Correlation between serum 1,25-dihydroxy-vitamin D and 25-hydroxyvitamin D in response to analytical procedures; a systematic review and meta-analysis

Objectives: In this study, the aim is to provide a more detailed understanding of vitamin D metabolism by evaluating the correlation between 1,25-dihydroxyvitamin D (1,25(OH)2D) and 25-hydroxyvitamin D (25(OH)D) according to the variations in measurement methods and clinical conditions. Methods: We searched PubMed, Embase, and Web of Science for studies reporting correlation results between 1,25(OH)2D and 25(OH)D. We performed a meta-analysis based on the correlation results of 1,25(OH)2D and 25(OH)D in di ﬀ erent clinical conditions. We included a total of 63 studies and our laboratory ’ s results in the meta-analysis. The studies were categorized into high-quality methods group (HQMG), medium-quality methods group (MQMG), and low-quality methods group (LQMG) based on the 25(OH)D and 1,25(OH)2D measurement. Results: In the healthy, renal disease, and other disease groups, the highest correlation values were observed in the studies categorized as HQMG, with values of 0.35 (95 % CI; 0.23 – 0.48), 0.36 (95 % CI; 0.26 – 0.42), and 0.36 (95 % CI; 0.22 – 0.48), respectively. Signi ﬁ cant statistical heterogeneity was observed in the healthy, renal disease, and other disease groups, with I2 values of 92.4 , 82.7, and 90.7 %, respectively (p<0.001). Both Funnel plots and the results of Egger ’ s and Begg ’ s tests indicated no statistically signi ﬁ cant bias across all studies. Conclusions: A signi ﬁ cantly low correlation was found between 25(OH)D and 1,25(OH)2D. However, higher correlations were found in the studies categorized as HQMG. Various factors, including methodological inadequacies and disparities, might contribute to this. In the future, with more accurate and reproducible measurements of 1,25(OH)2D, a clearer understanding of vitamin D metabolism will be achieved.


Introduction
Vitamin D deficiency (VDD) is the most common worldwide, and more than 1 billion people are known to have a deficiency [1].Although VDD was known long ago, rickets and osteomalacia were first distinctly described in 1645 [2].The experimental animal studies have clarified its synthesis, mechanism, and treatments for its deficiencies [3][4][5][6].It has been stated that there has been a considerable increase in the number of individuals affected by VDD in recent years, especially common in the elderly and those who stay indoors for longer periods, people with pigmented skin, pregnant women, vegans, and children of developmental age.

