A Study of MK-0822 in Postmenopausal Women With Osteoporosis to Assess Fracture Risk (MK-0822-018)

An adverse event is defined as any untoward medical occurrence in a person or clinical investigation participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. The incidence rate of participants with adverse events (number of participants with an event per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

An adverse event is defined as any untoward medical occurrence in a person or clinical investigation participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. The incidence rate of participant discontinuations from treatment due to adverse events (number of participants with an event per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

BMD was measured by dual-energy x-ray absorptiometry (DXA) at the total hip starting at screening, and at yearly intervals until the end of the study (second extension study) for all participants who entered the second extension study. Least squares (LS) means percent change in BMD from original baseline are provided through Month 108 (Year 9). At Months 96 (Year 8) and 108 (Year 9), approximately 3% or fewer participants in each treatment group had BMD data, and results at those time points should be viewed with caution. NOTE: The mean percent change in BMD from baseline in participants originally randomized to placebo includes BMD results obtained after those participants were switched to open-label odanacatib, which occurred at different times relative to their start of blinded study medication in the base study.

Time frame: Baseline and once yearly, up to approximately 108 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (total hip BMD) was used for analysis (base study + first extension-dbp + second extension).

Morphometric vertebral fractures were confirmed by central adjudication on lumbar and thoracic spinal radiographs acquired at baseline and at Months 6, 12, and subsequent yearly time points. Incident fractures (i.e., occurring after baseline) in previously normal vertebral bodies and/or those occurring after baseline in previously fractured vertebral bodies (Genant SQ Grade 1-3) (T4 to L4) were confirmed by QM and either SQ or binary SQ (yes/no) methodologies. A time-to-event methodology was used to evaluate results: life-table estimates of the incidence proportion (cumulative incidence) of participants with at least one morphometric vertebral fracture were assessed at Months 6, 12, and subsequent yearly intervals; an interval-censored approach was then used to determine incidence rate (number of participants with a morphometric fracture per 100 person-years of follow-up).

Time frame: Up to approximately 74 months of observation

Population: The FAS including all randomized participants, who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study + first extension-double blind period \[dbp\]) with at least one spine radiograph on treatment (and at baseline) was used for analysis.

Osteoporotic clinical hip fractures was confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Hip fractures were fractures of the proximal femur confirmed as being located in the hip (i.e., sub-region not specified; cervical, and intertrochanteric). Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 74 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study + first extension-dbp) was used for analysis.

Osteoporotic non-vertebral clinical fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Non-vertebral fractures were assessed across multiple anatomical sites including clavicle, distal femur and shaft, fibula, hip, humerus, pelvis, radius, ribs, sacrum, tibia, ulna, and wrist. Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 74 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study + first extension-dbp) was used for analysis.

Morphometric vertebral fractures were assessed by central adjudication on lumbar and thoracic spinal radiographs acquired at baseline and at Months 6, 12, and subsequent yearly time points. Incident fractures (i.e., occurring after baseline) in previously normal vertebral bodies and/or those occurring after baseline in previously fractured vertebral bodies (Genant semiquantitative \[SQ\] Grade 1-3) (T4 to L4) were confirmed by quantitative morphometric (QM) and either SQ or binary SQ (yes/no) methodologies. A time-to-event methodology was used to evaluate results: life-table estimates of the incidence proportion (cumulative incidence) of participants with at least one morphometric vertebral fracture were assessed at Months 6, 12, and subsequent yearly intervals; an interval-censored approach was then used to determine incidence rate (number of participants with a morphometric fracture per 100 person-years of follow-up).

Time frame: Up to approximately 60 months of observation

Population: The Full-Analysis-Set (FAS) including all randomized participants, who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study) with at least one spine radiograph on treatment (and at baseline) was used for analysis.

Osteoporotic clinical hip fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Hip fractures were fractures of the proximal femur confirmed as being located in the hip (i.e., sub-region not specified; cervical, and intertrochanteric). Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 60 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study) was used for analysis.

