Effects of 25-Hydroxyvitamin D3 Therapy on Bone Turnover Markers

ORIGINAL

ARTICLE

E n d o c r i n e

C a r e

Effects of 25-Hydroxyvitamin D3 Therapy on Bone Turnover Markers and PTH Levels in Postmenopausal Osteoporotic Women Treated with Alendronate Jose´ M. Olmos, Jose´ L. Hernańdez, Javier Llorca, Daniel Nan, Carmen Valero, and Jesu´s Gonza´lez-Macías Department of Internal Medicine (J.M.O., J.L.H., D.N., C.V., J.G.-M.), Hospital Universitario Marque´s de Valdecilla-Instituto de Formación e Investigación Marqués de Valdecilla (IFIMAV), Universidad de Cantabria, Red Temática de Investigación Cooperativa en Envejecimiento y Fragilidad (RETICEF); and Epidemiology Unit (J.L.), Medical School, Universidad de Cantabria, Centro de Investigación Biomédica en Red (CIBER) Epidemiología y Salud Pu´blica, E-39008 Santander, Spain

Objective: Our objective was to know the extent to which a fall in bone turnover markers is influenced by serum 25-hydroxyvitamin D (25OHD) levels in patients on alendronate (ALN) treatment. Design, Participants, and Setting: A total of 140 postmenopausal osteoporotic women were randomized to receive either ALN or ALN plus 25OHD3 (ALN⫹VitD) over a 3-month period. Serum 25OHD, PTH, C-terminal telopeptide of type I collagen (CTX), and amino-terminal propeptide of type I collagen (P1NP) were measured at baseline and at the end of the 3 months. Results: 25OHD rose four times above baseline levels in the ALN⫹VitD group, whereas no changes were seen in the ALN group. Administering ALN resulted in a significant decline in both serum CTX (53 ⫾ 24%) and P1NP (46 ⫾ 19%). After ALN⫹VitD, the fall in CTX amounted to 61 ⫾ 20% (P ⫽ 0.06 compared with ALN) and P1NP to 50 ⫾ 23% (P ⫽ 0.35). When patients were divided into those below and above 20 ng/ml of baseline serum 25OHD, in those below, CTX decreased by 48 ⫾ 26% in the ALN group and by 61 ⫾ 17% in the ALN⫹VitD group (P ⫽ 0.015). For P1NP, the corresponding figures were 43 ⫾ 20 and 50 ⫾ 23% (P ⫽ 0.2). In patients above 20 ng/ml, no differences were seen regarding CTX (58 ⫾ 21% decrease in the ALN group and 60 ⫾ 23% in the ALN⫹VitD group; P ⫽ 0.7) or P1NP (49 ⫾ 18 and 50 ⫾ 20%; P ⫽ 0.9). Conclusions: Administration of 25OHD3 is not an indispensable requirement for bisphosphonates to develop their bone antiresorptive effect. In fact, in patients with vitamin D sufficiency, no benefit is observed when the vitamin is added. However, in patients with vitamin D deficiency, an approximately 25% greater fall in the bone resorption marker CTX is seen with its administration. (J Clin Endocrinol Metab 97: 4491– 4497, 2012)

isphosphonates (BPs) are the most prescribed drugs in the management of osteoporosis. Their efficacy has been extensively studied in randomized controlled trials that have almost always included calcium and vitamin D supplements as adjunctive therapy. Indeed, vitamin D repletion has often been a prerequisite inclusion criterion (1). Current guidelines recommend the use of calcium and vitamin D supplementation along with major anti-osteo-

B

porotic drugs (2, 3), because low 25-hydroxyvitamin D (25OHD) levels are frequent in osteoporotic patients (4). However, a clear knowledge about the role of vitamin D in osteoporosis response to BPs, or even about the necessity for vitamin D supplementation with the use of these agents, is still lacking. Contradictory results with fractures, bone mineral density (BMD), and bone turnover markers (BTMs) as outcome variables, have been

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2012 by The Endocrine Society doi: 10.1210/jc.2012-2999 Received August 6, 2012. Accepted September 10, 2012. First Published Online October 5, 2012

Abbreviations: ALN, Alendronate; BMD, bone mineral density; BP, bisphosphonate; BTM, bone turnover marker; CTX, C-terminal telopeptide of type I collagen; CV, coefficients of variation; GFR, glomerular filtration rate; 25OHD, 25-hydroxyvitamin D; P1NP, aminoterminal propeptide of type I collagen.

