Predictors of discontinuation of osteoporosis treatment: sub-analysis of the Japanese osteoporosis intervention trial-05 (JOINT-05)

Introduction To identify predictors of discontinuing treatment with teriparatide (TPTD) and alendronate (ALN), data from a randomized, controlled trial (JOINT-05) involving postmenopausal Japanese women at high risk of fracture were re-analyzed. Materials and Methods Participants received sequential therapy with once-weekly TPTD for 72 weeks followed by ALN for 48 weeks (TPTD-ALN group) or monotherapy with ALN for 120 weeks (ALN group). Background data including comorbidities, fracture prevalence, cognitive function, quality of life, activities of daily living, bone metabolism parameters, and nutrient intake were collected. The endpoints were 3 types of discontinuations by the reason: a poor compliance, adverse events (AEs), or any reason including those unrelated to AEs or poor compliance. Odds ratios (ORs) of baseline predictors of discontinuation were evaluated by single or multiple regression analysis. Results A total of 234 (49.0%) patients in the TPTD-ALN group and 167 (34.2%) patients in the ALN group discontinued. In the TPTD-ALN group, a lower serum calcium level was a significant predictor of compliance-related discontinuation. Serum 25-hydroxyvitamin D levels were lower in patients with lower serum calcium levels than with higher serum calcium levels. In the ALN group, poor cognitive function was significantly associated with compliance-related discontinuation, and higher body mass index and alcohol intake were predictors of AE-related discontinuation. Predictors of discontinuation were drug-specific. Lower serum calcium levels and poor cognitive function were predictors of discontinuing once-weekly TPTD and ALN, respectively. Conclusion When starting TPTD and ALN treatment, careful attention to patients with lower serum calcium levels and poor cognitive function, respectively, may be needed for better treatment continuity.