Osteoporotic non-vertebral clinical fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Non-vertebral fractures were assessed across multiple anatomical sites including clavicle, distal femur or shaft, fibula, hip, humerus, pelvis, radius, ribs, sacrum, tibia, ulna, and wrist. Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 60 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study) was used for analysis.

Compartment-specific effects of osteoporosis were assessed by measuring trabecular vBMD at the lumbar spine (L1 total vertebral body) using quantitative computed tomography. The percent change from baseline at Month 60 (base study + extension study) was then assessed using a longitudinal data analysis model including terms for treatment, stratum (prior vertebral fracture \[yes/no\]) and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 60

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (lumbar spine vBMD) was used for analysis (PN032-Base + Extension).

Compartment-specific effects of osteoporosis were assessed by measuring trabecular vBMD at the lumbar spine (L1 total vertebral body) using quantitative computed tomography. The percent change from baseline at Month 24 (base study) was then assessed using a longitudinal data analysis model including terms for treatment, stratum (prior vertebral fracture \[yes/no\]) and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (lumbar spine vBMD) was used for analysis (PN032-Base).

Sarcopenia is the age-related loss of skeletal muscle mass and associated loss of strength. Progression of sarcopenia was assessed using aLBM as measured by total body DXA. The change from baseline at yearly intervals was then assessed using a longitudinal data analysis model with terms for treatment, stratum (sarcopenia, non-sarcopenia), time and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline and once yearly up to 4 years

Population: The FAS for change from baseline at specific time-points including all randomized participants who took at least one dose of study medication and had the necessary baseline and on-treatment measurements available (aLBM) was used for analysis (PN035).

Gait speed is a component of the Short Physical Performance Battery (SPPB) Score. Participants are asked to walk a distance of 4 meters at their normal pace. The test is performed 2 times, and the walk done in the shortest time is used for scoring. The activity is timed and then reduced to a categorical 0 to 4 scale based on time achieved. Higher scores indicate an improved function level. The change from baseline at yearly intervals was then assessed using a longitudinal data analysis model with terms for treatment, stratum (sarcopenia, non-sarcopenia), time and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline and once yearly up to 4 years.

Population: The FAS for change from baseline at specific time-points including all randomized participants who took at least one dose of study medication and had the necessary baseline and on-treatment measurements available (gait speed) was used for analysis (PN035).

The Short Physical Performance Battery (SPPB) Score is used to assess physical function in older persons. The SPPB consists of 3 types of physical activities: standing balance, gait speed, and chair rise. Component activities are timed and then reduced to a categorical 0 to 4 scale based on time achieved. A higher composite score (range 0 to 12) indicates an improved function level. The change from baseline at yearly intervals was then assessed using a longitudinal data analysis model with terms for treatment, stratum (sarcopenia, non-sarcopenia), time and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline and once yearly up to 4 years

Population: The FAS for change from baseline at specific time-points including all randomized participants who took at least one dose of study medication and had the necessary baseline and on-treatment measurements available (SPPB) was used for analysis (PN035).

An adverse event is defined as any untoward medical occurrence in a person or clinical investigation participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment.

Time frame: Up to approximately 34 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (second extension study).

An adverse event is defined as any untoward medical occurrence in a person or clinical investigation participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment.

Time frame: Up to approximately 34 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for the analysis (second extension study).

BMD was measured by DXA at the distal one-third radius at randomization and at yearly intervals until the end of the study (base study + first extension-dbp) in an approximate 10% random subset of participants at selected sites.

Time frame: Baseline and once yearly, up to approximately 74 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (distal one-third radius BMD) was used for analysis (base study + first extension-dbp).

BMD was measured by DXA for all participants at the femoral neck-hip at screening, Months 6, 12, and subsequent yearly intervals until the end of the study (base study + first extension-dbp).

Time frame: Baseline, Month 6, and once yearly, up to approximately 74 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (femoral neck BMD) was used for analysis (base study + first extension-dbp).

