Roche trials database

Protocol number:

ML19357

Title of Study:

A randomized, double-blind study of the effect of oral monthly Bonviva on in vivo bone micro-architecture parameters in post-menopausal women with osteopenia

Sponsor:

Hoffmann-La Roche

Company division:

Pharmaceutical

Product name:

Bonviva/Boniva

Generic name:

ibandronate [Bonviva/Boniva]

Therapeutic area:

• Post-Menopausal Osteopenia

Clinical study summary:

This randomized, multi-centre, placebo-controlled study was designed to evaluate the effect of oral monthly Bonviva/Boniva (ibandronate) on bone microarchitecture parameters assessed by 3-dimensional peripheral quantitative computer tomography (3D-qQCT), bone mineral density and safety in post-menopausal patients with osteopenia. Patients were randomized to one of two treatment arms; Arm A: Bonviva/Boniva, Arm B: Placebo.

Study center(s):

4 centers in France

Phase of development:

III

Objectives:

The primary objective was to evaluate the effect of oral monthly ibandronate (Bonviva/Boniva) on bone microarchitecture parameters, assessed after one year by computer tomography (3D-qQCT).

The secondary objectives were as follows:

• Parallel evaluation of the effect on bone markers and bone mineral density,
• description of the patients’ population in terms of World Health Organization (WHO) index for osteoporotic fracture risk:
  • age, previous fractures,
  • lumbar and femoral T-score,
  • family history of osteoporosis,
  • low body mass index,
  • previous treatments inducing osteoporosis,
  • early menopause, smoking habits,

• evaluation of the correlation between WHO fracture risk index and bone microarchitecture parameters,
• safety profile.

Methodology:

This was a randomized, double-blind, placebo-controlled study. After screening phase of one month, patients received monthly doses of Bonviva/Boniva or corresponding placebo for 14 month. During the treatment period, bone mineral density, bone microarchitecture markers by computer tomography (3D-qQCT) adverse events, vital signs, laboratory and clinical parameters, and history of fractures were assessed. Adverse events were assessed up to 28 days after the last dose of study treatment.

Number of patients (planned/analyzed):

150 patients planned; 149 patients analyzed

Diagnosis and main criteria for inclusion:

Post-menopausal women, aged 55-75 years diagnosed osteopenia.

Test product, dose and mode of administration or test procedure:

Ibandronate 150 mg once monthly

Duration of treatment:

24 months

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

Placebo to ibandronate once monthly

Criteria for evaluation (efficacy, safety):

Efficacy

Primary: Trabecular bone volume fraction (BV/TV) at distal radius of non-dominant arm after one year of treatment.

Secondary: BV/TV at distal radius of non-dominant arm after six months and two years of treatment; description of the patients’ population characteristics according to WHO osteoporotic fracture risk index.

After 6 months, one year and two years of treatment:

• Total bone density,
• trabecular and cortical density,
• trabecular number,
• trabecular thickness,
• trabecular separation,
• trabecular distribution,
• cortical thickness,
• section area of distal radius and distal tibia;
• BV/TV at distal tibia; correlation between WHO fracture risk index and bone microarchitecture parameters;
• correlation between lumbar BMD evolution, bone markers, total bone density, trabecular density and cortical density at each measurement site (distal radius and distal tibia)
• serum CTX and P1NP at 3, 6, 12, 18 and 24 months
• BMD: lumbar spine (mean on L1 - L4), hip and wrist at one and two years.

Safety

Adverse events, laboratory test results, vital signs

Statistical methods:

The type I error was set at 5% two-sided for all analyses.

The value at one year of BV/TV at the distal radius was analyzed using an ANCOVA model, including the value at one year of BV/TV at the distal radius as the response variable, the treatment group, the center, the interaction treatment*center and the baseline value (screening) of BV/TV at the distal radius as covariates. The interaction treatment*center being not significant, it was excluded from the final model.

The BMD at each site (lumbar, femoral neck, total femur, wrist) was analyzed at M12 and M24 using an ANCOVA with treatment, center, interaction treatment*center and the baseline value as covariates.

All the microarchitecture parameters for distal radius and for distal tibia were analyzed at M6, M12 and M24 using an ANCOVA with factor treatment, center, interaction treatment*center and baseline value as covariates. If the interaction treatment*center was significant, an analysis by center was performed.

Bone markers (sCTX and P1NP) were analyzed at M3, M6, M12, M18 and M24 using an ANCOVA with baseline value as covariate.

The correlation between the evolution from baseline of the BMD of each site and the evolution of microarchitecture parameters and bone markers at M12 and M24 was performed using the Pearson correlation coefficients.

The correlation between the evolution from baseline of bone markers and the evolution of microarchitecture parameters at M6, M12 and M24 was performed using the Pearson correlation coefficients.

