6, 118) | 0 (0, 9) |
| PTH (pg/mL), median (IQR) | 294 (153, 658) | 63 (52, 98) | 54 (40, 69) |
| FGF-23 (pg/mL), median (IQR) | 695 (96, 6652) | 78 (63, 113) | 57 (53, 70) |
| Calcium (mg/dL), mean (SD) | 9.0 (0.6) | 9.2 (0.3) | 9.2 (0.3) |
| Albumin (g/dL), mean (SD) | 4.0 (0.3) | 3.9 (0.3) | 4.2 (0.2) |
| VDBP (μg/mL), mean (SD) | 237 (37) | 220 (33) | 211 (26) |
| 24,25(OH)2D3 (ng/mL), mean (SD) | 0.31 (0.21) | 0.81 (0.36) | 1.61 (0.79) |
| 25(OH)D3 (ng/mL), mean (SD) | 20.3 (8.6) | 21.1 (8.2) | 24.5 (8.9) |
| Total 25(OH)D (ng/mL), mean (SD) | 23.0 (8.8) | 21.7 (8.1) | 25.4 (9.0) |
| 25(OH)D3 clearance (mL/day), mean (SD) | 290 (170) | 314 (73) | 371 (92) |
| Deuterated metabolite-to-parent AUC ratioa, mean (SD) | 0.04 (0.03) | 0.09 (0.03) | 0.13 (0.04) |
VDMR, vitamin D metabolite ratio; 24,25(OH)2D, 24,25-dihydroxyvitamin D; 25(OH)D, 25-hydroxyvitamin D; CKD, chronic kidney disease; RAAS-I, renin-angiotensin-aldosterone system inhibitor; BP, blood pressure; BMI, body mass index; EBV, estimated blood volume; eGFR, estimated glomerular filtration rate; UACR, urine albumin to creatinine ratio; PTH, parathyroid hormone; IQR, interquartile range; FGF-23, fibroblast growth factor-23; VDBP, vitamin D binding protein.
The molar ratio of the area under the curve (AUC) for deuterated 24,25(OH)2D3 to the AUC of deuterated 25(OH)D3.
Table 2.
Measures of 25-Hydroxyvitamin D3 Clearance and Their Correlation
| Baseline VDMR (pg/ng) | 25(OH)D3 clearance (mL/day) |
Deuterated metabolite- to-parent AUC ratioa |
|
|---|---|---|---|
| Mean (SD) | |||
| All participants | 39 (22) | 325 (123) | 0.09 (0.05) |
| Kidney disease | |||
| Healthy controls | 54 (17) | 360 (108) | 0.12 (0.04) |
| CKD | 34 (13) | 313 (86) | 0.08 (0.03) |
| Kidney failure | 12 (6) | 263 (163) | 0.03 (0.03) |
| p-valueb | <0.01 | 0.02 | <0.01 |
| Race | |||
| Black | 33 (20) | 341 (102) | 0.09 (0.06) |
| White | 41 (22) | 318 (131) | 0.09 (0.04) |
| p-valueb | 0.08 | 0.38 | 0.81 |
| Pearson Correlation with Baseline VDMRc | |||
| All participants | 0.51 | 0.76 | |
| Kidney disease | |||
| Healthy controls | 0.06 | 0.43 | |
| CKD | 0.44 | 0.70 | |
| Kidney failure | 0.37 | 0.22 | |
| Race | |||
| Black | 0.66 | 0.78 | |
| White | 0.54 | 0.83 | |
The molar ratio of the area under the curve (AUC) for deuterated 24,25(OH)2D3 to the AUC of deuterated 25(OH)D3.
Global Wald test comparing clearance measures across kidney disease groups or race.
Analyses excluded two participants with unusually high 25(OH)D3 clearances (919 and 801 mL/day).
VDMR, vitamin D metabolite ratio; 25(OH)D, 25-hydroxyvitamin D; CKD, chronic kidney disease.
