Evaluation of the age-specific relationship between PTH and vitamin D metabolites

A commonly used method for determining vitamin D sufficiency is the suppression of excess PTH secretion. Conventionally, the main circulating vitamin D metabolite 25(OH)D is used for this assessment, however, the cut-off data for this parameter vary widely in the literature. The role of other metabolites as markers of vitamin D status is actively debated. The aim of our study was to assess the relationship between PTH, age and parameters characterizing vitamin D status, both “classical” – 25(OH)D3, and “non-classical” – 24,25(OH)2D3 and 25(OH)D3/24,25(OH)2D3 (vitamin D metabolite ratio, VMR). This prospective non-controlled cohort study included 162 apparently healthy Caucasian adult volunteers. When PTH was binarized according to the median value, at VMR < 14.9, 25(OH)D3 > 9.7 ng/mL and 24,25(OH)2D3 > 0.64 ng/mL there was a pronounced relationship between PTH and age (p = 0.001, p = 0.023 and p = 0.0134 respectively), with the prevalence of higher PTH levels in older individuals and vice versa. Moreover, at an age of <40.3 years, there was a pronounced relationship between PTH and VMR (p < 0.001), and similarly at an age of <54.5 years, there was a pronounced relationship between PTH and 25(OH)D3 (p = 0.002) as well as between PTH and 24,25(OH)2D3 (p = 0.0038): in younger people, higher PTH values prevailed only in the range of vitamin D insufficiency, while in the older age group this relationship was not demonstrated and PTH values were in general above the median. VMR controlled the correlation between PTH and age more strongly than metabolites 25(OH)D3 and 24,25(OH)2D3 (p = 0.0012 vs. p > 0.05 and p = 0.0385 respectively). The optimal threshold was found equal to 11.7 for VMR such that the relationship between PTH and age in the subset of participants with VMR < 11.7 was characterized by a correlation coefficient of ρ = 0.68 (p < 0.001), while the cohort with VMR > 11.7 was characterized by a very weak correlation coefficient of ρ = 0.12 (p = 0.218), which is non-significant. In summary, our findings suggest that the relationship between PTH and vitamin D is age-dependent, with a greater susceptibility to elevated PTH among older individuals even with preserved renal function, likely due to the resistance to vitamin D function. We propose VMR can be considered as a potential marker of vitamin D status. These findings require confirmation in larger population-based studies.


Introduction
The main biological role of vitamin D is the absorption of dietary calcium.One of the most commonly used criteria in determining vitamin D sufficiency is the suppression of the excess parathyroid hormone (PTH) secretion, since the increase in PTH concentration reflects compensation for the inadequate absorption of calcium in the intestine in the state of vitamin D insufficiency.Conventionally, the main circulating vitamin D metabolite 25(OH)D is used for this assessment, however, the cut-off data for this parameter vary widely in the literature (Sai et al., 2011).The role of other metabolites as markers of vitamin D status is actively debated, but their relationship with PTH has been studied to a much lesser extent.
Under the conditions of the vitamin D deficiency, the synthesis of the major predominantly inactive metabolite (24,25(OH) 2 D) is minimized in favor of producing sufficient amounts of the active form of vitamin D (1,25(OH) 2 D) and maintaining adequate calcium absorption in the intestine, as was first shown by Tanaka and DeLuca (Tanaka and DeLuca, 1981).In severe vitamin D deficiency (defined as 25(OH)D < 10 ng/ mL), the concentration of 24,25(OH) 2 D becomes undetectable (Kaufmann et al., 2014), while a significant decrease in the percentage of individuals with 24,25(OH) 2 D levels below the detection threshold is observed with 25(OH)D values >20 ng/mL (Cavalier et al., 2020).
It should be noted that very low serum concentrations of 24,25 (OH) 2 D may be observed due to a number of reasons, among which the differential diagnosis between vitamin D deficiency and an inactivating mutation of CYP24A1 (the enzyme responsible for the synthesis of 24,25 (OH) 2 D from 25(OH)D).To solve this problem, the calculation of the ratio of vitamin D metabolites (25(OH)D/24,25(OH) 2 D, vitamin D metabolite ratio, VMR) is proposed.Under normal conditions VMR is inversely correlated with 25(OH)D and is described by an exponential relationship, while 24,25(OH) 2 D has a strong positive correlation and linear relationship with 25(OH)D (Jones and Kaufmann, 2021).
The relationship between vitamin D and PTH may be influenced by various external and internal factors, with age being one of the prominent candidates.Older age is widely considered to be a risk factor for vitamin D deficiency (Holick et al., 2011;Dedov et al., 2021), which is explained primarily by decreased cutaneous synthesis (Wacker and Holick, 2013).On the other hand, a decrease in renal function, being common in older people, leads to a decrease in the synthesis of 1,25 (OH) 2 D and the consumption of 25(OH)D correspondingly (Dusso et al., 2011).In addition, it has long been established that calcium absorption decreases with age (Nordin et al., 1970).Some previously conducted studies aimed at clarifying the effect of age on the relationship between PTH and vitamin D, but they were mainly focused on 25(OH)D and 1,25 (OH) 2 D (Vieth et al., 2003;Valcour et al., 2012).
The aim of this study was to assess the relationship between PTH, age and parameters characterizing vitamin D status (both "classical" -25 (OH)D 3 , and "non-classical" -24,25(OH) 2 D 3 and VMR).