J Clin Endocrinol Metab, December 2012, 97(12):4491– 4497

jcem.endojournals.org

4491

4492

Olmos et al.

Effects of 25OHD3 on Alendronate Efficacy

reported (5–11). A threshold value for 25OHD serum levels to advise vitamin D administration has been intended to be defined, once again with differing results (6, 12). Besides, the rationale for this potential interaction between BP effects and vitamin D administration is unclear, and little work has been done on the subject. Theoretically, the increase in serum PTH levels, and the osteoid undermineralization induced by vitamin D deficiency could be entertained among the possibilities to consider. In any case, whether vitamin D has an additive or a permissive role on the antiresorptive effect of BPs is not known. To add information on this issue, we carried out a 3-month study analyzing the extent to which the fall in BTM is influenced by serum 25OHD3 levels in patients on alendronate (ALN) treatment. To this aim, two groups of osteoporotic postmenopausal women were studied, one receiving ALN alone and the other ALN plus calcifediol (25OHD3). Calcifediol was given instead of vitamin D because of its pharmacokinetics advantages and the availability of studies carried out establishing its effects on PTH and C-terminal telopeptide of type I collagen (CTX) levels (13, 14). Results were analyzed considering the whole population as well as two groups of patients classified according to their baseline levels of 25OHD (above or below 20 ng/ml). The BTMs amino-terminal propeptide of type I collagen (P1NP) and CTX were the variables assessed. The study period was restricted to 3 months because it was considered that a longer period of time could perhaps be unethical, even if the favorable effect of vitamin D on ALN antiresorptive effect is still questionable.

Subjects and Methods Subjects Patients were enrolled as they came to our Osteoporosis Unit on a consecutive basis. They were community-dwelling postmenopausal women (at least 1 yr since menopause) with a BMD T-score below ⫺2.5 in at least one location (spine, femoral neck, or total hip) or a previous vertebral fracture, who had never been on treatment for osteoporosis. Patients who had previously taken calcium or vitamin D supplements were allowed to participate in the study. Reasons for exclusion were a history of metabolic bone disease (other than osteoporosis), malignancy, abnormal hepatic function tests, malabsorption, uncontrolled hyperthyroidism, serum creatinine higher than 1.7 mg/dl (151 ␮mol/liter), or other treatments known to affect bone metabolism, such as L-T4, anticonvulsants, glucocorticoids, or oral estrogens, within the preceding year. The local ethics committee approved the study protocol, and all the patients signed the informed consent.


Treatment protocol After the baseline assessment, patients were randomly assigned using a random number table to receive over 3 months: 1) ALN 70 mg once a week plus calcifediol 0.266 mg weekly (ALN⫹VitD group) or 2) ALN 70 mg once a week (ALN group). Subjects were instructed to take ALN in the morning under fasting conditions with a full glass of water and to remain upright for 30 min before the first food or beverage of the day.

Biochemical assessment Blood and urine samples were collected in the morning, under fasting conditions, at baseline and 3 months after starting treatment. At baseline, serum 25OHD, intact PTH, BTMs CTX and P1NP, total calcium, phosphate, creatinine, and albumin and 24-h urinary calcium and creatinine excretion were measured. Total calcium measurements were corrected for albumin concentration following a previously published formula (15). At the end of the 3 months of follow-up, serum 25OHD, PTH, and BMTs were assessed again. Serum 25OHD concentration was quantified by RIA using a commercial kit (DiaSorin, Stillwater, MN). The limit of detection was 4 ng/ml, and its intraassay and interassay coefficients of variation (CV) were 5 and 8%, respectively. Serum PTH 1– 84 was measured by electrochemiluminescence (DiaSorin LIAISON; normal range ⬍ 65 pg/ml) with an intra- and interassay CV of 3 and 7%, respectively. Regarding BTMs, P1NP was determined by RIA (Orion Diagnostica, Espoo, Finland; normal range 19 –102 ng/ ml; intra- and interassay CV 4 and 6%) and CTX by EIA (Nordic, Copenhagen, Denmark; normal range 0.166 –1.351 ng/ml; intra- and interassay CV 6 and 5%). Serum total calcium, phosphate, creatinine, and albumin as well as urinary calcium and creatinine excretion were measured by standard automated methods in a Technicon Dax autoanalyzer (Technicon Instruments, Milwaukee, WI). Glomerular filtration rate (GFR) was estimated according to the Modification of Diet in Renal Disease (MDRD-4) study formula (16).