3.2. Accuracy, Precision and Bias of the VDMR
Overall, the VDMR showed strong correlation with 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio (r = 0.51 and 0.76, respectively; Table 2 and Figure 1). When stratified by kidney disease group, the correlation between the VDMR and 25(OH)D3 clearance was much weaker among healthy controls than in CKD and kidney failure (r = 0.06 vs. 0.44 and 0.37, respectively), while the correlation between the VDMR and the deuterated metabolite-to-parent AUC ratio was relatively stronger for all kidney disease groups (r = 0.22-0.70; Table 2 and Figure 2). When stratified by race, the VDMR remained strongly correlated with both 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio (r = 0.54-0.83). Among all participants, the P30 for 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio were 79% (95% CI: 70, 88) and 62% (95% CI: 52, 73), respectively. The RMSE between predicted and measured 25(OH)D3 clearance, and predicted and measured deuterated metabolite-to-parent AUC ratio were 80 (95% CI: 67, 96) mL/day and 0.03 (95% CI: 0.02, 0.04), respectively.
Figure 1. VDMR vs 25-Hydroxyvitamin D3 Clearance and The Deuterated Metabolite-to-Parent AUC Ratioa.
aThe molar ratio of the area under the curve (AUC) for deuterated 24,25(OH)2D3 to the AUC of deuterated 25(OH)D3.
Analyses excluded two participants with unusually high 25(OH)D3 clearances (919 and 801 mL/day). VDMR, vitamin D metabolite ratio; 25(OH)D, 25-hydroxyvitamin D.
Figure 2. VDMR vs 25-Hydroxyvitamin D3 Clearance and The Deuterated Metabolite-to-Parent AUC Ratioa by Kidney Disease Group and Race.
aThe molar ratio of the area under the curve (AUC) for deuterated 24,25(OH)2D3 to the AUC of deuterated 25(OH)D3.
Analyses excluded two participants with unusually high 25(OH)D3 clearances (919 and 801 mL/day). VDMR, vitamin D metabolite ratio; CKD, chronic kidney disease; 25(OH)D, 25-hydroxyvitamin.
For participants with the same 25(OH)D3 clearance, age, sex, race, BMI, serum albumin and VDBP concentration, statistically significant lower VDMR was observed in those with CKD and kidney failure than in healthy controls (−14.4 (95% CI: −20.9, −8.0) pg/ng and −29.5 (95% CI: −36.5, −22.4) pg/ng, respectively; Table 3). Significantly lower VDMR was also observed in Black than White participants, and in participants with higher BMI and lower serum albumin when adjusted for 25(OH)D3 clearance and clinical covariates. In a parallel set of multivariable regression models that adjusted for the deuterated metabolite-to-parent AUC ratio, all of the same covariates biased the VDMR and in the same direction, except for BMI, which was not a statistically significant determinant of bias (Table 3).
Table 3.
Covariate Determinants of Bias of the VDMR Compared with 25-Hydroxyvitamin D3 Clearance
| 25(OH)D3 Clearance | Deuterated metabolite-to-parent AUC ratioa |
|||
|---|---|---|---|---|
| Estimate of VDMR bias in pg/ng (95%CI) |
p-value | Estimate of VDMR bias in pg/ng (95%CI) |
p-value | |
| Age (per decade increment) | 0.50 (−2.2, 3.2) | 0.71 | 0.60 (−2.2, 3.4) | 0.67 |
| Male sex | −5.0 (−11.6, 1.6) | 0.13 | 0.2 (−4.8, 5.3) | 0.93 |
| Black race | −8.6 (−13.7, −3.4) | <0.01 | −8.2 (−13.5, −2.9) | <0.01 |
| BMI (per 5 kg/m2 increment) | −1.9 (−3.7, −0.1) | 0.04 | 0.2 (−1.5, 2.0) | 0.80 |
| Kidney disease | ||||
| CKD | −14.4 (−20.9, −8.0) | <0.01 | −12.2 (−18.4, −6.2) | <0.01 |
| Kidney failure | −29.5 (−36.5, −22.4) | <0.01 | −24.4 (−34.5, −14.4) | <0.01 |
| Serum albumin (per 0.5 g/dL increment) | 8.3 (2.2, 14.4) | <0.01 | 7.5 (1.5, 13.5) | 0.02 |
| VDBP (per 10 μg/mL increment) | −0.7 (−1.8, 0.4) | 0.22 | −0.8 (−2.0, 0.3) | 0.16 |
The molar ratio of the area under the curve (AUC) for deuterated 24,25(OH)2D3 to the AUC of deuterated 25(OH)D3.
