Clinical Trial Result Information

Protocol number
MF 4411

Title of Study
Multicenter, double-blind, placebo-controlled, randomized study on the efficacy and safety of ibandronate over 3 years in patients with postmenopausal osteoporosis using a continuous oral and an intermittent oral dosing regimen.

Sponsor
Roche Global Development

Company division
Pharmaceutical

Product name
Bonviva/Boniva

Generic name
ibandronate

Therapeutic area
Postmenopausal Osteoporosis

Clinical study summary
This was a 3-year, multicenter, double-blind, randomized, placebo-controlled study with 3 parallel treatment groups. Patients were randomized to one of the following 3 treatment groups:

1. Bonviva 2.5 mg po daily
2. Bonviva 20 mg po intermittent
3. Placebo po

Patients received the study medication for a total of 3 years.

Study center(s)
73 centers in Canada, Denmark, France, Germany, the Netherlands, Norway, Poland, Russia, the United Kingdom, and the United States.

Phase of development
III

Objectives
To investigate the long-term efficacy and safety of continuous and intermittent oral administration of Bonviva in the long-term treatment of postmenopausal osteoporosis, using the incidence of new vertebral fractures as the primary efficacy variable.

Methodology
Patients attended regular visits at 3 month intervals during the 3 year study. Lateral radiographs of the thoraco-lumbar spine were taken annually for the assessment of vertebral fractures. Evaluation of the X-rays was performed by 2 central reading facilities. BMD measurements of the lumbar spine, the hip, and of the distal forearm (for a subset of patients) were carried out every 6 months during the first 2 years and then again at the final visit. Height and bone pain were recorded at each visit. Laboratory assessments were done at 3 month intervals during the first half year, every 6 months until the end of the second year and then once more at the final visit. Safety information regarding adverse events and concomitant medications were recorded continuously throughout the study.

Number of patients (planned/analyzed)
A total of 2946 patients were randomized.

Diagnosis and main criteria for inclusion
Women with postmenopausal osteoporosis aged 55-80 years, with time since menopause ≥5 years. Bone-mineral density T-score in at least one vertebra of the lumbar spine (L₁–L₄) between –2.0 and 5.0, and with 1-4 prevalent vertebral fractures (T₄–L₄).

Test product, dose and mode of administration or test procedure
Patients in the 2.5 mg group received daily doses of oral 2.5 mg Bonviva (ibandronate). Patients receiving intermittent treatment were administered 12 oral doses of 20 mg Bonviva (ibandronate) every other day at the start of each 3 month cycle. On days without active study medication they were given an identically looking placebo. All patients received oral doses of calcium (500 mg/day) and vitamin D3 (400 IU/day) as supplements, throughout the 3 year study period.

Duration of treatment
3 years.

Reference therapy, dose and mode of administration or reference procedure
Placebo

Criteria for evaluation (efficacy, safety)
Efficacy: The primary efficacy parameter was the incidence of new vertebral fractures, defined as the proportion of patients with at least one new fracture. Secondary variables included: general incidence of clinical fractures, bone mineral density (BMD) of the lumbar spine, proximal femur, and distal forearm; height, pain and disability; urinary calcium, CTx and NTx; serum osteocalcin, bone-specific alkaline phosphatase, and parathyroid hormone concentrations.
Safety: Incidence of clinical adverse events, concomitant medications, clinical laboratory assessments (hematology and blood chemistry), laboratory abnormalities.

Statistical methods
All efficacy parameters were summarized for both the ITT and PP populations. The primary analysis group for all fracture analyses was the ITT population; for BMD and biochemical markers of bone turnover, the PP population was primary. Analysis of the primary efficacy variable was performed using the life-table survival method. A confirmatory analysis of the primary endpoint was done by testing the homogeneity of the time-to-event curves. The reduction in relative risk was estimated by using a proportional hazards model considering an interaction between treatment and ‘high’ or ‘low’ lumbar spine BMD at baseline. Analyses of time-to-event for clinical fractures, used Kaplan Meier survival analysis to evaluate pairwise differences between treatment groups. An ANOVA was used to investigate the treatment effect on the relative 3-year BMD changes, adjusting for baseline BMD as a covariate. Kruskal-Wallis tests were used for the comparison of treatment groups, based on 3-year data, for relative changes in markers of bone turnover. Differences in the reduction of height were assessed by a Wilcoxon test.