Bone mass measurement BMD was measured by dual-energy x-ray absorptiometry (Hologic QDR 4500; Bedford, MA) at the lumbar spine, femoral neck, and total hip at baseline. In vivo precision was 0.4 –1.5% at the different measurement sites. Results are expressed as grams per square centimeter and T-score (defined as the number of SD below the mean value for young women). T-scores were calculated according to the Third National Health and Nutrition Examination Survey (NHANES III) reference database for femur measurements (17). Quality control was performed following the usual standards (18).

Statistical analysis The estimate of sample size was based on data from a preliminary group of patients. Assuming a 25% difference between the two treatment protocols regarding the drop in BMTs, a sample size of 130 patients was estimated to be sufficient to provide a statistical power of 80%, given a two-sided type I error of 0.05. With an assumed dropout rate of around 5%, the total number of patients needed was estimated at 136. The primary endpoint was the percent change in bone markers from baseline at month 3. Missing BTMs at either baseline or after baseline were not imputed. Paired Student’s t test or Wil-



coxon test was used to assess the significance of the treatment difference at each time point. ANOVA was performed to test the differences in serum PTH and BTM levels, adjusted for GFR, in both treatment arms. We considered P values ⬍0.05 statistically significant. All analyses were conducted using SPSS for Windows (SPSS Inc., Chicago, IL).

Results Of the 147 women who were enrolled in the study, 140 (95%) completed the 3-month treatment and were included in the analysis. Seventy-two had been initially allocated to ALN alone and 75 to ALN plus 25OHD3; in the first group, five patients were withdrawn because of protocol violations and one left the study for personal reasons; in the second group, one patient was withdrawn for protocol violation. Women in the two arms of the study had similar baseline characteristics (Table 1). Likewise, there were no differences in personal history of fractures in adulthood, tobacco or alcohol use, physical exercise, number of falls in the previous year, or number of chronic diseases; family history of osteoporotic fractures among first-degree relatives was also similar (data not shown). In particular, no differences regarding 25OHD concentraTABLE 1. Baseline characteristics, biochemical results, and BMD in patients who completed 3 months of therapy and were included in the analysis ALNⴙVit D (n ⴝ 73) 69 ⫾ 8 65 ⫾ 11 153 ⫾ 6 27.8 ⫾ 4.5 661 ⫾ 309 9.7 ⫾ 0.4

Age (yr) Weight (kg) Height (cm) BMI (kg/m2) Dairy calcium (mg/d) Corrected calcium (mg/dl) Phosphate (mg/dl) 3.6 ⫾ 0.5 Creatinine (mg/dl) 0.94 ⫾ 0.2 GFR (ml/min 䡠 1.73 m2) 65.4 ⫾ 14.7 Urine calcium (mg/24 h) 145 ⫾ 68 Urine calcium/creatinine 0.16 ⫾ 0.1 (mg/mg) 25OHD (ng/ml) 21 ⫾ 12 PTH (pg/ml) 60 ⫾ 22 P1NP (ng/ml) 50 ⫾ 26 CTX (ng/ml) 0.671 ⫾ 0.31 BMD, LS (g/cm2) 0.773 ⫾ 0.108 BMD, LS (T-score) ⫺2.69 ⫾ 1.20 BMD, FN (g/cm2) 0.593 ⫾ 0.074 BMD, FN (T-score) ⫺2.34 ⫾ 0.69 BMD, TH (g/cm2) 0.734 ⫾ 0.097 BMD, TH (T-score) ⫺01.70 ⫾ 0.81

ALN (n ⴝ 67) 67 ⫾ 8 66 ⫾ 12 154 ⫾ 6 27.6 ⫾ 4.9 684 ⫾ 293 9.6 ⫾ 0.4

P 0.3 0.7 0.1 0.7 0.6 0.8

3.5 ⫾ 0.5 0.96 ⫾ 0.2 63.6 ⫾ 13.8 140 ⫾ 85 0.15 ⫾ 0.1

0.3 0.5 0.5 0.7 0.5

21 ⫾ 9 70 ⫾ 34 49 ⫾ 28 0.644 ⫾ 0.30 0.808⫾.114 ⫺2.45 ⫾ 1.04 0.611 ⫾ 0.06 ⫺2.17 ⫾ 0.55 0.749 ⫾ 0.08 ⫺1.61 ⫾ 0.65