Estimates were derived from linear regression models that included 25(OH)D3 clearance or deuterated metabolite-to-parent AUC ratio, age, sex, race, BMI, kidney disease group, serum albumin, and VDBP as independent variables. Analyses excluded two participants with unusually high 25(OH)D3 clearances (919 and 801 mL/day).
VDMR, vitamin D metabolite ratio; 25(OH)D, 25-hydroxyvitamin D; BMI, body mass index; CKD, chronic kidney disease; VDBP, vitamin D binding protein.
4. DISCUSSION
This is the first study to evaluate and show that the VDMR correlates strongly with 25(OH)D3 clearance measured using pharmacokinetic methods. The stronger correlation with the deuterated metabolite-to-parent AUC ratio than overall 25(OH)D3 clearance is unsurprising, given that the VDMR is essentially a cross-sectional ratio of plasma 24,25(OH)2D3 to 25(OH)D3 exposure over time. It also suggests there are significant non-CYP24A1-mediated pathways of 25(OH)D3 clearance not captured by the VDMR. Importantly, race, kidney disease, and serum albumin significantly biased the VDMR relative to 25(OH)D3 clearance and the deuterated metabolite-to-parent AUC ratio. Specifically, for the same 25(OH)D3 clearance or deuterated metabolite-to-parent AUC ratio, lower VDMR was observed in participants with CKD and kidney failure than healthy controls; in Black than White participants; and in those with lower serum concentrations of albumin.
CYP24A1 is a ubiquitous enzyme found in all vitamin D target tissues. Its expression is potently induced by 1,25(OH)2D in a mechanism thought to be protective against vitamin D toxicity, such that CYP24A1 knockout mice exhibit 1,25(OH)2D excess resulting in death,30 and CYP24A1 expression is used in basic science as a readout for tissue-level vitamin D activity.31-36 Researchers and clinicians have taken these observations to presume the VDMR reflects CYP24A1-mediated 25(OH)D3 clearance, but this central assumption has not been validated until now. Our findings add to the promise of the VDMR as a surrogate measure of vitamin D status, including in patients with advanced CKD as proposed by a group of leading nephrologists.12 This includes stronger associations with adverse outcomes than 25(OH)D,9 and independence from VDBP concentration10 and potentially race.8
We identified key determinants of bias of the VDMR. For participants with the same value of 25(OH)D3 clearance, systematically lower VDMR was seen in CKD and kidney failure than healthy controls; in Black than White participants; and in those with lower serum albumin, even after adjusting for clinical covariates. These findings indicate that analyses using the VDMR should adjust for these characteristics, and have important implications for the interpretation of study results.
Given the bias of lower VDMR with kidney disease, it is likely that previously reported differences in VDMR and 24,25(OH)2D3 overestimate the true differences in 25(OH)D3 clearance associated with lower eGFR.13,37 Nonetheless, we demonstrated a strong association between eGFR and measured 25(OH)D3 clearance in the parent CLEAR study,18 thus the bias in VDMR does not fundamentally change our conclusion that kidney disease is a state of reduced 25(OH)D3 clearance.
The narrative with respect to race is less conclusive. Prior to the CLEAR study, our group and others postulated lower CYP24A1 activity in Black than White participants on the basis of lower VDMR in Black than White populations.4,11,37 Contrary to our expectation, in the parent CLEAR study we observed higher 25(OH)D3 clearance and deuterated metabolite-to-parent AUC ratio in Black than White participants among those with normal eGFR. This important finding underscores a pitfall in interpreting the VDMR and other metabolite-to-parent concentration ratios as surrogate measures of enzymatic activity, and emphasizes the need for other studies for confirmation.