The correlation between bone markers at inclusion and the evolution from baseline of microarchitecture parameters at M6, M12 and M24 was performed using the Pearson correlation coefficients.

The correlation between WHO fracture risk index and bone microarchitecture parameters at each measurement site (distal radius and distal tibia) was performed at inclusion using a linear regression.

The numbers of patients with AEs during the treatment period were compared between groups using the Fisher’s exact test.

Summary (efficacy, safety, other results):

Efficacy

A total of 149 patients were included and randomized: 72 in the Bonviva/Boniva group and 77 in the placebo group. As one patient in the placebo group did not take any treatment, the safety population, as well as the ITT population, included 148 patients: 72 receiving Bonviva/Boniva and 76 receiving placebo. The PP population included 134 patients: 66 in the Bonviva/Boniva group and 68 in the placebo group. A total of 18 patients withdrew prematurely: 10 in the placebo group and 8 in the Bonviva/Boniva group, mostly because of consent withdrawal or occurrence of adverse events.

Demographic and baseline characteristics were similar in both treatment groups. Age ranged from 54 to 75 years old, with a median of 62 years old and mean BMI was 25 (±4) kg/m² for the total Intent-to-treat (ITT) population. Fourteen patients (9.5%) had a previous history of osteoporotic fracture, more in the Bonviva/Boniva than in the placebo group: 12.5% compared to 6.6%. One patient out of four had at least one risk factor of osteoporosis. The most frequent risk factors were history of proximal femur fracture in first degree relative (11%), history of fracture in minor traumatism (7%) and low BMI (7%).

All patients had bone densitometry measured by DXA at screening visit. Results were similar in both groups. Mean T-score was -1.4 (±0.8) for lumbar spine, -1.5 (±0.6) for femoral neck, -1.0 (±0.7) for total femur and -1.0 (±1.4) for wrist (distal radius).

All patients had computer tomography (3D-qQCT) performed at screening. Comparable results for volumetric BMD, trabecular and cortical parameters were obtained in both groups. Mean value of markers of bone resorption was similar in both groups at baseline with a mean sCTX of 0.58 (±0.21) ng/ml and a mean P1NP of 52 (±17) μg/L.

In the ITT population, after one year of treatment, there was no statistically significant difference in BV/TV at the distal radius between women receiving Bonviva/Boniva and women receiving the placebo. Adjusted mean values [95%CI] were 0.108 [0.106; 0.111] and 0.107 [0.105; 0.109] respectively (p=0.254). Mean change from screening visit was close to 0. Similar results were obtained in the PP population: no significant difference was observed between the two groups (p=0.477).

The sensitivity analysis, performed on patients of the ITT population who had all results (no handling of missing data), led to the same results. There was no difference between Bonviva/Boniva and placebo group and no change from baseline.

After one year of treatment, mean lumbar and hip BMD were significantly higher in the group receiving Bonviva/Boniva compared to the group on placebo (p<0.001) : adjusted mean BMD was 0.88 g/cm² at lumbar spine, 0.68 g/cm² at femoral neck, 0.81 g/cm² for total femur in the Bonviva/Boniva group compared to 0.85 g/cm², 0.67 g/cm² and 0.81 g/cm² respectively in the placebo group.

In the placebo group mean BMD values at M12 remained similar to values at baseline while mean BMD increase from baseline in the Bonviva/Boniva group was 0.031 g/cm² at lumbar spine, 0.014 g/cm² at femoral hip and 0.016 g/cm² for total femur. These results were confirmed after two years of treatment.

At the radius site, there was no difference in mean BMD between the two groups after one year of treatment (adjusted mean BMD of 0.51 g/cm² in both groups), but mean BMD was significantly higher in the group receiving Bonviva/Boniva compared to the group receiving the placebo after two years of treatment (p=0.009), with no change from baseline in mean BMD in the Bonviva/Boniva group and a decrease from baseline in the placebo group. Similar results were obtained in PP population.

In the distal radius, there was no significant difference in trabecular and cortical microarchitecture parameters after 12 months as well as after 24 months of treatment between patients who received Bonviva/Boniva and patients who received placebo, except for trabecular area with a higher mean value in the Bonviva/Boniva group (p=0.005). Still this result must be analysed with caution as the interaction treatment*center was significant and the analysis by center showed no significant difference between the two groups in the two centers Lyon and Paris, which included most patients.

On the contrary, statistically significant differences in bone microarchitecture parameters were observed in the distal tibia. Cortical density, cortical thickness and cortical area were significantly greater at all time points in the Bonviva/Boniva group compared to placebo group. Such differences were not observed on trabecular parameters in the distal tibia except for trabecular area which was higher in the placebo group at all time points. Furthermore, the total bone density was significantly higher in the Bonviva/Boniva group compared to the placebo group at all time points.