The regulation of vitamin D metabolism
Vitamin D metabolism and regulation are very complex (Figure 1).Vitamin D3 is produced in the skin by ultraviolet-B (290-315 nm) irradiation of 7-dehydrocholesterol (7-DHC).Irradiation of 7-DHC produces pre-D3 (which later becomes vitamin D3), lumisterol, and tachysterol [7].Melanin in the skin absorbs UV radiation, potentially reducing the skin's ability to produce vitamin D from sunlight.This could be a key factor contributing to lower 25-hydroxyvitamin D (25(OH)D) levels (which is a well-established indicator of vitamin D levels) in Black and Hispanic individuals living in regions with less direct sunlight [8].The 25(OH)D levels can exhibit significant seasonal fluctuations, with elevated concentrations in the summer and decreased levels in the winter.
Vitamin D is initially metabolized to 25(OH)D, predominantly in the liver, and then further converted to 1,25-dihydroxyvitamin D (1,25(OH)2D), mainly in the kidney.This final form, 1,25(OH)2D, is the primary active form responsible for most of vitamin D's effects in the body [9].
The 25-hydroxylase-CYP2R1 is highly controlled by a variety of diseases (obesity, diabetes mellitus, starvation, infection, inflammation, cancer, etc.), and although many regulatory factors have now been identified, significant gaps remain.Genetic silencing mutations in CYP2R1 can cause rickets, osteomalacia, or other clinical conditions.Serum 25(OH)D may not reflect only the vitamin D produced by diet and skin.In particular, hepatic 25(OH)D synthesis has a complex regulation involving many possible hormones and factors.
The enzyme 1,25-dihydroxylase CYP27B1 is mainly present in the renal proximal tubule, and its production is influenced by changes in parathyroid hormone (PTH), fibroblast growth factor 23 (FGF23), 1,25(OH)2D, calcium, and phosphate levels [10].A specific region in the enhancer region of renal CYP27B1 is responsible for responding to PTH, FGF23, and 1,25(OH)2D regulation [11].However, this region is not open to such regulation in non-kidney tissues like skin, immune cells, and adipose tissue.In non-renal tissues, a different enhancer region of CYP27B1 is regulated by various factors such as interferon-γ (IFN-γ), cytokines like tumor necrosis factor-α (TNF-α), or leptin [12,13].These feedback loops play a vital role in regulating 1,25(OH)2D production in the kidney, which differs from CYP27B1 in other cell types, including distal renal tubular cells that are minimally affected by PTH [14].CYP24A1 regulates 1,25(OH) 2D levels in other tissues and is stimulated by TNF-α and IFN-γ.However, in some conditions like sarcoidosis, where macrophages produce excess 1,25(OH)2D without proper CYP24A1 regulation, it can lead to hypercalcemia and hypercalciuria [15][16][17].
25(OH)D and 1,25(OH)2D are carried in the bloodstream by a protein called vitamin D-binding protein (DBP).This ) is freely available [9].Total 25(OH)D or 1,25(OH)2D includes three fractions: DBP-bound, weakly bound to albumin (also called bioavailable, as they can easily dissociate from albumin), and free compounds.Changes in DBP levels due to genetic factors, hormonal status, or liver and kidney conditions can impact the accuracy of 25(OH)D levels as a marker of vitamin D status, especially in pregnancy, estrogencontaining contraceptives, and liver or kidney diseases.Through the measurement of total 25(OH)D, DBP, and albumin by the principles of protein-ligand binding kinetics, it is possible to calculate both the 25(OH)D/1,25(OH)2D-DBP and 25(OH)D/1,25(OH)2D-albumin affinities [9,18].However, these are used limitedly due to their impracticality and complexity.

The clinical decision limits for vitamin D
Determining the clinical decision limits of 25(OH)D is complicated.In 1997, the Food and Nutrition Board of the Institute of Medicine identified serum 25(OH)D as a good marker for evaluating vitamin D status [19].However, there was not enough data at that time to fully understand its normal range in the body.Estimated values were insufficient, and it has since become evident that vitamin D deficiency is more prevalent than previously thought.It is affected by various factors such as seasons, diet, medications, and inaccurate measurement methods.Studies have investigated the relationship between serum PTH levels and 25(OH)D levels.It is known that PTH levels increase at low levels of 25(OH)D values but decrease at 75-110 nmol/L levels [20][21][22].In this context, high PTH levels suggest the body is adapting to lower calcium intake.However, whether this adaptation indicates better health is still uncertain [23].To date, no definitive functional change in 25(OH)D levels is known at the point where PTH stabilizes at the lower end of the healthy reference value (or clinical decision threshold level) of 25(OH)D.Vitamin D significantly increases circulating 1,25(OH)2D concentrations, but in vitamin D users, this increase is suppressed by calcium co-administration.
The relationship between 25(OH)D and 1,25(OH)2D is multifaceted and complex.In this study, we planned to conduct a meta-analysis and systematic review to elucidate the relationship between 25(OH)D and 1,25(OH)2D, aiming to gain a better understanding of vitamin D metabolism.This correlation was designed considering two different scenarios.In the first scenario, the measurement methods for 25(OH)D and 1,25(OH)2D, along with the analytical limitations associated with these methods, were considered.In the second scenario, various clinical conditions, such as in healthy individuals, kidney diseases (where 1,25(OH)2D synthesis takes place), and systemic diseases, were individually evaluated by considering their specific effects on vitamin D metabolism.