Summary (efficacy, safety, other results)
Efficacy: In comparison to placebo, after 3 years of treatment Bonviva highly significantly reduced the risk of a new incident vertebral fracture by 61.6% for the 2.5 mg daily and by 49.9% for the 20 mg intermittent group, respectively. The p-values for the treatment effect (estimated by a proportional hazards model without an interaction term for the BMD status at baseline) were p = 0.0003 (2.5 mg) and p = 0.0005 (20 mg), respectively. The reduction was robust and was consistently observed in a variety of subgroups.

Significant reductions of new vertebral fractures were also observed in both treatment groups after 2 years of treatment. A strong reduction was already evident in the 2.5 mg group after the first year (reduction of 57.9%; p = 0.0561).
Significant reductions were observed with both regimens in the incidence of new or worsening vertebral fractures and in the incidence of clinical vertebral fractures.
There was a reduced incidence of osteoporotic clinical fractures in both Bonviva treatment groups but these reductions were not significant. In a high risk group (with a baseline femoral neck T score < –3) the reduction in osteoporotic clinical fractures reached statistical significance in both treatment groups.
Bonviva treatment significantly reduced height loss in comparison to placebo in both active treatment groups (2.5 mg: p < 0.0001; 20 mg: p = 0.0115). Bonviva treatment led to an increase in bone mineral density that was highly significant in both active treatment groups. In comparison to placebo, lumbar spine BMD(L₂-L₄) increased over 3 years in the Per-Protocol population by 5.3% (2.5 mg, p < 0.0001) or 4.4% (20 mg p < 0.0001), while total hip BMD increased by 4.0% (2.5 mg, p < 0.0001) or 3.6% (20 mg p < 0.0001), respectively. Bonviva treatment rapidly led to a sustained decrease in the urinary or serum concentrations of 4 biochemical markers of bone turnover, reaching a nadir at 3 6 months. Levels of CTX, NTX, osteocalcin, and BSAP were all significantly suppressed in both active treatment groups in comparison to placebo (p < 0.0001 in all cases). There were no measurable effects of Bonviva therapy on bone pain, physical activity, and movement disability.

Safety: Bonviva therapy was well tolerated and there were very similar incidences (18.5% – 18.8%) of AEs leading to withdrawal in the placebo and the 2 active treatment groups. The overall incidence of adverse events was similar between all treatment groups. Most adverse events recorded during the study (> 90%) were considered unrelated to the study medication. A few AEs considered related to study medication occurred somewhat more frequently in the Bonviva groups than in the placebo group. These were dyspepsia, diarrhea (both Bonviva groups), myalgia, rash (2.5 mg only), and nausea (20 mg only). No clinically relevant changes in laboratory parameters were observed. The distribution of marked laboratory abnormalities was balanced between treatment groups. Bonviva therapy reduced the rate of bone turnover without any impairment of bone mineralization.

Conclusions
This study met its primary objective and was able to demonstrate both for continuous and for intermittent Bonviva administration a significant reduction in the rate of patients with new incident vertebral fractures at 3 years of treatment (2.5 mg: p = 0.0003; 20 mg: p = 0.0005). The magnitude of relative risk reduction in comparison to placebo was estimated to be 61.6% for the 2.5 mg daily and 49.9% for the 20 mg intermittent group, respectively. Significant reductions were also observed in the incidence of new or worsening vertebral fractures and in the incidence of clinical vertebral fractures. There was a significant reduction in height loss in both Bonviva groups. Highly significant treatment benefits (p < 0.0001) were observed for both Bonviva groups for lumbar spine and total hip BMD and for 4 biochemical markers of bone turnover. Both Bonviva treatment regimens were well tolerated. Bonviva therapy was safe and did not lead to major imbalances in the distribution of adverse events or laboratory abnormalities between Bonviva and placebo groups.

Date of report
5/1/2002