0.6 0.1 0.8 0.6 0.1 0.2 0.1 0.1 0.3 0.4

Values are expressed as mean ⫾ SD. The ALN⫹Vit D group includes patients treated with ALN plus calcifediol; the ALN group includes patients treated with ALN alone. Corrected calcium is albumincorrected serum total calcium. GFR was calculated according to the MDRD-4 formula. FN, Femoral neck; LS, lumbar spine; TH, total hip.

4493

tions were observed (21 ⫾ 12 vs. 21 ⫾ 9 ng/ml) in the two groups of treatment. We did not find any difference regarding serum PTH levels or BTMs. General results As expected, after 3 months of treatment, patients allocated to ALN⫹VitD showed a significant increase in serum concentration of 25OHD, which rose four times above baseline levels (the percentage of patients achieving 25OHD levels above 20 ng/ml was 100% and above 30 ng/ml was 96%). On the contrary, no significant changes were seen in patients treated with ALN alone. Serum PTH increased in the ALN group (P ⬍ 0.001) but remained at the baseline level in patients receiving ALN⫹VitD (P ⫽ 0.24). The increase in PTH in the ALN group varied with baseline 25OHD level. In patients below 20 ng/ml, it amounted to 12.5% and in those above to 27% (P ⫽ 0.04). A significant decline in both serum CTX and P1NP levels was seen after ALN alone administration, amounting to 53 ⫾ 24 and 46 ⫾ 19%, respectively. After ALN⫹VitD, the corresponding figures were 61 ⫾ 20% for CTX and 50 ⫾ 23% for P1NP (Table 2). Differences between ALN and ALN⫹VitD groups regarding the fall in CTX was almost significant (P ⫽ 0.06), but this was not so for P1NP (P ⫽ 0.35). In addition, a significant interaction (P ⫽ 0.01) between 25OHD baseline levels and 25OHD3 administration was observed regarding CTX response to ALN, suggesting that patients with different 25OHD levels may show a different response. No significant interaction was seen for P1NP (P ⫽ 0.1). BTMs decrease after ALN administration with and without vitamin D in patients above or below 20 ng/ml 25OHD Because of the interaction observed between baseline 25OHD and vitamin D administration regarding CTX decrease, the analysis was repeated after dividing the population into patients above or below 20 ng/ml (Fig. 1). This value was chosen because it is considered to define the threshold for vitamin D deficiency. In patients above 20 ng/ml, CTX decreased by 58 ⫾ 21% in the ALN group and by 60 ⫾ 23% in the ALN⫹VitD group (P ⫽ 0.7); regarding P1NP, the corresponding figures were 49 ⫾ 18 and 50 ⫾ 20% (P ⫽ 0.9). For patients with basal 25OHD below 20 ng/ml, the CTX decrease was 48 ⫾ 26% in the ALN group and 61 ⫾ 17% in the ALN⫹VitD group (P ⫽ 0.015); for P1NP, the decreases amounted to 43 ⫾ 20 and 50 ⫾ 23%, respectively (P ⫽ 0.2). Table 3 shows CTX and P1NP changes after GFR adjustment. Adjusted PTH changes are also given.

4494

Olmos et al.



TABLE 2. Serum levels of 25OHD, PTH, and BTMs at baseline and at 3 months, according to the group of therapy ALNⴙVitD 25OHD (ng/ml) PTH (pg/ml) P1NP (ng/ml) CTX (ng/ml)

Baseline 21 ⫾ 12 60 ⫾ 22 50 ⫾ 26 0.671 ⫾ 0.31

ALN

3 months 82 ⫾ 31b 57 ⫾ 22 24 ⫾ 16b 0.258 ⫾ 0.19b

Baseline 21 ⫾ 9 70 ⫾ 34 49 ⫾ 28 0.644 ⫾ 0.30

3 months 20 ⫾ 9 84 ⫾ 37a 26 ⫾ 14b 0.290 ⫾ 0.22b

The ALN⫹Vit D group includes patients treated with ALN plus calcifediol; the ALN group includes patients treated with ALN alone. a

P ⬍ 0.001, baseline vs. after treatment.

b

P ⬍ 0.0001, baseline vs. after treatment.