Materials and methods
This meta-analysis was conducted following the guidelines recommended by the PRISMA statement [29].

The strategy of publication search
In this meta-analysis study, comprehensive search strategies were created to identify publications.The selection and extraction criteria of publications Four independent reviewers, who were blinded to the study details, read the titles and abstracts of all reports found through electronic searches (MAS, FDA, NYS, DY).For studies that seemed to fulfill the inclusion criteria, or when the title and abstract provided insufficient data for a definitive decision, the complete report was acquired.The reliability between reviewers was assessed using Cohen's kappa test, setting an acceptable threshold at 0.74.Discussions among the reviewers resolved disagreements about whether to include or exclude certain studies.The relationship between 1,25(OH)2D and 25(OH)D levels was analyzed using various clinical conditions and analytical techniques.
For this meta-analysis, studies were included if they reported correlation test results (such as Pearson correlation without or with log transformation, Spearman correlation, or regression analysis) conducted in serum or plasma, were published in English, had accessible full texts, and were conducted on human subjects.Studies were excluded if they lacked correlation values but provided interpretations, were animal studies, or involved other biological samples (such as cord blood, cerebrospinal fluid, etc.).

The classification of analytical methods
High-quality methods group (HQMG): Automated and traceable 25(OH)D and 1,25(OH)2D measurements, commercial liquid chromatographymass spectrometry/mass spectrometry (LC-MS/MS) measurement (ImmuTube ® LC-MS/MS assay), in-house LC-MS/MS measurement, and automated repeatable immunoassay (like LIASON) methods were used and given analytical performance for both tests (limit of detection, limit of quantification, repeatability, linearity, etc.).
Low-quality methods group (LQMG): The methods have no or insufficient information regarding 25(OH)D and 1,25(OH)2D measurements.

The classification of clinical conditions
The clinical conditions were evaluated in three groups: healthy, kidney diseases, and other illnesses.

The results of 1,25(OH)2D and 25(OH)D in our laboratory
In addition to the studies, the total of 25.457 results of 5424 patients who applied for various reasons between 2015 and 2023 and had measurements of 1,25(OH)2D and 25(OH)D in the Acıbadem Labmed Laboratory have also been included.Patients have been presented in three groups, like the other groups, based on their ICD codes, diagnoses, clinical information, and other laboratory test results.

Presentation of data
Data extraction was achieved by the four authors of the metaanalysis (MAS, DY, NSY, FDA).Detailed data from 63 articles, along with our laboratory results, are included in this meta-analysis.The following information was extracted from the studies: the year of the research, place where it was conducted, study design, sample size, gender, age, diseases, analytical measurement procedures, sample size, gender, correlation results between 25(OH)2D and 25(OH)D, and analytical quality considerations.Data were compiled into evidence tables, and a descriptive summary was formulated to assess the volume of data, various study characteristics, and outcomes (Table 1).
For measurements of 25(OH)D and 1,25(OH)2D in this study, information about the manufacturers, methods, sample types, analytical performances, and interferences belonging to the most commonly preferred brands are provided in Supplemental Table 1 (for 25(OH)D) and Supplemental Table 2 (for 1,25(OH)2D).

Statistical analysis
Meta-analysis was achieved on the correlation between 25(OH)D and 1,25(OH)2D.These analyses were done using Stata MP17 4.6.241(Stata Corp LLC, Texas, USA).Random effects meta-analyses were performed using the DerSimonian-Laird method.The χ2 and I2 statistics were used to assess the statistical heterogeneity among the included studies.Forest plots were drawn to describe the weighted correlation with 95 % confidence intervals (CI).In addition, the Funnel plot and Egger and Beggs tests were applied to explore the sources of bias.