Study population and design
This was a prospective non-controlled cohort study.The study group included 162 apparently healthy Caucasian adult volunteers without a history of metabolic skeletal diseases (including osteoporosis) or disorders of calcium-phosphorus homeostasis.The exclusion criteria were: vitamin D supplementation or therapy that is presumably associated with alterations in vitamin D metabolism in the three month period prior to the study; severe obesity (body mass index (BMI) >35 kg/m 2 ); pregnancy; the presence of granulomatous disease, malabsorption syndrome, liver failure, or chronic kidney disease.All participants were recruited in the period from July 2019 to June 2023.Serum samples were either transferred directly to the laboratory for biochemical analyzes and PTH measurement or were stored at − 80 • C avoiding repeated freeze-thaw cycles for measurement of vitamin D metabolites at a later date.The study protocol was approved by the Ethics Committee of Endocrinology Research Centre, Moscow, Russia, on April 10, 2019 (abstract of record No. 6), all participants signed an informed consent to participate in the study.

Laboratory measurements
The serum vitamin D metabolites levels (25(OH)D 3 , 25(OH)D 2 and 24,25(OH) 2 D 3 ) were determined by ultra-high performance liquid chromatography in combination with tandem mass spectrometry (UPLC-MS/MS) using an in-house developed method, described earlier for 2019-2022 (Povaliaeva et al., 2020) and 2022-2023(Zhukov et al., 2022;Usoltseva et al., 2023).With this technique, the laboratory participates in the DEQAS quality assurance program (lab code 2388) and the results fall within the target range for the analysis of 25(OH)D and 1,25(OH) 2 D metabolites in human serum.Proficiency certificates for the determination of 25(OH)D were issued annually from 2020 to 2024.
PTH levels were evaluated by electrochemiluminescence immunoassay (ELECSYS, Roche, Switzerland; reference range for this and subsequent laboratory parameters are given in the Results section for easier reading).Biochemical parameters of blood serum were assessed by ARCHITECT c8000 analyzer (Abbott, Illinois, United States) using reagents from the same manufacturer according to standard methods.Calibration and quality control were performed in accordance with the manufacturers' recommendations.

Other measurements
At the baseline visit, participants provided medical history via a questionnaire.Serving of dairy products was defined as 100 g of cottage cheese, 200 mL of milk, 125 g of yogurt or 30 g of cheese.Participants' weight was measured in light indoor clothing with a medical scale to the nearest 100 g, and their height was measured with a wall-mounted stadiometer to the nearest centimeter.Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared.The glomerular filtration rate was calculated according to the recommendations of the NKF-ASN Task Force 2021 (Delgado et al., 2022).

Statistical analysis
Statistical analysis was performed using Statistica version 13.0 (StatSoft, Oklahoma, United States), Data Master (Azforus, Moscow, Russia) and Python statistics and graphics modules.The statistical techniques were aimed to assess the nonlinear relationship of factors.Significance assessment was based on nonparametric permutation test and the Occam's razor principle.The procedure for statistical calculations is described in details in the Appendix.Continuous variables were summarized as median and interquartile range (IQR), binary variables were presented as numbers and percentages.A p-value of <0.05 was considered statistically significant.