BTMs decrease after ALN administration with and without vitamin D in patients above or below 65 pg/ml of serum PTH levels at 3 months Because the 25OHD effect on BTMs is at least partially mediated through its action on PTH secretion, ALN and ALN⫹VitD responses were compared dividing patients into two groups: those above or below 65 pg/ml in the concentration of PTH at 3 months, this figure being considered the upper limit of normality. In patients below 65 pg/ml, CTX decreased by 54 ⫾ 26% in the ALN group and by 59 ⫾ 19% in the ALN⫹VitD group (P ⫽ 0.45); regarding P1NP, the corresponding figures were 44 ⫾ 22 and 50 ⫾ 21% (P ⫽ 0.30). For patients with final PTH level above 65 pg/ml, CTX decrease was 53 ⫾ 23% in the ALN group and 64 ⫾ 21% in the ALN⫹VitD group (P ⫽ 0.06); for P1NP, the decreases amounted to 47 ⫾ 17 and 50 ⫾ 28% respectively (P ⫽ 0.65). Relationship between baseline 25OHD and ALN-alone response In view of the previous results, the possibility that the response to ALN alone (without calcifediol) was different when administered to patients above or below 20 ng/ml was considered. Regarding CTX, ALN alone led to a 48 ⫾

FIG. 1. Mean percent decrease in concentration of CTX after administration of ALN plus calcifediol (ALN⫹VitD) or with ALN alone in the total population and in patients above or below 20 ng/ml of serum 25OHD.

26% decrease in patients with vitamin D deficiency and a 58 ⫾ 21% fall in those with vitamin D sufficiency (P ⫽ 0.07). As for P1NP, the respective figures were 43 ⫾ 20 and 49 ⫾ 18%, differences not being significant (P ⫽ 0.23).

Discussion Our study shows that ALN administration without 25OHD3 induces a deep and marked decline in serum CTX (53%), indicating that vitamin D is not an indispensable requirement for BPs to be effective in decreasing bone resorption. When calcifediol is given to the whole population, only a moderately greater decrease in CTX decline is observed. However, this greater decrease in CTX becomes deeper when only patients with basal 25OHD below 20 ng/ml are considered, although no difference is observed in patients with 25OHD above this figure. The difference in CTX decline between ALN⫹VitD and ALN groups in patients below 20 ng/ ml, amounts about a quarter of that seen in the latter, indicating that vitamin D contribution is only moderate in relation to that induced by ALN alone. In line with these findings, we have also seen in our study that when ALN is given without 25OHD3, CTX decline in patients with basal 25OHD above 20 ng/ml is greater (about 20%) than in those below, although only with a marginal significance (P ⫽ 0.07), perhaps influenced by the sample size. The possibility that inadequate levels of vitamin D impair the response to and effectiveness of BPs has frequently been suggested, although there is no evidence supporting the idea that vitamin D supplements add to ALN administration more than their own and independent effect on bone resorption. And certainly, no data exist indicating that vitamin D deficiency prevents BPs from acting on bone or that such a hypothetical prevention disappears after vitamin D repletion. Studies published on this subject are scarce, often contradictory, and deal with disparate outcomes (fractures, BMD, and BTMs).



4495

TABLE 3. Changes in PTH and bone remodeling markers after 25OHD3 administration, adjusted for GFR P1NP (ng/ml) CTX (ng/ml) PTH (pg/ml)

ALNⴙVitD ⫺26.1 (⫺30.7–21.6) ⫺0.4 (⫺0.5– 0.4) ⫺2.9 (⫺8.2–2.4)

ALN ⫺23.8 (⫺28.6 –19.1) ⫺0.4 (⫺0.4 – 0.3) 14.3 (8.8 –19.7)

P 0.5 0.2 ⬍0.001

Pⴕ 0.7 0.4 0.6

The ALN⫹Vit D group includes patients treated with ALN plus calcifediol; the ALN group includes patients treated with ALN alone. GFR was calculated according to the MDRD-4 formula. The 95% confidence intervals for the difference are shown in parentheses. P⬘ reflects the value for the interaction between 25OHD3 effect and GFR.

