BMD response in the odanacatib phase 2 trial: Prespecified threshold and subgroup analyses

Background: Odanacatib (ODN) is a cathepsin K inhibitor in phase 3 development for osteoporosis. Treatment with ODN 50 mg once weekly for 2 years resulted in placebo (PBO)-subtracted mean increases in lumbar spine (LS) and total hip (TH) bone mineral density (BMD) of 5.7% and 4.1% respectively, P≤0.001 vs. placebo. Five years of continuous treatment with ODN 50 mg resulted in BMD increases from baseline of 11.9% at LS and 8.5% at TH, and sustained reductions in bone resorption markers. In contrast, levels of bone formation markers were near baseline. Objectives:To evaluate BMD response by thresholds and in pre-specified subgroups. Methods: 399 postmenopausal women with low baseline BMD were randomized to ODN 3, 10, 25, 50 mg or PBO weekly, plus vitamin D and calcium. Percent change from baseline in LS BMD was the primary endpoint. Prespecified BMD-change threshold analyses and subgroup analyses by age, ethnicity, and baseline BMD and biomarkers are reported for women who received weekly placebo or ODN 50 mg (the dose being investigated in the phase 3 trial). Results: At the LS, 0%, 95%, 76%, and 41% of participants treated with ODN 50 mg weekly for 2 years had BMD changes of ≤-3%, >0%, >3%, and >6%, respectively, compared with 24%, 48%, 18%, and 7% in the PBO group. At the TH, the proportions were 0%, 85%, 48%, and 19% in the ODN group and 33%, 41%, 8%, and 0% in the PBO group. Subgroup analysis of % changes from baseline in LS BMD after 2 years revealed no statistically significant treatment-subgroup interactions between women who at baseline were <65 vs. ≥65 years old, Caucasian vs. non-Caucasian, or above or below the median values of baseline LS, hip trochanter, or femoral neck BMD or baseline bone resorption markers CTX or DPD. However, there were greater treatment effects in LS BMD in women with lower baseline TH BMD, higher baseline bone resorption marker NTX, or higher baseline bone formation markers BSAP and P1NP (7.2%, 6.6%, 7.5%, 6.9%, respectively) compared with women with the opposite baseline characteristics (4.3%, 4.6%, 4.2%, 4.4%, respectively). TH BMD subgroup analyses showed similar trends of smaller magnitude. Conclusion s: Treatment with ODN 50 mg was considerably more likely than treatment with PBO to increase LS and TH BMD to each threshold level. Subgroup analysis of BMD change suggests that participants with lower BMD or higher bone turnover at baseline may have a greater % BMD increase with ODN treatment.