Although 24-hydroxylation of 25(OH)D by CYP24A1 is believed to be the primary mechanism for 25(OH)D clearance, other metabolic pathways of clearance have been identified. For instance, our group recently showed the liver to be the primary site of 25(OH)D conjugation into 25-hydroxyvitamin D-3-O-sulfate (25(OH)D-sulfate).38 As the most abundant vitamin D metabolite at low 25(OH)D concentrations, 25(OH)D-sulfate may serve as an alternative storage form of 25(OH)D,39 and appears to be protected from elimination by the kidney. Yet, remarkably little is known about 25(OH)D-sulfate formation, and how it may differ by population or disease state. Additionally, while CYP24A1 is predominantly a 24-hydroxylase in humans, it is a multifunctional enzyme capable of metabolizing 25(OH)D into other vitamin D products, including calcioic acid and 25(OH)D3-26,23-lactone.40,41 Differences in the activity of any of these alternative pathways of clearance by race and kidney disease could be potential reasons for our observed bias in the VDMR compared with 25(OH)D3 clearance. Pharmacokinetic studies aimed at defining less studied pathways of clearance are needed to improve our understanding of vitamin D biology and may identify new biomarker targets and therapeutic approaches for managing vitamin D disorders.
Unexpectedly, serum albumin concentration biased the VDMR, even after adjusting for clinical covariates including kidney disease. We speculated that differential binding of 24,25(OH)2D3 and 25(OH)D3 to albumin may potentially be responsible for this bias. However, the VDMR was not biased by VDBP concentration, which binds the majority (>85%) of circulating vitamin D metabolites,29 and undermines this speculation. This finding remains unexplained.
Strengths of this study include the gold-standard pharmacokinetic methods used to quantify 25(OH)D3 clearance, and the larger number of Black participants, and participants with CKD and kidney failure relative to prior 25(OH)D tracer studies.42-45 We also recognize several limitations. First, this was a single center study of 87 participants with some unexpected results (e.g., bias of the VDMR by race and serum albumin) that leave room for uncertainty and should be confirmed in future studies. Next, non-Black minorities were not represented in our study and should be included in future studies. Next, we recognize that self-categorized race is a social construct, and that observed racial differences in vitamin D metabolism may be related to genetic ancestry, which was not assessed in this study. Next, we observed high between-batch variability in our deuterated metabolites. If our outcome measurements were more precise, the performance of the VDMR may have been better than we observed. Next, as the deuterated metabolite-to-parent AUC ratio should ideally be calculated from AUCs extrapolated to infinite time, the use of AUC0-56 may lead to errors in the estimation of the ratio.18 Lastly, while the deuterated metabolite-to-parent AUC ratio is likely a good approximation of CYP24A1-mediated 25(OH)D3 clearance, it is influenced by phenomena we could not measure, such as d-24,25(OH)2D3 clearance.
In conclusion, this study validates the VDMR as a surrogate measure of 25(OH)D3 clearance. Black race, kidney disease, and lower albumin are associated with systematically lower VDMR values, suggesting all analyses using the VDMR should be adjusted for these covariates.
Supplementary Material
HIGHLIGHTS.
The vitamin D metabolite ratio is a strong surrogate measure of 25-dihydroxyvitamin D3 clearance.
The vitamin D metabolite ratio is biased by race, kidney disease, and serum albumin.
ACKNOWLEDGEMENTS
This work was supported by grants R01DK099199, R01DK099199-S1, R01GM63666, 2T32DK007467, 1F32DK128986-01, the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1 TR002319, P30 DK040561 and an unrestricted grant from Northwest Kidney Centers. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Abbreviations:
- 25(OH)D