Results
The flowchart of the study, according to the PRISMA statement, is presented in Figure 2. Initially, the search strategy retrieved 1388 references.After screening the titles and abstracts, 1325 articles were excluded due to unrelated topics.The entire texts of the remaining 264 articles were assessed, and 63 studies were included in the meta-analysis.In this meta-analysis, correlation analysis results of a total of 25.147 people, consisting of 63 studies and our laboratory data, were evaluated.The meta-analysis outcomes are represented according to clinical conditions in Figures 3-5.
Accordingly, in the healthy group, a total of 24 studies were evaluated.Among these, ten were classified as HQMG, eleven as MQMG, and three as LQMG.The correlation values were calculated as 0.35 (95 % CI, 0.23-0.48)with 91.2 % of heterogeneity (I2) for HQMG, 0.21 (95 % CI, 0.10-0.31)with 91.3 % for MQMG, and as 0.22 (95 % CI, 0.03-0.42)with 38.0 % for LQMG.The correlation value was determined in the total healthy group as 0.26 (95 % CI, 0.18-0.34)with 92.4 % of I2.It was observed that the correlation value was the highest in HQMG, followed by MQMG, and lastly, LQMG.Significant heterogeneity was detected in all groups except for the LQMG group and in the overall evaluation of the study.
In the case of renal diseases, a total of 19 studies were assessed.Among these, nine were categorized as HQMG, seven as MQMG, and three as LQMG.The correlation values were calculated as 0.34 (95 % CI, 0.26-0.42)with 78.6 % of I2 for HQMG, 0.28 (95 % CI, 0.17-0.38)with 75.0 % for MQMG, and 0.27 (95 % CI, 0.25-0.37)with 92.6 % for LQMG.In the overall analysis, encompassing both healthy and renal disease groups, the correlation value was determined to be 0.31 (95 % CI; 0.25-0.37)with 82.7 % of I2.It was observed that the correlation value was the highest in HQMG, followed by MQMG, and lastly, LQMG.Notably, significant heterogeneity was detected in all groups except for the LQMG group and in the comprehensive study assessment.
In the context of other diseases, 36 studies were examined.Among these, 12 were classified as HQMG, 13 as MQMG, and 11 as LQMG.The correlation values were calculated as 0.36 (95 % CI; 0.22-0.48)with 94.8 % of I2 for HQMG, as 0.19 (95 % CI; 0.09-0.30)with 71.6 % for MQMG, and as 0.16 (95 % CI; 0.01-0.32)with 89.6 % for LQMG.The correlation value was determined to be 0.25 (95 % CI; 0.17-0.32)with 90.7 % of I2 in the comprehensive analysis covering healthy and disease groups.It was observed that the correlation value was the highest in HQMG, followed by MQMG, and lastly, LQMG.Importantly, significant heterogeneity was detected in all groups.
As a result, the correlation values obtained from measurements conducted with HQMG are higher than those of MQMG and LQMG.
The results of the assessment for publication bias in the conducted study are presented in Figure 6.According to both the Funnel plots and the results of Egger and Begg's tests, it was determined that there was no statistically significant bias.