Adami et al. (5) reported in a first publication of the ICARO (Incidence and Characterization of Inadequate Clinical Responders in Osteoporosis) study that a significant determinant for an inadequate clinical response (occurrence of new vertebral or nonvertebral fragility fractures in patients prescribed ALN, risedronate, or raloxifene) was a less frequent coadministration of calcium and vitamin D supplements. However, in a second publication (6), they stated that inadequate clinical responders were comparable to patients who did not sustain clinical fractures with regard to calcium and vitamin D supplements. Díez-Pe´rez et al. (7), in a study designed to identify which patients are more likely to fracture while on treatment, found the risk was significantly greater in patients with 25OHD levels below 20 ng/ml. As for BMD changes, Antoniucci et al. (8) have found, in a post hoc analysis of the Fracture Intervention Trial study (FIT) study, that BMD response to therapy at total hip or spine did not vary with vitamin D status at baseline. It must be taken into account that in the FIT study, participants who had daily calcium intake of less than 1000 mg were provided with a supplement of 500 mg calcium and 250 IU cholecalciferol, and therefore, it could be argued that the lack of differences in BMD response could be due to this vitamin D administration. Nevertheless, 250 IU is a low dose not considered to be sufficient to successfully treat vitamin D deficiency. In contrast to these negative results, Ishijima et al. (9) found that patients on ALN treatment and with basal serum 25OHD concentration above 25 ng/ml show BMD changes significantly greater than those below (a difference that was not found when the discrimination point was set at 30 ng/ml). On the other hand, Carmel et al. (10), categorizing patients as BP nonresponders if they had a persistent T-score below ⫺3, a greater than 3% decrease in BMD, or an incident fracture on BP therapy, conclude that the serum 25OHD cutoff point for the best response to these drugs is 33 ng/ml. Regarding BTMs, Chung et al. (11) have recently compared the effect of risedronate plus cholecalciferol with that of risedronate alone on BTMs in 164 Korean patients with osteoporosis. They did not find significant differences in the CTX decrease after 16 wk of treat-

ment. However, no stratification for 25OHD baseline levels was performed, and therefore, a different pattern of response below or above a certain 25OHD cutoff point could not be put forward. Peris et al. (19), classifying patients as inadequate responders to anti-osteoporotic treatment on the basis of BMD loss and/or the presence of fragility fractures, conclude that 25OHD serum levels influence adequate response to bisphosphonate treatment. Concerning the BMTs more suitable for this type of studies, it is worth mentioning that P1NP results were negative in our study in clear contrast with those of CTX. We believe this is merely due to its 3-month duration. BPs typically lower indices of bone resorption and, because of the coupling between osteoclasts and osteoblasts, they also lower indices of bone formation by about 12 wk. Therefore, P1NP changes could be expected to be borderline with the 3-month period of our study that we did not want to extend for ethical reasons. The mechanistic explanation for the interaction between vitamin D and the antiresorptive effect of ALN is not entirely clear. Vitamin D deficiency is considered to result in an increase in bone turnover through its stimulatory effect on PTH secretion. Consequently, vitamin D administration in patients with vitamin D deficiency should bring about a decline in PTH and a fall in bone turnover. Accordingly, the synergistic effect of vitamin D on the antiresorptive effect of ALN in patients with vitamin deficiency could just reveal the coincidence of two independent phenomena: the antiresorptive effect of ALN and the antiresorptive effect that vitamin D has under these circumstances by means of decreasing PTH (20, 21). Barone et al. (22), assessing the BMD response to ALN alone or ALN plus calcitriol, have shown that subjects who normalized serum PTH had a greater increase in lumbar spine BMD. In any case, it must be taken into account that in patients on BPs, the relationship between vitamin D and PTH is complicated by the fact that BPs themselves raise serum PTH levels. In fact, in our patients, serum PTH increased in the ALN group and remained at the baseline level only in patients receiving ALN⫹VitD. Of course, this lack of rise in PTH may be one of the reasons for the greater CTX fall observed in the ALN⫹VitD group.

4496

Olmos et al.