25-hydroxyvitamin D
- 1,25(OH)2D
1,25-dihydroxyvitamin D
- 24,25(OH)2D
24,25-dihydroxyvitamin D
- VDMR
vitamin D metabolite ratio
- CLEAR
Clearance of 25-Hydroxyvitamin D in Chronic Kidney Disease
- d-25(OH)D3
deuterated 25-hydroxyvitamin D3
- LC-MS/MS
liquid chromatography-tandem mass spectrometry
- d-24,25(OH)2D3
deuterated 24,25-dihydroxyvitamin D3
- PTAD
4-phenyl-1,2,4-triazline-3,5-dione
- k
terminal rate constant
- C56
concentration at 56 days
- AUC
area under the serum concentration-time curve
- eGFR
estimated glomerular filtration rate
- CKD
chronic kidney disease
- BMI
body mass index
- EBV
estimated blood volume
- PTH
parathyroid hormone
- UACR
urine albumin to creatinine
- VDBP
vitamin D binding protein
- RMSE
root mean square error
- P30
percentage of predicted values within 30% of the measured value
- 25(OH)D-sulfate
25-hydroxyvitamin D-3-O-sulfate
Footnotes
Declarations of interest: None
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Holick MF. Vitamin D Deficiency. New England Journal of Medicine. 2007;357(3):266–281. [DOI] [PubMed] [Google Scholar]
- 2.Rosen CJ. Clinical practice. Vitamin D insufficiency. N Engl J Med. 2011;364(3):248–254. [DOI] [PubMed] [Google Scholar]
- 3.Institute of Medicine Committee to Review Dietary Reference Intakes for Vitamin D, Calcium. The National Academies Collection: Reports funded by National Institutes of Health. In: Ross AC, Taylor CL, Yaktine AL, Del Valle HB, eds. Dietary Reference Intakes for Calcium and Vitamin D. Washington (DC): National Academies Press (US). National Academy of Sciences.; 2011. [PubMed] [Google Scholar]
- 4.Hsu S, Hoofnagle AN, Gupta DK, et al. Race, Ancestry, and Vitamin D Metabolism: The Multi-Ethnic Study of Atherosclerosis. The Journal of Clinical Endocrinology & Metabolism. 2020;105(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bosworth C, de Boer IH. Impaired vitamin D metabolism in CKD. Semin Nephrol. 2013;33(2):158–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jones G, Prosser DE, Kaufmann M. 25-Hydroxyvitamin D-24-hydroxylase (CYP24A1): Its important role in the degradation of vitamin D. Archives of Biochemistry and Biophysics. 2012;523(1):9–18. [DOI] [PubMed] [Google Scholar]
- 7.Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF. Human 25-Hydroxyvitamin D3-24-Hydroxylase, a Multicatalytic Enzyme. Biochemistry. 1996;35(25):8465–8472. [DOI] [PubMed] [Google Scholar]
- 8.Berg AH, Powe CE, Evans MK, et al. 24,25-Dihydroxyvitamin D3 and Vitamin D Status of Community-Dwelling Black and White Americans. Clinical Chemistry. 2015;61(6):877–884. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ginsberg C, Katz R, de Boer IH, et al. The 24,25 to 25-hydroxyvitamin D ratio and fracture risk in older adults: The cardiovascular health study. Bone. 2018;107:124–130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ginsberg C, Hoofnagle AN, Katz R, et al. The Vitamin D Metabolite Ratio Is Independent of Vitamin D Binding Protein Concentration. Clinical Chemistry. 2021;67(2):385–393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bansal N, Katz R, Appel L, et al. Vitamin D Metabolic Ratio and Risks of Death and CKD Progression. Kidney Int Rep. 2019;4(11):1598–1607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Melamed ML, Chonchol M, Gutiérrez OM, et al. The Role of Vitamin D in CKD Stages 3 to 4: Report of a Scientific Workshop Sponsored by the National Kidney Foundation. American journal of kidney diseases : the official journal of the National Kidney Foundation. 2018;72(6):834–845. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bosworth CR, Levin G, Robinson-Cohen C, et al. The serum 24,25-dihydroxyvitamin D concentration, a marker of vitamin D catabolism, is reduced in chronic kidney disease. Kidney Int. 2012;82(6):693–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Graeff-Armas LA, Kaufmann M, Lyden E, Jones G. Serum 24,25-dihydroxyvitamin D(3) response to native vitamin D(2) and D(3) Supplementation in patients with chronic kidney disease on hemodialysis. Clin Nutr. 2018;37(3):1041–1045. [DOI] [PubMed] [Google Scholar]
