Common use of STATISTICAL PROCEDURES Clause in Contracts

STATISTICAL PROCEDURES. The purpose of this section is to outline prospectively the types of analyses and presentations of data that will answer the study objectives outlined in the protocol, and to explain how the data will be handled and analyzed, adhering to commonly accepted standards and practices of biostatistical analysis in the pharmaceutical industry. The primary objective of this study is to determine the safety and efficacy of BA058 80 µg when compared to a matching placebo (Placebo) for prevention of vertebral fracture in otherwise healthy ambulatory postmenopausal women at risk of fracture from severe osteoporosis. The secondary objectives of this study are to determine the safety and efficacy of BA058 80 µg when compared to Placebo for prevention of non-vertebral fractures and change in vertical height. Additional secondary efficacy outcomes include BMD (spine, hip and femoral neck) and safety (hypercalcemia) when compared to teriparatide in otherwise healthy ambulatory postmenopausal women at risk of fracture from severe osteoporosis. The specific objectives of this study are to: · Determine the comparative efficacy of 18 months of treatment with BA058 80 µg on reduction of vertebral fracture incidence in otherwise healthy ambulatory postmenopausal women at risk of fracture from severe osteoporosis when compared with Placebo. · Determine the comparative efficacy of 18 months of treatment with BA058 80 µg on lumbar spine, hip, and femoral neck bone mineral density (BMD) in otherwise healthy ambulatory postmenopausal women with severe osteoporosis when compared to teriparatide. · Determine the comparative efficacy of 18 months of treatment with BA058 80 µg on reduction of non-vertebral fracture incidence in otherwise healthy ambulatory postmenopausal women at risk of fracture from severe osteoporosis when compared with Placebo. · Determine the overall safety and tolerability of 18 months of treatment with BA058 80 µg, and specifically the number of patients with hypercalcemic events, in otherwise healthy postmenopausal women with severe osteoporosis when compared to teriparatide and Placebo. · Provide additional evidence of bone safety through histomorphometric assessment of bone biopsy samples in a subset of patients from the BA058 80 µg and Placebo groups. · Provide additional evidence of renal safety through radiological assessment by CT scan of a subset of patients from the BA058 80 µg, Placebo and teriparatide groups. A sample size of 622 patients per treatment arm provides 90% power at a two-sided alpha of 0.05 to detect a difference of 4% between treatments, assuming a vertebral fracture rate of 7% in placebo patients and 3% in BA058 80 µg for injection-treated patients when the large scale approximation of the binomial method is employed. This superiority assessment infers a relative risk reduction of 57% and presupposes the availability of a pretreatment and post-treatment radiological assessment. This population analysis would therefore be considered a modified ITT and will constitute the primary analysis population for this study. To ensure an analysis size of 622 patients, an overall sample size of 800 patients per treatment arm will be recruited, anticipating that approximately 20% of patients may not have a second evaluable X-ray film available for analysis. Should the projected fracture rate of 7% in placebo patients not be achieved, the sample size retains greater than 90% power at an alpha of 0.05 to detect a 4% difference between treatments based on placebo fracture incidence of 6% or 5%. For statistically-powered secondary endpoint assessments, the sample size will have more than 90% power (n=275) at a two-sided alpha of 0.05 to detect a 1.15 percent difference between BA058 and teriparatide for spinal BMD based on a superiority hypothesis. Similarly, for total analyzable hip BMD, the sample size will provide more than 90% power (n=25) at a two-sided alpha of 0.05 to detect a 2.45 percent difference between BA058 and teriparatide treatment effect and to detect a 2.00 percent difference between BA058 and teriparatide for femoral neck BMD (n=125) based on the same hypothesis. For differences in the number of patients in the BA058 and teriparatide treatment groups reporting one or more events of hypercalcemia, both above the upper limit of normal and at a value of 0.3 mg/mL above the upper limit of normal, the medically significant elevation, the study sample size will also provide more than 90% power to detect such a difference using a two-sided alpha of 0.05. Additional and other secondary endpoints will also be satisfied by these study sizes and will be included in the details provided in the Statistical Analysis Plan (SAP).