BMD was measured by DXA for all participants at the lumbar spine at randomization, Months 6, 12, and subsequent yearly intervals until the end of the study (base study + first extension-dbp). Measurements were made on at least 3 vertebrae for all time points; vertebrae with fractures were excluded from analyses.

Time frame: Baseline, Month 6, and once yearly, up to approximately 74 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (lumbar spine BMD) was used for analysis (base study + first extension-dbp).

BMD was measured by DXA for all participants at the total hip at screening, Months 6, 12, and subsequent yearly intervals until the end of the study (base study + first extension-dbp).

Time frame: Baseline, Month 6, and once yearly, up to approximately 74 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (total hip BMD) was used for analysis (base study + first extension-dbp).

BMD was measured by DXA for all participants at the trochanter-hip at screening, Months 6, 12, and subsequent yearly intervals until the end of the study (base study + first extension-dbp).

Time frame: Baseline, Month 6, and once yearly, up to approximately 74 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (trochanter BMD) was used for analysis (base study + first extension-dbp).

Height was measured by wall-mounted stadiometer at randomization and at yearly intervals across the base study and the two extension studies.

Time frame: Baseline and once yearly, up to approximately 108 months of observation

Population: Study was terminated early due to observed increase in risk of stroke in Protocol MK-0822-018. As a result, efficacy analyses in 2nd Extension were limited to change in BMD and incidence of osteoporotic clinical fractures; Base Study + First Extension + Second Extension: Change in Height from Baseline Stature was not analyzed.

BMD was measured by DXA at the femoral neck starting at screening, and at yearly intervals until the end of the study (2nd extension study) for all participants who entered the 2nd extension study. LS means percent change in BMD from original baseline are provided through Month 108. At Months 96 and 108, approximately 3% or fewer participants in each treatment group had BMD data, and results at those time points should be viewed with caution. NOTE: The mean percent change in BMD from baseline in participants originally randomized to placebo includes BMD results obtained after those participants were switched to open-label odanacatib, which occurred at different times relative to their start of blinded study medication in the base study. A longitudinal model with terms for treatment, stratum, region and interaction between treatment and time as fixed effects (LS means weighted for region and stratum size) was used for analysis.

Time frame: Baseline and once yearly, up to approximately 108 months of observation

Population: The FAS population consisted of all randomized participants who entered the second extension study and received at least one dose of open-label treatment and had the necessary on-treatment information (femoral neck BMD).

BMD was measured by DXA at the lumbar spine starting at randomization, and at yearly intervals until the end of the study (2nd extension study) for all participants who entered the 2nd extension study. Measurements were made on at least 3 vertebrae for all time points; vertebrae with fractures were excluded from analyses. At Months 96 and 108, approximately 3% or fewer participants in each treatment group had BMD data, and results at those time points should be viewed with caution. NOTE: The mean percent change in BMD from baseline in participants originally randomized to placebo includes BMD results obtained after those participants were switched to open-label odanacatib, which occurred at different times relative to their start of blinded study medication in the base study. A longitudinal model with terms for treatment, stratum, region & interaction between treatment and time as fixed effects (LS means weighted for region & stratum size) was used for analysis.

Time frame: Baseline and once yearly, up to approximately 108 months of observation

Population: The FAS population consisted of all randomized participants who entered the second extension study and received at least one dose of open-label treatment and had the necessary on-treatment information (lumbar spine BMD).

BMD was measured by DXA at the trochanter starting at screening, and at yearly intervals until the end of the study (2nd extension study) for all participants who entered the 2nd extension study. LS means percent change in BMD from original baseline are provided through Month 108. At Months 96 and 108, approximately 3% or fewer participants in each treatment group had BMD data, and results at those time points should be viewed with caution. NOTE: The mean percent change in BMD from baseline in participants originally randomized to placebo includes BMD results obtained after those participants were switched to open-label odanacatib, which occurred at different times relative to their start of blinded study medication in the base study. A longitudinal model with terms for treatment, stratum, region and interaction between treatment and time as fixed effects (LS means weighted for region and stratum size) was used for analysis.