On the other hand, PTH may not be the only explanation for the synergistic effect between vitamin D and BPs. Vitamin D no doubt acts on bone through changes in PTH secretion, but we cannot dismiss the likelihood that other mechanisms are also involved in the possible influence of hypovitaminosis D in the antiresorptive effect of BPs, as may be the case for osteoid mineralization, and perhaps for a direct effect of a vitamin D-active metabolite on bone cells. In this regard, it should be taken into account that BPs work through their binding to mineralized surfaces, and therefore, it could be thought that in situations of bone undermineralization they were unable to carry out their effect properly. In fact, our results suggest that a mechanism independent of PTH is plausible. Thus, after stratifying the patients by their final PTH levels, we have seen that among women with values above normal (⬎65 pg/ml), those who had been on ALN plus 25OHD3 tended to have a greater CTX decrease than those on ALN alone, despite the fact that they were in the same PTH stratum. Our paper has some weaknesses. For instance, because the main objective of osteoporosis treatment is fracture prevention, a study addressing this objective could have been preferable. However, apart from the very large number of patients needed for this, such an approach would have implied several years of followup, which would have been unethical. On the other hand, the aim of our study was to know whether vitamin D deficiency impairs the BP effect on bone, and this effect may be measured quite precisely by means of BTMs, particularly bone resorption markers. Second, a larger number of patients could have allowed us to perform further analysis grouping patients by small segments of 25OHD serum levels. Although that would have been of interest, this analysis would not have added to the main aim of our study, mentioned above. In conclusion, our results suggest that BPs may carry out their antiresorptive effect even in vitamin D deficiency states, although 25OHD3 administration would add some additional effect in deficient patients. This additional effect, however, proved to be only of a moderate magnitude as compared with that of BPs. We consider that a decrease in PTH secretion is a most plausible explanation, but more studies are needed to unravel whether other mechanisms are also involved in this vitamin D-BPs interaction.

Acknowledgments Address all correspondence and requests for reprints to: Jose´ M. Olmos, Department of Internal Medicine, Hospital Universita-


rio Marque´s de Valdecilla, Avenida Valdecilla, s/n-39008 Santander, Spain. E-mail: [email protected]. This work was supported by grants from the Instituto de Salud Carlos III-FIS, Spain (PI08/0183 and PI11/01092). Disclosure Summary: The authors have nothing to disclose.