An analysis of correlation was performed using the Pearson correlation coefficients. The correlation coefficients were low and no correlation was shown between the evolution of BMD at each site and the evolution of total bone density, trabecular density and cortical density in distal radius and distal tibia after 12 months of treatment as well as after 24 months of treatment.

Markers of bone resorption sCTX and P1NP were measured at all visits. There was a statistically significant difference between the two groups at all time points with a marked decrease in these two markers in the group receiving Bonviva/Boniva while mean value remained similar to baseline value in the group receiving the placebo.

At M12, adjusted mean [95%CI] sCTX was 0.25 ng/ml [0.20; 0.31] in the Bonviva/Boniva group compared to 0.67 ng/ml [0.62; 0.72] in the placebo group and adjusted mean [95%CI] P1NP was 22.88 μg/L [19.84; 25.91] in the Bonviva/Boniva group compared to 49.64 μg/L [46.69; 52.59] in the placebo group.

An analysis of correlation was performed using the Pearson correlation coefficients to evaluate a potential correlation between the evolution of bone markers and the evolution of bone microarchitecture parameters.

The correlation coefficients were low and no correlation was shown between the evolution of bone microarchitecture parameters and evolution of bone markers in distal radius and distal tibia after 12 months of treatment as well as after 24 months of treatment.

Another correlation analysis was performed between values of bone markers at inclusion and evolution of bone microarchitecture parameters at M6, M12 and M24 for distal radius and distal tibia. Once again, Pearson’s correlation coefficients were low and no correlation was highlighted. Correlation analysis between the evolution of BMD at each site and evolution of bone markers showed correlation coefficients no higher than 0.34 in absolute value.

The mean WHO Fracture Risk Assessment Tool (FRAX index) was 5.2% for 10-year probability of major osteoporotic fracture in both groups at screening. The mean FRAX index for 10-year probability of hip fracture was 1.1% in the Bonviva/Boniva group and 1.2% in the placebo group.

A correlation was sought between fracture risk and bone microarchitecture parameters in distal radius and distal tibia using linear regression. No correlation was highlighted.

Safety

From the first treatment intake to the last intake + 28 days, 136 patients (92%) reported at least one treatment emergent adverse event (TEAE; adverse events that started between the first treatment intake and 28 days after the last treatment intake) for a total of 511 TEAEs, numbers being comparable in the two treatment groups.

The most frequent TEAEs were back pain (16.2% of the safety population), arthralgia (12.2%), and bronchitis (14.9%). More cases of fractures were reported in the placebo group compared to the Bonviva/Boniva group (8 vs. 2) and falls were reported only in the placebo group.

For 12 patients (8 in the Bonviva/Boniva group and 4 in the placebo group), TEAEs were considered related to study treatment. Most TEAEs considered related to Bonviva/Boniva were expected AEs with the use of bisphosphonates. They were completely resolved except one case of severe rheumatoid arthritis that was persisting.

Ten patients (6.8%) had to discontinue the treatment because of an adverse event: four in the Bonviva/Boniva group (5.6%) and six in the placebo group (7.9%). In the placebo group, most treatment discontinuations were due to occurrence of fractures. One of the TEAEs leading to treatment discontinuation was considered related to treatment: rheumatoid arthritis and one had an unknown relationship to treatment: rash erythematous, both in the Bonviva/Boniva group.

A total of 28 patients (18.9%) experienced at least one emergent adverse event that was considered serious: 15 in the Bonviva/Boniva group (20.8%) and 13 in the placebo group (17.1%). Only one of them was considered related to study treatment: visual acuity reduced in the placebo group, and for another one: chest pain in the Bonviva/Boniva group, the relationship to treatment was unknown. One death was reported during the study, in the Bonviva/Boniva group, due to road traffic accident and not related to the study treatment.

Weight was similar in both groups and mean change from baseline was less than 1 kg at each time point. All clinical examinations were normal at screening visit. At M3 and M6 five patients had abnormal clinical examination. This number increased two fold at M12 and M18 and after two years, more than one patient out of four had an abnormal clinical examination, similarly in both treatment groups.

Conclusions:

This placebo-controlled study had not demonstrated an effect of one year treatment with oral monthly Bonviva/Boniva on bone microarchitecture parameters measured by 3D-pQCT at the distal radius in postmenopausal osteopenic women. In the distal tibia, no effect was shown on trabecular parameters but a significant effect was observed on cortical density, cortical thickness and cortical area after one year of treatment.

Areal BMD measured by DXA was significantly improved with Bonviva/Boniva and volumetric BMD measured by 3D-pQCT was also improved in the distal tibia as soon as after 6 months of treatment with Bonviva/Boniva and in the distal radius after 2 years of treatment.

Markers of bone resorption sCTX and P1NP were significantly decreased with Bonviva/Boniva.

Once monthly 150 mg Bonviva/Boniva was well tolerated for two years.

Date of report:

03.03.2011

Click here for the protocol registry listing of this trial.