Time frame: Baseline and once yearly, up to approximately 108 months of observation

Population: The FAS population consisted of all randomized participants who entered the second extension study and received at least one dose of open-label treatment and had the necessary on-treatment information (trochanter BMD).

Morphometric vertebral fractures were assessed by central adjudication on lumbar and thoracic spinal radiographs acquired at baseline (base study) and at Months 6, 12, and subsequent yearly time points. Incident fractures (i.e., occurring after baseline) in previously normal vertebral bodies and/or those occurring after baseline in previously fractured vertebral bodies (Genant semiquantitative \[SQ\] Grade 1-3) (T4 to L4) were confirmed by quantitative morphometric (QM) and either SQ or binary SQ (yes/no) methodologies. A time-to-event methodology was used to evaluate results: life-table estimates of the incidence proportion (cumulative incidence) of participants with at least one morphometric vertebral fracture were assessed at Months 6, 12, and subsequent yearly intervals; an interval-censored approach was then used to determine incidence rate (number of participants with a morphometric fracture per 100 person-years of follow-up).

Time frame: Up to approximately 108 months of observation

Population: Study was terminated early due to observed increase in risk of stroke in Protocol MK-0822-018. As a result, efficacy analyses in 2nd Extension were limited to change in BMD and incidence of osteoporotic clinical fractures; Base Study + 1st Ext + 2nd Ext: Time From Baseline to First Morphometrically-Assessed Vertebral Fracture was not analyzed.

Osteoporotic clinical fractures (combining vertebral and non-vertebral) were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Vertebral fractures were assessed for all vertebral levels (C7, T1 to T12, L1 to L5). Non-vertebral fractures were assessed across multiple anatomical sites including clavicle, distal femur or shaft, fibula, hip, humerus, pelvis, radius, ribs, sacrum, tibia, ulna, and wrist. Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 74 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study + first extension) was used for analysis.

Osteoporotic vertebral clinical fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Vertebral fractures were assessed for all vertebral levels (C7, T1 to T12, L1 to L5). Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 74 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study + first extension) was used for analysis.

Height was measured by wall-mounted stadiometer at randomization and at yearly intervals in the base study.

Time frame: Baseline and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (height loss) was used for analysis (base study).

BMD was measured by DXA at the distal one-third radius at randomization and at yearly intervals until the end of the study (base study) in an approximate 10% random subset of participants at selected sites.

Time frame: Baseline and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (distal one-third radius BMD) was used for analysis (base study).

BMD was measured by DXA at the distal one-third radius at randomization and at yearly intervals until the end of the study (base study) in an approximate 10% random subset of bisphosphonate-intolerant participants at selected sites. Bisphosphonates are anti-resorptive agents used in the treatment of osteoporosis. Bisphosphonate-intolerance was defined by contraindication or history of intolerance to bisphosphonates, or physician's determination of unsuitability for bisphosphonate treatment.

Time frame: Baseline and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (distal one-third radius BMD) was used for analysis (base study).

BMD was measured by DXA for all participants at the femoral neck-hip at screening, Months 6, 12, and subsequent yearly intervals until the end of the study (base study).

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (femoral neck BMD) was used for analysis (base study).

BMD was measured by DXA in bisphosphonate-intolerant participants at the femoral neck-hip at screening, and at yearly intervals until the end of the study (base study). Bisphosphonates are anti-resorptive agents used in the treatment of osteoporosis. Bisphosphonate-intolerance was defined by contraindication or history of intolerance to bisphosphonates, or physician's determination of unsuitability for bisphosphonate treatment.

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (femoral neck BMD) was used for analysis (base study).

BMD was measured by DXA for all participants at the lumbar spine at randomization, Months 6, 12, and subsequent yearly intervals until the end of the study (base study). Measurements were made on at least 3 vertebrae for all time points; vertebrae with fractures were excluded from analyses. at

Time frame: Baseline, Month 6, and once yearly, approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (lumbar spine BMD) was used for analysis (base study).