Discussion
The relationship between 25(OH)D and 1,25(OH)2D is quite complex.In addition to the different reasons mentioned above, it is especially related to the measurement of 1,25(OH) 2D.The characteristics of 25(OH)D and 1,25(OH)2D methods are presented in Supplemental Tables 1 and 2.
Since 2010, the U.S. National Institutes of Health, Office of Dietary Supplements (NIH-ODS), through the Vitamin D Standardization Program (VDSP), has been working to standardize the measurement of serum total 25(OH)D, which is the primary indicator of vitamin D levels.Studies have shown that the results of assays used to determine serum total 25(OH)D, comprising both 25-hydroxyvitamin D2 [25(OH)D2] and 25-hydroxyvitamin D3 [25(OH)D3], may vary depending on the specific assay method employed [93][94][95].
The VDSP is a cooperative venture involving the National Institutes of Health, National Institute of Standards and Technology (NIST), Office of Dietary Supplements (NIH-ODS) [96], Centers for Disease Control and Prevention (CDC), as well as the national survey laboratories in multiple countries, and vitamin D investigators worldwide [97].
The VDSP has enforced a reference measurement system that includes reference measurement procedures conducted at NIST and CDC, along with NIST Standard Reference Materials [98][99][100][101][102]. Additionally, it comprises the CDC  The VDSP has set strict criteria for assay performance, ensuring that measurement variability and bias meet the standards of a coefficient of variation (CV) of ≤10 % and a mean bias of ≤ 5 % [106,107].Despite highly successful standardization efforts in 25(OH)D measurements, the issues still need to be solved.In the VDSP's Intra-laboratory Study for the Assessment study, 12 assays were compared, and 9 out of the 12 assays demonstrated a mean bias within ≤ 5 %.Samples with high levels of 25(OH)D2 were essential in evaluating the effectiveness of the immunoassays, highlighting possible differences in response or recovery between 25(OH)D2 and 25(OH)D3 in various assays [108].
Serious problems were also encountered in the LC-MS/MS method, which was presented as a better method.Only 53 % of the LC-MS/MS assays met the VDSP criterion of mean %bias ≤5 %.Four assays showed a mean %bias between 12 % and 21 % among those that did not.A regression study using the concentrations of four vitamin D metabolites in 50 single donor samples found that implementing several LC-MS/MS assays was affected by the presence of 3-epi-25(OH)D3 [109].Significant correlation discrepancies and high bias values have also been reported for 25(OH)D measurements by immunoassay, chromatography, and mass spectrometry [110][111][112].
It has been observed that the analytical measurement difficulties are much greater in 1,25(OH)2D measurements compared to 25(OH)D.1,25(OH)2D is a compound found in very low concentrations (pmol/L) in circulation and is highly lipophilic.Furthermore, the structurally similar metabolic precursor 25(OH)D circulates at nmol/L concentrations, making assay specificity an analytical concern.Significant advancements have been made in measuring 1,25(OH)2D.In 1974, a radioreceptor assay (RRA) was developed, utilizing the competitive binding of 1,25(OH)2D and a tritiated tracer to its nuclear receptor isolated from the calf thymus [113].The first RIA that measured 1,25(OH)2 D was introduced in 1978 [114].RIA for 1,25(OH)2 D using a radio iodinated (125 I) tracer was invented [115].The assay involves acetonitrile extraction and purification of endogenous 1,25(OH)2 D by solid phase chromatography and quantification by RIA.
In this study, our laboratory results are particularly crucial.According to the correlation results obtained with automated systems in a quite extensive patient group, 1,25(OH)2D and 25(OH)D measurements were found to be significantly higher in healthy individuals compared to the groups with renal and other diseases, with correlation coefficients of 0.50 (0.42-0.58), 0.26 (0.16-0.36), and 0.26 (0.23-0.29) respectively.A moderately significant correlation was observed in the healthy group.We believe that these results, obtained through the use of the same systems across all groups and with a sufficient amount of data, represent the best data currently available that demonstrates the current relationship.
Exacerbating the issue of low concentration is the poor ionization of the analyte, coupled with the potential complications arising from the derivatization required for sufficient analytical sensitivity.Additionally, poor sample preparation techniques can have a significant adverse impact on clinical performance in LC-MS/MS-based methods [125].It is worth noting that LC-MS/MS is a complex and specialized technique that requires advanced equipment Under standardized conditions, a recently introduced automated immunoassay demonstrates strong agreement with measurements obtained using a liquid chromatographytandem mass spectrometry reference method (LC-MS/MS) [126].Nonetheless, recent findings from DEQAS reveal that coefficients of variation within specific tests and mean 1,25(OH)2D levels between different test procedures can exhibit fluctuations of over 20 % [127].
According to a study conducted by Zittermann and colleagues, the measurement of circulating 1,25(OH)2D was carried out using two different methods: an LC-MS/MS method provided by Immundiagnostik and an automated immunoassay test provided by DiaSorin.The study found a correlation (r=0.534) and an agreement (62 %) between the two methods and highlighted the need for additional standardization studies [128].A recent meta-analysis has also revealed that the measurement procedure can significantly  impact circulating 1,25(OH)2D levels.These differences in measurement make it challenging to compare results between labs and establish consistent reference values for circulating 1,25(OH)2D levels.Consequently, automation and standardization are crucial for improving the reliability of testing procedures [129].It is worth noting that 80.8 % of the 1,25(OH)2D assays included in the meta-analysis were RIA and radioreceptor assays.
Upon closer examination, it is observed that there is a statistically insignificant or low correlation between 25(OH) D and 1,25(OH)2D in general.However, as can be seen in our results, the highest correlation in all three disease groups is found in HQMG.The highest correlations in the HQMG group are observed in all groups.The results in the renal diseases group are exciting.This may be due to this patient group taking Vitamin D supplements.However, the lower correlation in other diseases suggests that the relationship is highly complex and disrupted by different mechanisms in various diseases.
Even though the HQMG shows the highest correlation values in the healthy group, this is still a weak correlation.However, vastly different and heterogeneous results are present.The most significant factors here are methodological challenges, the short half-life of 1,25(OH)2D, and its complex regulations.While there is a direct enzymatic transformation of 25(OH)D into 1,25(OH)2D, a relationship between their serum levels may be noted.1,25(OH)2D can directly suppress the production of 1α-hydroxylase and indirectly by reducing PTH levels and promoting FGF23 production.This feedback mechanism is crucial for preventing hypercalcemia.As a result, the level of 1,25(OH)2D is not influenced by the circulating amount of 25(OH)D [130,131].We anticipate that with improved methodologies, the correlation value could increase in the future.
This study has notable limitations.In the search that was carried out, the term 'correlation' was explicitly looked for in the title, keywords, and abstract.There might be studies that do not mention the term "correlation" but discuss it within the text.While some studies may have used successful measurement procedures, they may not have been explicitly stated or may have been inadequately presented.Another significant limitation is that we did not consider age and gender differences in our analysis.We refrained from making distinctions based on age and gender as we believed it might reduce the number of studies in each group.In the studies included in this meta-analysis, different correlation analyses (Pearson correlation without or with log transformation, Spearman correlation, or regression analysis) were used.We included all of these correlation analyses in our study.
When all the results are evaluated, this study is the first meta-analysis conducted considering differences in methodological and health situations between 25(OH)D and 1,25(OH)2D.Both in the examination of Vitamin D metabolism and the relationship between 25(OH)D and 1,25(OH) 2D, differences in methodological and health situations are crucial and must be considered.