- 15.Kaufmann M, Gallagher JC, Peacock M, et al. Clinical utility of simultaneous quantitation of 25-hydroxyvitamin D and 24,25-dihydroxyvitamin D by LC-MS/MS involving derivatization with DMEQ-TAD. The Journal of clinical endocrinology and metabolism. 2014;99(7):2567–2574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cappellani D, Brancatella A, Kaufmann M, et al. Hereditary Hypercalcemia Caused by a Homozygous Pathogenic Variant in the CYP24A1 Gene: A Case Report and Review of the Literature. Case Rep Endocrinol. 2019;2019:4982621–4982621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kaufmann M, Morse N, Molloy BJ, et al. Improved Screening Test for Idiopathic Infantile Hypercalcemia Confirms Residual Levels of Serum 24,25-(OH)2D3 in Affected Patients. Journal of Bone and Mineral Research. 2017;32(7):1589–1596. [DOI] [PubMed] [Google Scholar]
- 18.Hsu S, Zelnick LR, Lin YS, et al. Differences in 25-Hydroxyvitamin D Clearance by eGFR and Race: A Pharmacokinetic Study. Journal of the American Society of Nephrology. 2021;32(1):188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Laha TJ, Strathmann FG, Wang Z, de Boer IH, Thummel KE, Hoofnagle AN. Characterizing Antibody Cross-reactivity for Immunoaffinity Purification of Analytes prior to Multiplexed Liquid Chromatography–Tandem Mass Spectrometry. Clinical Chemistry. 2012;58(12):1711–1716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Strathmann FG, Laha TJ, Hoofnagle AN. Quantification of 1α,25-Dihydroxy Vitamin D by Immunoextraction and Liquid Chromatography–Tandem Mass Spectrometry. Clinical Chemistry. 2011;57(9):1279–1285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hoofnagle AN, Laha TJ, Donaldson TF. A rubber transfer gasket to improve the throughput of liquid–liquid extraction in 96-well plates: Application to vitamin D testing. Journal of Chromatography B. 2010;878(19):1639–1642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nadler SB, Hidalgo JH, Bloch T. Prediction of blood volume in normal human adults. Surgery. 1962;51(2):224–232. [PubMed] [Google Scholar]
- 23.Henderson CM, Lutsey PL, Misialek JR, et al. Measurement by a Novel LC-MS/MS Methodology Reveals Similar Serum Concentrations of Vitamin D–Binding Protein in Blacks and Whites. Clinical Chemistry. 2016;62(1):179–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hoofnagle AN, Eckfeldt JH, Lutsey PL. Vitamin D–Binding Protein Concentrations Quantified by Mass Spectrometry. New England Journal of Medicine. 2015;373(15):1480–1482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Yamazaki Y, Okazaki R, Shibata M, et al. Increased Circulatory Level of Biologically Active Full-Length FGF-23 in Patients with Hypophosphatemic Rickets/Osteomalacia. The Journal of Clinical Endocrinology & Metabolism. 2002;87(11):4957–4960. [DOI] [PubMed] [Google Scholar]
- 26.Levey AS, Stevens LA. Estimating GFR Using the CKD Epidemiology Collaboration (CKD-EPI) Creatinine Equation: More Accurate GFR Estimates, Lower CKD Prevalence Estimates, and Better Risk Predictions. American Journal of Kidney Diseases. 2010;55(4):622–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Efron B The Jackknife, the Bootstrap, and Other Resampling Plans. Philadelphia, Pa.: Society for Industrial and Applied Mathematics (SIAM, 3600 Market Street, Floor 6, Philadelphia, PA 19104), 1982. [Google Scholar]
- 28.Mawer EB, Backhouse J, Holman CA, Lumb GA, Stanbury SW. The distribution and storage of vitamin D and its metabolites in human tissues. Clin Sci. 1972;43(3):413–431. [DOI] [PubMed] [Google Scholar]
- 29.Bikle DD, Gee E, Halloran B, Kowalski MA, Ryzen E, Haddad JG. Assessment of the free fraction of 25-hydroxyvitamin D in serum and its regulation by albumin and the vitamin D-binding protein. J Clin EndocrinolMetab. 1986;63(4):954–959. [DOI] [PubMed] [Google Scholar]