BMD was measured by DXA in bisphosphonate-intolerant participants at the lumbar spine at randomization, and at yearly intervals until the end of the study (base study). Measurements were made on at least 3 vertebrae for all time points; vertebrae with fractures were excluded from analyses. Bisphosphonates are anti-resorptive agents used in the treatment of osteoporosis. Bisphosphonate-intolerance was defined by contraindication or history of intolerance to bisphosphonates, or physician's determination of unsuitability for bisphosphonate treatment.

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (lumbar spine BMD) was used for analysis (base study).

BMD was measured by DXA for all participants at the total hip at screening, Months 6, 12, and subsequent yearly intervals until the end of the study (base study).

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (total hip BMD) was used for analysis (base study).

BMD was measured by DXA in bisphosphonate-intolerant participants at the total hip at screening, and at yearly intervals until the end of the study (base study). Bisphosphonates are anti-resorptive agents used in the treatment of osteoporosis. Bisphosphonate-intolerance was defined by contraindication or history of intolerance to bisphosphonates, or physician's determination of unsuitability for bisphosphonate treatment.

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (total hip BMD) was used for analysis (base study).

BMD was measured by DXA for all participants at the trochanter-hip at screening, Months 6, 12, and subsequent yearly intervals until the end of the study (base study).

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (trochanter BMD) was used for analysis (base study).

BMD was measured by DXA in bisphosphonate-intolerant participants at the trochanter-hip at screening, and at yearly intervals until the end of the study (base study). Bisphosphonates are anti-resorptive agents used in the treatment of osteoporosis. Bisphosphonate-intolerance was defined by contraindication or history of intolerance to bisphosphonates, or physician's determination of unsuitability for bisphosphonate treatment.

Time frame: Baseline, Month 6, and once yearly, up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (trochanter BMD) was used for analysis (base study).

BSAP is an enzyme produced by matrix-synthesizing osteoblasts during the synthesis of collagenous bone matrix (type 1 collagen) used as a biochemical marker of bone formation. Serum BSAP was assessed at randomization, Month 6, and at yearly intervals until the end of the study (base study) in an approximate 10% random subset participants at selected sites. The log-transformed fraction from baseline in BSAP was determined using a longitudinal model with terms for treatment, stratum, and interaction between treatment and time as fixed effects (LS means weighted for stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 6, and once yearly up to 4 years

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (BSAP) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (base study).

P1NP is a cleavage fragment produced during the synthesis of collagenous bone matrix (type 1 collagen) used as a biochemical marker of bone formation. Serum P1NP was assessed at randomization, Month 6, and at yearly intervals until the end of the study (base study) in an approximate 10% random subset participants at selected sites. The log-transformed fraction from baseline in P1NP was determined using a longitudinal model with terms for treatment, stratum, and interaction between treatment and time as fixed effects (LS means weighted for stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 6, and once yearly up to 4 years

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (P1NP) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (base study).

s-CTx, a biochemical marker of bone resorption by osteoclasts reflecting collagen breakdown products, was assessed at randomization, Month 6, and at yearly intervals until the end of the study (base study) in an approximate 10% random subset participants at selected sites. The log-transformed fraction from baseline in s-CTx was determined using a longitudinal model with terms for treatment, stratum, region and interaction between treatment and time as fixed effects (LS means weighted for region and stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 6, and once yearly up to 4 years

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (s-CTx) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (base study).

u-NTx, a biochemical marker of bone resorption, was assessed at randomization, Month 6, and at yearly intervals until the end of the study (base study) in an approximate 10% random subset participants at selected sites. Urine NTx measurements (in bone collagen equivalents \[BCE\]) were normalized to urine Cr concentration (i.e., u-NTx/Cr ratio) and the log-transformed fraction from baseline in u-NTx/Cr ratio was then determined using a longitudinal model with terms for treatment, stratum, region and interaction between treatment and time as fixed effects (LS means weighted for region and stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 6, and once yearly up to 4 years

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (u-NTx/Cr) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (base study).