Results
The study group consisted predominantly of young women characterized by a moderately healthy lifestyle and suboptimal calcium intake (Table 1).
At the first stage the relationship between PTH and a combination of age and vitamin D status was studied using optimal valid partitioning (OVP) (Kuznetsova et al., 2014) technique described in Appendix.PTH was binarized in the following way: PTH_B = 0 (lower than median value) and PTH_B = 1 (higher than median value).The aim of performed OVP analysis was to separate in the best way groups with PTH_B = 1 and PTH_B = 0 by age and vitamin D status parameters and to evaluate the significance of this separation.Significance was evaluated using permutation test (Pesarin and Salmaso, 2010) with the number of permutations equal to 5000 and Occam's razor principle (Senko et al., 2022).The results of this analysis when vitamin D status is described by its metabolites 25(OH)D 3 , 24,25(OH) 2 D 3 , as well as VMR are presented in Figs. 1, 2, 3, respectively.
On the contrary, at an age of <54.5 years there was a pronounced relationship between PTH_B and 25(OH)D 3 : in the area of 25(OH)D 3 < 9.7 (quadrant IV) 14 cases of PTH_B = 1 versus 2 cases with PTH_B = 0, which differed greatly from the proportions in the previously mentioned quadrant I.The significance of the effect for age was estimated at p = 0.002, the significance of the effect for 25(OH)D 3 was estimated at p = 0.023.A similar layout was observed for 24,25(OH) 2 D 3 (Fig. 2).In participants with 24,25(OH) 2 D 3 > 0.64 there was a strong association between PTH_B and age.At an age of <54.5 years (quadrant I), cases with PTH_B = 0 predominated: 72 cases of PTH_B = 0 versus 38 cases with PTH_B = 1.At an age >54.5 years (quadrant II), cases with PTH_B = 1 predominated: 18 cases of PTH_B = 1 versus 3 cases with PTH_B = 0.
On the contrary, at an age of <54.5 years, there was a pronounced relationship between PTH_B and 24,25(OH) 2 D 3 : at 24,25(OH) 2 D 3 < 0.64 (quadrant IV) 20 cases of PTH_B = 1 versus 6 cases with PTH_B = 0, which was markedly different from the proportions in the previously mentioned quadrant I.The significance of the effect for age was estimated at p = 0.0038, the significance of the effect for 24,25(OH) 2 D 3 was estimated at p = 0.0134.
A somewhat stronger effect was observed when examining the association of PTH with the combination of age and VMR.It is seen from Fig. 3 that in participants with VMR < 14.9 there was a strong association between PTH_B and age.At an age of <40.3 years (quadrant IV), cases with PTH_B = 0 predominated: 56 cases of PTH_B = 0 versus 20 cases with PTH_B = 1.At an age >40.3 years (quadrant III), cases with PTH_B = 1 predominated: 18 cases of PTH_B = 1 versus 3 cases with PTH_B = 0.
On the contrary, at an age of <40.3 years, there was a pronounced relationship between PTH_B and VMR: at VMR > 14.9 (quadrant I) 29 cases of PTH_B = 1 versus 13 cases with PTH_B = 0, which was markedly different from the proportions in the previously mentioned quadrant IV.The significance of the effect for age was estimated at p < 0.001, the significance of the effect for VMR was estimated at p = 0.001.
At the second stage nonparametric technique evaluating the significance of the correlation between Y and X in groups that are formed by the third variable Z was used to assess the relationship between PTH, age and vitamin D status.The technique is discussed in Appendix.All pvalues were calculated using a previously discussed permutation test with 10,000 permutations.
The relationship between PTH and age in the entire study group was characterized by a weak positive correlation (r = 0.24, p = 0.0025), as depicted in Fig. 4.
Next, PTH and age were considered as Y and X variables.In Table 3 p-values p y , p x and p z are given when parameters 25(OH)D 3 , 24,25 (OH) 2 D 3 and VMR are used as variable Z.
It is seen from the table that the effect is significant at p = 0.0012 and at p = 0.0385 when VMR and 24,25(OH) 2 D 3 are used respectively as variable Z.The effect is not significant at p < 0.05 when 25(OH)D 3 is used as Z.Thus, VMR controls the correlation between PTH and age more strongly than the metabolites 25(OH)D 3 and 24,25(OH) 2 D 3 .The optimal threshold for VMR was found equal to 11.7.The linear Fig. 3. Association of PTH with the combination of age and vitamin D metabolite ratio (VMR).Red crosses correspond to PTH_B = 0 (lower than median value); green circles correspond to PTH_B = 1 (higher than median value).Presented on a logarithmic scale.dependence between PTH and age in the sub-set of 52 cases with VMR < 11.7 is seen from Fig. 5.In Fig. 6, there is no significant association between age and PTH in the sub-set of 110 cases with VMR > 11.7.