- 30.Masuda S, Byford V, Arabian A, et al. Altered Pharmacokinetics of 1α,25-Dihydroxyvitamin D3 and 25-Hydroxyvitamin D3 in the Blood and Tissues of the 25-Hydroxyvitamin D-24-Hydroxylase (Cyp24a1) Null Mouse. Endocrinology. 2005;146(2):825–834. [DOI] [PubMed] [Google Scholar]
- 31.Tomon M, Tenenhouse HS, Jones G. Expression of 25-hydroxyvitamin D3-24-hydroxylase activity in Caco-2 cells. An in vitro model of intestinal vitamin D catabolism. Endocrinology. 1990;126(6):2868–2875. [DOI] [PubMed] [Google Scholar]
- 32.Akeno N, Saikatsu S, Kawane T, Horiuchi N. Mouse vitamin D-24-hydroxylase: molecular cloning, tissue distribution, and transcriptional regulation by 1alpha,25-dihydroxyvitamin D3. Endocrinology. 1997;138(6):2233–2240. [DOI] [PubMed] [Google Scholar]
- 33.Liu PT, Stenger S, Li H, et al. Toll-like receptor triggering of a vitamin D-mediated human antimicrobial response. Science. 2006;311(5768):1770–1773. [DOI] [PubMed] [Google Scholar]
- 34.Flanagan JN, Zheng S, Chiang KC, et al. Evaluation of 19-nor-2alpha-(3-hydroxypropyl)-1alpha,25-dihydroxyvitamin D3 as a therapeutic agent for androgen-dependent prostate cancer. Anticancer Res. 2009;29(9):3547–3553. [PubMed] [Google Scholar]
- 35.Stubbs JR, Idiculla A, Slusser J, Menard R, Quarles LD. Cholecalciferol supplementation alters calcitriol-responsive monocyte proteins and decreases inflammatory cytokines in ESRD. J Am Soc Nephrol. 2010;21(2):353–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Koszewski NJ, Horst RL, Goff JP. Importance of apical membrane delivery of 1,25-dihydroxyvitamin D3 to vitamin D-responsive gene expression in the colon. Am J Physiol Gastrointest Liver Physiol. 2012;303(7):G870–878. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.de Boer IH, Sachs MC, Chonchol M, et al. Estimated GFR and Circulating 24,25-Dihydroxyvitamin D3 Concentration: A Participant-Level Analysis of 5 Cohort Studies and Clinical Trials. American Journal of Kidney Diseases. 2014;64(2):187–197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Wong T, Wang Z, Chapron BD, et al. Polymorphic Human Sulfotransferase 2A1 Mediates the Formation of 25-Hydroxyvitamin D3-3-O-Sulfate, a Major Circulating Vitamin D Metabolite in Humans. Drug Metab Dispos. 2018;46(4):367–379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Higashi T, Shimada K, Toyo’oka T. Advances in determination of vitamin D related compounds in biological samples using liquid chromatography–mass spectrometry: A review. Journal of Chromatography B. 2010;878(20):1654–1661. [DOI] [PubMed] [Google Scholar]
- 40.Prosser DE, Kaufmann M, O'Leary B, Byford V, Jones G. Single A326G mutation converts human CYP24A1 from 25-OH-D<sub>3</sub>-24-hydroxylase into -23-hydroxylase, generating 1α,25-(OH)<sub>2</sub>D<sub>3</sub>-26,23-lactone. Proceedings of the National Academy of Sciences. 2007;104(31):12673–12678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Kaufmann M, Schlingmann KP, Berezin L, et al. Differential diagnosis of vitamin D-related hypercalcemia using serum vitamin D metabolite profiling. J Bone Miner Res. 2021;36(7):1340–1350. [DOI] [PubMed] [Google Scholar]
- 42.GRAY RW, WEBER HP, DOMINGUEZ JH, LEMANN J JR. The Metabolism of Vitamin D3 and 25-Hydroxyvitamin D3 in Normal and Anephric Humans. The Journal of Clinical Endocrinology & Metabolism. 1974;39(6):1045–1056. [DOI] [PubMed] [Google Scholar]
- 43.Jones KS, Assar S, Vanderschueren D, Bouillon R, Prentice A, Schoenmakers I. Predictors of 25(OH)D half-life and plasma 25(OH)D concentration in The Gambia and the UK. Osteoporos Int. 2015;26(3):1137–1146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Batchelor AJ, Compston JE. Reduced plasma half-life of radio-labelled 25-hydroxyvitamin D3 in subjects receiving a high-fibre diet. Br J Nutr. 1983;49(2):213–216. [DOI] [PubMed] [Google Scholar]
- 45.Haddad JG Jr., Rojanasathit S. Acute administration of 25-hydroxycholecalciferol in man. J Clin Endocrinol Metab. 1976;42(2):284–290. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.