Figure 1 :
Figure 1: Vitamin D metabolism and regulation.
Articles published between 2005 and 2023 without language restrictions are in MEDLINE (via PubMed), Embase, and Web of Science.Using the keywords "(25-hydroxy D OR 25-hydroxycholecalciferol OR 25OHD OR 25-OH vitamin D OR calcidiol) AND (1,25-dihydroxy vitamin D OR 1,25-dihydroxycholecalciferol OR 1,25-dihydroxy vitamin D OR calcitriol)) NOT (animal)) NOT (review)) NOT (case report)) AND (correlation))." Vitamin D Standardization Certification Program and partnerships with the College of American Pathologists and Vitamin D external quality assessment scheme [103-105].

Figure 2 :
Figure 2: The flowchart of the study is based on the PRISMA* statement.*PRISMA (preferred reporting items for systematic reviews and meta-analyses).From: Page et al. [29].

Figure 4 :
Figure 4: Forest graph of the correlation between 25(OH)D and 1,25(OH)2D in the renal diseases group.

Figure 5 :
Figure 5: Forest graph of the correlation between 25(OH)D and 1,25(OH)2D in other disease groups.

Figure 6 :
Figure 6: Funnel plots, Egger's and Begg's test results of all groups.

Table  :
Detailed data from  articles and our laboratory results are included in this meta-analysis.These data consist of the year of the research, place where it was conducted, study design, sample size, gender, age, diseases, analytical measurement procedures, sample size, gender, correlation results between (OH)D and (OH)D, and analytical quality performances.