br Materials and methods br Study design and
2. Materials and methods
2.1. Study design and participants
We have previously described the study design and eligibility of the IBIS-II bone sub-study . Entry criteria were designed to include women aged 45–60 years who had a relative risk of breast cancer that was at least two times higher than in the general population, those aged 60–70 years who had a risk that was at least 1.5 times higher, and those aged 40–44 years who had a risk that was four times higher. The IBIS-II bone sub-study enrolled 1410 women and they were stratified into three groups according to the lowest baseline T-score at either femoral neck or lumbar spine (Fig. 1).
Women who were osteopenic (−2.5 ≤ T-score < −1.0) were en-tered into stratum II and were additionally randomised to receive oral risedronate (35 mg per week) or matching placebo (N = 500). For this analysis only women from stratum II and those who did not develop breast cancer or died were included. All women were advised, but not required, to take vitamin D and calcium supplements. Exclusion criteria for the bone sub-study included previous bilateral hip fractures, women with any type of metabolic bone disease, and women who have reg-ularly taken medication affecting bone N-octanoyl-L-Homoserine lactone within the past 12 months prior to study entry. Breast cancer development was the primary endpoint of the IBIS-II trial and women were excluded from the current analysis if they developed breast cancer at any point during the active treatment period. The trial was approved by the UK North West Multicentre Research Ethics Committee and was done in accordance with the Declaration of Helsinki, under the principles of good clinical practice. Participants provided written informed consent.
BMD was assessed by follow-up DXA scans at the lumbar spine and hip at 1, 3 and 5 years. BMD measurements at the femoral neck are more variable as it relies on a smaller region of the bone than total hip. Therefore, we report here BMD measurements for the total hip. A final
DXA scan at 7 years (two years after treatment cessation) was also performed, but these results are not reported here. T-scores were cal-culated using either the GE-Lunar  or Hologic  manufacturer's reference age-specific ranges for the lumbar spine (L1 to L4) and the NHANES III reference range for the femoral neck region . For quality assurance, all baseline and follow-up DXA scans were reviewed centrally by two clinical scientists with expertise in bone densitometry (GMB and RP). Women in stratum II of the bone sub-study were ran-domised to receive oral risedronate or matching placebo using ran-domly chosen blocks of size six, eight, or ten to maintain balance. Compliance in stratum II was determined full if women took their al-located risedronate at least 80% over 5 years of follow-up (at least 208 weeks over 5 years). Women who had a BMD loss of 6% or more at the 12 month's visit were furthermore required to have a safety DXA scan at 24 months of follow-up. Similarly, those with a BMD loss of over 10% at 36 months had a safety scan at 48 months, and those with a BMD loss of over 16% at the 60 month's visit, had an interval scan at 72 months. Further details on assessments in the bone sub-study can be found in our previous publication .
Blood samples were taking in the non-fasting state and the serum separated and stored at −70C. The total N-Terminal Propeptide of Type I Collagen (PINP) was measured using the Cobas e411 automated im-munoassay (Roche Diagnostics, Penzburg, Germany). The precision of PINP was assessed by measuring a serum quality control sample daily. The coefficient of variation was calculated and expressed as a percen-tage. The mean (SD) PINP concentration of duodenum QC was 47.7 (2.4) ng/ ml and the coefficient of variance was 5%.
2.3. Statistical analysis
The primary objective of this analysis was to compare the effect of risedronate versus placebo on the BMD change between baseline and 5 years at both the lumbar spine and total hip for women in stratum II who were randomised to anastrozole. Secondary objectives included comparison of baseline and randomised treatment effects on PINP measurements at study entry and 12 and 60 months of follow-up.
All results are expressed as percent mean BMD changes (with cor-responding 95% confidence intervals) at the lumbar spine and total hip between baseline and 5 years. BMD changes and differences between treatment groups were assessed using paired t-tests for normal dis-tribution. Pearson's correlation coefficient analysis was used to examine the relationship between and the values of BMD and PINP. Adverse
Baseline characteristics for osteopenic women according to main and risedronate randomisation.
Previous HRT use (%)
Never smokers (%)
PINP, median (IQR)
Data are median (interquartile range) or percentage (%). Abbreviations: P = Placebo, R = Risedroante, A = Anastrozole, IQR = interquartile range, BMI = body mass index, kg = kilogram, m = meter, HRT = hormone replacement therapy, PINP = N-Terminal Propeptide of Type I Collagen.