An adverse event is defined as any untoward medical occurrence in a person or clinical investigation participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. The incidence rate of participants with adverse events (number of participants with an event per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

An adverse event is defined as any untoward medical occurrence in a person or clinical investigation participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. The incidence rate of participant discontinuations from treatment due to adverse events (number of participants with an event per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

Osteoporotic vertebral clinical fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Vertebral fractures were assessed for all vertebral levels (C7, T1 to T12, L1 to L5). Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence rate of participants with fracture (number of participants with a fracture per 100 person-years of follow-up) is provided.

Time frame: Up to approximately 60 months of observation

Population: The FAS including all randomized participants who took at least one dose of study medication with follow-up from start of prime therapy to study termination (base study) was used for analysis.

The time to first TIMI-adjudicated 3-point MACE (a composite outcome measure of 1. cardiovascular death, 2. non-fatal definite MI, or 3. non-fatal definite stroke) was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated 4-point MACE (a composite outcome measure of 1. cardiovascular death, 2. non-fatal definite MI, 3. non-fatal definite stroke, or 4. hospitalization for unstable angina) was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated all-cause death was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. The analysis includes data obtained from vital status data collections of participants no longer followed in the base study. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated any reported episode of atrial fibrillation or atrial flutter was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Participants with known history of atrial fibrillation or atrial flutter were included and electrocardiogram confirmation was not required. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated cardiovascular death was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. The analysis includes data obtained from vital status data collections of participants no longer followed in the base study. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated fatal definite myocardial infarction was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated fatal or non-fatal definite myocardial infarction was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated fatal or non-fatal definite stroke was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated fatal definite stroke was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated hospitalization for unstable angina was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated new or presumed new onset atrial fibrillation or atrial flutter was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Participants with known history of atrial fibrillation or atrial flutter were excluded and electrocardiogram confirmation was not required. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The time to first TIMI-adjudicated new or presumed new onset atrial fibrillation or atrial flutter was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Participants with known history of atrial fibrillation or atrial flutter were excluded and electrocardiogram confirmation was required. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

Height was measured by wall-mounted stadiometer at randomization and at yearly intervals in the base study.

Time frame: Up to approximately 60 months of observation

Population: The FAS population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (height) was used for analysis (base study).

aBMD was measured at the femoral neck at baseline and Month 60 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 60

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (femoral neck aBMD) was used for analysis (PN032-Base + Extension).

aBMD was measured at the lumbar spine (L1 to L4) at baseline and Month 60 using DXA . If a vertebra was fractured at baseline or became fractured during the study, its BMD measurement was excluded from analysis. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 60

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (lumbar spine aBMD) was used for analysis (PN032-Base + Extension).

aBMD was measured at the total hip at baseline and Month 60 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 60

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (total hip aBMD) was used for analysis (PN032-Base + Extension).

aBMD was measured at the trochanter at baseline and Month 60 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 60

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (trochanter aBMD) was used for analysis (PN032-Base + Extension).

Compartment-specific effects of osteoporosis were assessed by measuring cortical vBMD at the total hip using quantitative computed tomography. The percent change from baseline at Month 60 (base study + extension study) was then assessed using a longitudinal data analysis model including terms for treatment, stratum (prior vertebral fracture \[yes/no\]) and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 60

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (total hip cortical vBMD) was used for analysis (PN032-Base).

aBMD was measured at the the distal one-third radius at baseline and Month 24 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (distal one-third radius aBMD) was used for analysis (PN032-Base).

aBMD was measured at the femoral neck at baseline and Month 24 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (femoral neck aBMD) was used for analysis (PN032-Base).

aBMD was measured at the total hip at baseline and Month 24 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (total hip aBMD) was used for analysis (PN032-Base).

aBMD was measured at the trochanter at baseline and Month 24 using DXA. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (trochanter aBMD) was used for analysis (PN032-Base).

aBMD was measured at the lumbar spine (L1 to L4) at baseline and Month 24 using DXA . If a vertebra was fractured at baseline or became fractured during the study, its BMD measurement was excluded from analysis. The percent change from baseline was assessed using a longitudinal data analysis model including terms for treatment, stratum and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (lumbar spine aBMD) was used for analysis (PN032-Base).