References 1. Boonen S, Vanderschueren D, Haentjens P, Lips P 2006 Calcium and vitamin D in the prevention and treatment of osteoporosis: a clinical update. J Intern Med 259:539 –552 2. Kanis JA, Burlet N, Cooper C, Delmas PD, Reginster JY, Borgstrom F, Rizzoli R; European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO) 2008 European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 19:399 – 428 3. Dawson-Hughes BD, Lindsay R, Khosla S, Melton LJ III, Favus M, Baim S 2008 Clinicians’s guide to prevention and treatment of osteoporosis. Washington, DC: National Osteoporosis Foundation; www.nof.org/professionals/clinical-guidelines 4. Karaplis AC, Chouha F, Djandji M, Sampalis JS, Hanley DA 2011 Vitamin D status and response to daily 400 IU vitamin D3 and weekly alendronate 70 mg in men and women with osteoporosis. Ann Pharmacother 45:561–568 5. Adami S, Isaia G, Luisetto G, Minisola S, Sinigaglia L, Gentilella R, Agnusdei D, Iori N, Nuti R, ICARO Study Group 2006 Fracture incidence and characterization in patients on osteoporosis treatment: The ICARO study. J Bone Miner Res 21:1565–1570 6. Adami S, Isaia G, Luisetto G, Minisola S, Sinigaglia L, Silvestri S, Agnusdei D, Gentilella R, Nuti R; ICARO Study Group 2008 Osteoporosis treatment and fracture incidence: the ICARO longitudinal study. Osteoporos Int 19:1219 –1223 7. Díez-Pe´rez A, Olmos JM, Nogue´s X, Sosa M, Díaz-Curiel M, Pe´rezCastrilloń JL, Pe´rez-Cano R, Munõz-Torres M, Torrijos A, Jodar E, Del Rio L, Caeiro-Rey JR, Farrerons J, Vila J, Arnaud C, Gonza´lezMacías J 2012 Risk factors for prediction of inadequate response to antiresorptives. J Bone Miner Res 27:817– 824 8. Antoniucci DM, Vittinghoff E, Palermo L, Black DM, Sellmeyer DE 2009 Vitamin D insufficiency does not affect response of bone mineral density to alendronate. Osteoporos Int 20:1259 –1266 9. Ishijima M, Sakamoto Y, Yamanaka M, Tokita A, Kitahara K, Kaneko H, Kurosawa H 2009 Minimum required vitamin D level for optimal increase in bone mineral density with alendronate treatment in osteoporotic women. Calcif Tissue Int 85:398 – 404 10. Carmel AS, Sieh A, Bang H, Bockman RS 2012 The 25(OH)D level needed to maintain a favorable BP response is ⱖ33 ng/ml. Osteoporos Int 10.1007/s00198-011-1868-7 11. Chung HY, Chin SO, Kang MI, Moon SH, Yoon BK, Yoon HK, Chung YS, Park HH 2011 Efficacy of risedronate with cholecalciferol on 25-hydroxyvitamin D level and bone turnover in Korean patients with osteoporosis. Clin Endocrinol (Oxf) 74:699 –740 12. Kuchuk NO, van Schoor NM, Pluijm SM, Chines A, Lips P 2009 Vitamin D status, parathyroid function, bone turnover, and BMD in postmenopausal women with osteoporosis: Global perspective. J Bone Miner Res 24:693–701 13. Bishoff-Ferrari HA, Dawson-Hughes B, Stoklin E, Sidelnikov E, Willet WC, Edel JO, Stahelin HB, Wolfram S, Jetter A, Schwager J, Henschkowski J, Von Eckardstein A, Egli A 2012 Oral supplementation with 25(OH)D3 versus vitamin D3: effects on 25(OH)D levels, lower extremity function, blood pressure and markers of innate immunity. J Bone Miner Res 27:160 –169 14. Russo S, Carlucci L, Cipriani C, Ragno A, Piemonte S, Fiacco RD, Pepe J, Fassino V, Arima S, Romagnoli E, Minisola S 2011 Metabolic changes following 500 ␮g monthly administration of calcidiol: a study in normal females. Calcif Tissue Int 89:252–257


15. Berry EM, Gupta MM, Turner SJ, Burns RR 1973 Variations in plasma calcium with induced changes in plasma specific gravity, total protein, and albumin. Br Med J 4:640 – 643 16. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D 1999 A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of diet in renal disease study group. Ann Intern Med 130:461– 470 17. Looker AC, Orwoll ES, Johnston Jr CC, Lindsay RL, Wahner HW, Dunn WL, Calvo MS, Harris TB, Heyse SP 1997 Prevalence of low femoral bone density in old U.S. adults from NHANES III. J Bone Miner Res 12:1761–1768 18. Riancho JA, Valero C, Hernandez JL, Olmos JM, Paule B, Zarrabeitia A, Gonzalez-Macias J 2007 Biomechanical indices of the femoral neck estimated from the Standard DXA output: age- and sexrelated differences. J Clin Densitom 10:39 – 45 19. Peris P, Martínez-Ferrer A, Monegal A, Martínez de Osaba MJ,


4497

Muxi A, Guanãbens N 2012 25 hydroxyvitamin D serum levels influence adequate response to bisphosphonate treatment in postmenopausal osteoporosis. Bone 51:54 –58 20. Bischoff-Ferrari HA, Willett WC, Orav EJ, Oray EJ, Lips P, Meunier PJ, Lyons RA, Flicker L, Wark J, Jackson RD, Cauley JA, Meyer HE, Pfeifer M, Sanders KM, Sta¨helin HB, Theiler R, Dawson-Hughes B 2012 A pooled analysis of vitamin D dose requirements for fracture prevention. N Engl J Med 367:40 – 49 21. Holvik K, Madar AA, Meyer HE, Lofthus CM, Stene LC 2012 Changes in the vitamin D endocrine system and bone turnover after oral vitamin D3 supplementation in healthy adults: results of a randomised trial. BMC Endocr Disord 12:7 22. Barone A, Giusti A, Pioli G, Girasole G, Razzano M, Pizzonia M, Palummeri E, Bianchi G 2007 Secondary hyperparathyroidism due to hypovitaminosis D affects bone mineral density in response to alendronate in elderly women with osteoporosis: A randomized controlled trial. J Am Geriatr Soc 55:752–757