Clinical characteristics associated with bone mineral density improvement after 1-year alendronate/vitamin d3 or calcitriol treatment: Exploratory results from a phase 3, randomized, controlled trial on postmenopausal osteoporotic women in China

Abstract

Baseline and on-treatment characteristics, including age, obesity, calcium intake, and bone turnover markers, may predict the bone mineral density (BMD) response in women with postmenopausal osteoporosis (PMO) to 1 to 2 years of antiresorptive therapy and/or vitamin D supplementation. This study aimed to explore clinical characteristics associated with 12-month BMD improvement in Chinese women with postmenopausal osteoporosis (PMO).In this post hoc analysis of a previous phase 3 multicenter, randomized controlled trial, Chinese PMO women who were treated with once weekly alendronate 70 mg/vitamin D3 5600 IU (ALN/D5600) or once daily calcitriol 0.25 mcg, and had measurements of 1-year lumbar spine BMD (LS-BMD) and on-treatment bone turnover markers (BTMs) were included in the analysis.In Chinese PMO patients on ALN/D5600, 1-year LS-BMD change was negatively correlated with age (β = -0.00084, P < .01), dietary calcium (β = -0.0017, P = .07), and procollagen type 1 N-terminal propeptide (P1NP) change at month 6 (β = -0.000469, P = .0016), but positively with body mass index (BMI) (β = 0.00128, P = .08); baseline P1NP above the median was associated with a significantly greater BMD percentage change at the lumbar spine (P = .02) and the total hip (P = .0001). In the calcitriol group, a significant 1-year LS-BMD increase was associated with BMI (β = 0.0023, P = .02), baseline P1NP (β = 0.00035, P = .0067), history of prior vertebral fracture(s) (β = 0.034, P < .0001) and baseline serum 25(OH)D level (β = -0.00083, P = .02).The presented findings from Chinese postmenopausal osteoporotic women suggested clinically meaningful baseline and on-treatment characteristics predicting BMD improvement after 1 year of ALN/D5600 treatment, which differed from calcitriol treatment with baseline identifiable associations. The study remained exploratory and further accumulation of evidence is needed.

Figure 1

Correlation of 12-month mean BMD percentage change at the lumbar spine, femoral neck, and total hip with tertiles of BTMs at the baseline and 12 months in patients treated with 12-month ALN/D5600. A longitudinal data analysis with the unstructured covariance matrix was used to model the correlation among repeated measurements. The model includes the percent change in BMD from the baseline as a response variable and includes terms for time, baseline P1NP, or s-CTx stratum (≤median vs >median), and the interaction of time by baseline P1NP or s-CTx stratum (≤median vs >median). Error bars represent standard error. BMD = bone mineral density, BTM = bone turnover marker, P1NP = procollagen type 1 N-terminal propeptide, s-CTx = serum C-terminal telopeptide.

Figure 2

Correlation of 12-month mean BMD percentage change at the lumbar spine, femoral neck, and total hip with tertiles of plasma 25(OH)D at the baseline and 12 months and with absolute change in patients treated with 12-month ALN/D5600. A longitudinal data analysis with the unstructured covariance matrix was used to model the correlation among repeated measurements. The model includes percent change in BMD from the baseline as a response variable and includes terms for time, tertiles of baseline 25(OH)D, and the interaction of time by tertiles of baseline 25(OH)D. Error bars represent standard error. BMD = bone mineral density.