Discussion
This is an exploratory study aimed at the principles of the relationship between PTH and vitamin D. The results suggest that at higher values of 24,25(OH) 2 D 3 and lower values of VMR (which includes the range of vitamin D sufficiency), there was an association between PTH and age such that higher PTH values predominated in older individuals and vice versa.With values of 24,25(OH) 2 D 3 and VMR in the opposite range of the axis (including severe vitamin D deficiency), such a relationship was absent and PTH values above the median predominated at all ages.These findings are supported by the similar pattern observed for the conventional metabolite defining vitamin D status -25(OH)D 3 .Moreover, the influence of age was noted when analyzing the relationship between PTH and parameters of vitamin D metabolism: in younger people, higher PTH values prevailed only in the range of vitamin D insufficiency, while in the older age subgroups this relationship was not demonstrated and PTH values were in general above the median.
Furthermore, we have demonstrated a different nature of the relationship between PTH and age at VMR values less and more than the obtained "cut-off point", which is close to the lower quartile.With VMR values <11.7 (characteristic of the normal range of 25(OH)D 3 values and vitamin D sufficiency respectively), older patients were characterized by higher PTH values.This is consistent with the literature showing that older individuals have higher PTH levels than younger individuals even with comparable 25(OH)D levels (Vieth et al., 2003;Valcour et al., 2012).Our results, along with previous work by other groups, may indicate the need to consider age as an independent risk factor for secondary hyperparathyroidism, and also serve as confirmation of existing recommendations for higher vitamin D intake in older adults (Holick et al., 2011), as preventing PTH increase in them may require higher levels of vitamin D. It should be noted that at higher VMR values, which include the range of vitamin D deficiency, no association of PTH and age was noted, indicating a more complex regulation of PTH production in these conditions.Interestingly, VMR controlled the correlation between PTH and age more strongly than metabolites 25(OH)D 3 and 24,25 (OH) 2 D 3 , which strengthens its high potential as a marker of vitamin D status.
It might be assumed that the increased impact of the vitamin D status on serum PTH levels in older populations may reflect decreased calcium absorption and/or renal function observed with age.It seems unlikely to us that the results we observed could be associated with reduced production of the active metabolite 1,25(OH) 2 D 3 , since the included individuals had preserved renal function, while the production of 1,25 (OH) 2 D 3 appears to be determined primarily by the functional state of the kidneys, and not by the increasing age itself (Vieth et al., 2003).The hypothesis of relative resistance to the action of 1,25(OH) 2 D 3 in older individuals seems more likely to apply (Eastell et al., 1991;Pattanaungkul et al., 2015).
Our cohort was relatively small with a predominance of young people, and therefore it seems not feasible to project the obtained results to the general population.In addition, it was characterized by suboptimal dietary calcium intake, which may have affected the results.Other limitations of the study include the following: key parameters of calcium-phosphorus homeostasis (1,25(OH) 2 D, fibroblast growth factor-23, ionized calcium, magnesium) were not taken into account; a single measure of PTH is not optimal due to biologic variation especially in this population with suboptimal calcium intake; and the participants were not screened for other abnormalities (such as elevated serum transaminase, hyperglycemia, clotting factor, low BMD, etc).However, our data provide an important direction for future research, since an elevated PTH is associated with both bone health deterioration (Sahota et al., 2001;Sahota et al., 2004;Qu et al., 2020) and non-skeletal consequences (Yang et al., 2016;Anderson et al., 2011;Grandi et al., 2011;Wu et al., 2023) even out of context of primary hyperparathyroidism.A better understanding of the dynamics of the relationship between PTH and vitamin D and the factors influencing this relationship is an important step towards improving clinical approaches.

Conclusions
The relationship between PTH and vitamin D is age-dependent, with a greater susceptibility to elevated PTH among older individuals even with preserved renal function, likely due to resistance to vitamin D function.Interestingly, VMR controlled the correlation between PTH and age more strongly than metabolites 25(OH)D 3 and 24,25(OH) 2 D 3 , which strengthens its high potential as a marker of vitamin D status.The findings require confirmation in larger population-based studies.

Institutional review board statement
This study was performed in line with the principles of the Declaration of Helsinki.Approval was granted by the Ethics Committee of Endocrinology Research Centre, Moscow, Russia on April 10, 2019 (abstract of record No. 6).Informed consent was obtained from all individual participants included in the study.

Informed consent statement
Informed consent was obtained from all subjects involved in the study.

Declaration of competing interest
The authors have no relevant financial or non-financial interests to disclose.

Fig. 1 .
Abbreviations: PTH, parathyroid hormone; eGFR, estimated glomerular filtration rate.aReference ranges are specified according to kit manufacturers' recommendations.bReference range is given for total 25(OH)D according to the clinical guidelines(Holick et al., 2011;Dedov et al., 2021); the 25(OH)D 2 fraction is negligible (<0.5 ng/mL in absolute values) for the purposes of this study.c Reference ranges are given according to the literature data(Dirks et al., 2016;Tang et al., 2017).

Fig. 4 .
Fig. 4. Association between PTH and age in the entire study group (n = 162).

Fig. 5 .
Fig. 5.The relationship between PTH and age in the subset of participants with VMR < 11.7 (n = 52) was characterized by a correlation coefficient of ρ = 0.68, which is significant at the p < 0.001 level according to the standard Student's test.

Fig. 6 .
Fig. 6.The relationship between PTH and age in the subset of participants with VMR > 11.7 (n = 110) was characterized by a very weak correlation coefficient of ρ = 0.12 (p = 0.218), which is non-significant.

Table 1
General characteristics of the study group.

Table 2
Laboratory parameters of the study group.

Table 3
Statistical significance of the effect.