Background/Purpose: While bisphosphonates (BP) have been well studied in long-term trials of up to 4 years’ duration, relatively less is known about the immediate consequences of continuing vs. interrupting long-term treatment. This report describes changes in bone turnover and BMD in a 1-year trial of the calcium-sensing receptor antagonist MK-5442 in postmenopausal women who, after taking BP for ≥3 years, were randomized to continued alendronate (ALN) 70 mg weekly, switch to placebo (PBO), or switch to MK-5442. Primary results for MK-5442 are presented separately.

Methods: 526 postmenopausal women who had taken ALN for ≥12 months preceding the trial and an oral BP for ≥3 of the 4 years before the trial, with spine or hip BMD T-scores ≤-2.5 (or ≤-1.5 with ≥1 prior fragility fracture) and ≥-4.0, were recruited into a dose-finding study of MK-5442. Statistical tests of within-group changes and comparison between the PBO and ALN groups were performed post-hoc.

Results: At baseline, women switched from ALN to PBO (n=88) or continued on ALN 70 mg weekly (n=87) were of mean age 67 years and had mean T-scores at lumbar spine of -2.5 and total hip of -1.6,  and mean baseline urine NTX/Cr= 26.6 nmolBCE/mmolCr and serum P1NP= 26.0 ng/mL. Median length of previous BP use was 5.2 years. After 1 month, women switched from ALN to PBO experienced increases from baseline in urine NTX/Cr (28.4% vs. continued ALN, p<0.0001), while serum P1NP was unchanged. Both NTX/Cr and P1NP increased by 3 months (33.7% and 37.8%, respectively vs. ALN, both p<0.0001). After 12 months of PBO, least squares mean concentrations of NTX/Cr and P1NP rose to 42.2 nmolBCE/mmolCr and 40.1 ng/mL (both p<0.0001, Table). The markers were unchanged from baseline with continued ALN. After 12 months, the women who continued ALN had an increase in lumbar spine BMD while those switched to PBO experienced no change; total hip BMD did not change in those remaining on ALN but was reduced in women switched to PBO. BMD at both sites was significantly lower in women who switched to PBO vs. those who stayed on ALN (Table).

Conclusion: Discontinuation of alendronate after a median of 5 years resulted in an increase in NTX/Cr as early as 1 month and P1NP by 3 months. After 1 year, both bone turnover markers returned to levels similar to those expected in untreated postmenopausal women. These increases were accompanied by significantly lower spine and total hip BMD vs. continued treatment with ALN.