BSAP is an enzyme produced by matrix-synthesizing osteoblasts during the synthesis of collagenous bone matrix (type 1 collagen) used as a biochemical marker of bone formation. Serum BSAP was assessed at baseline and Month 24 in an approximate 10% random subset participants at selected sites in the P018 base study who were additionally included in the imaging substudy PN032-base study. The log-transformed fraction from baseline in BSAP was determined using a longitudinal model with terms for treatment, stratum, and interaction between treatment and time as fixed effects (LS means weighted for stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 24

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (BSAP) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (PN032-Base).

Compartment-specific effects of osteoporosis were assessed by measuring cortical vBMD at the total hip using quantitative computed tomography. The percent change from baseline at Month 24 was then assessed using a longitudinal data analysis model including terms for treatment, stratum (prior vertebral fracture \[yes/no\]) and interaction between treatment and time as fixed effects (LS means weighted for stratum size).

Time frame: Baseline, Month 24

Population: The FAS including consisting of all randomized participants who took at least one dose of blinded study treatment and have a baseline and at least one on-treatment measurement available (total hip cortical vBMD) was used for analysis (PN032-Base).

P1NP is a cleavage fragment produced during the synthesis of collagenous bone matrix (type 1 collagen) used as a biochemical marker of bone formation. Serum P1NP was assessed at baseline and Month 24 in an approximate 10% random subset participants at selected sites in the P018 base study who were additionally included in the imaging substudy PN032-base study. The log-transformed fraction change from baseline in P1NP was determined using a longitudinal model with terms for treatment, stratum, and interaction between treatment and time as fixed effects (LS means weighted for stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 24

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (P1NP) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (PN032-Base).

s-CTx, a biochemical marker of bone resorption by osteoclasts reflecting collagen breakdown products, was assessed at baseline and Month 24 in an approximate 10% random subset participants at selected sites in the P018 base study who were additionally included in the imaging substudy PN032-base study. The log-transformed fraction from baseline in s-CTx was determined using a longitudinal model with terms for treatment, stratum, and interaction between treatment and time as fixed effects (LS means weighted for stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 24

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (s-CTx) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (PN032-Base).

u-NTx, a biochemical marker of bone resorption by osteoclasts reflecting collagen breakdown products, was assessed at baseline and Month 24 in an approximate 10% random subset participants at selected sites in the P018 base study who were additionally included in the imaging substudy PN032-base study. Urine NTx measurements (in BCE) were normalized to urine Cr concentration (i.e., u-NTx/Cr ratio) and the log-transformed fraction from baseline in u-NTx/Cr ratio was then determined using a longitudinal model with terms for treatment, stratum, and interaction between treatment and time as fixed effects (LS means weighted for stratum size). Back-transformed weighted geometric LS means values for percent change from baseline are provided.

Time frame: Baseline, Month 24

Population: The per-protocol population including all randomized participants who took at least one dose of study medication and having the necessary on-treatment information (u-NTx/Cr) and excluding participants and/or data points that represent clinically important deviations from the protocol-specified criteria was used for analysis (PN032-Base).

Osteoporotic vertebral clinical fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Vertebral fractures were assessed for all lumbar vertebral levels (L1 to L5). Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence of participants in the second extension study with osteoporotic vertebral clinical fractures is provided. Due to early termination of the study, the cumulative incidence using a time-to-event methodology was not assessed across base and extension studies.

Time frame: Up to approximately 34 months of observation

Population: The All-Participants-as-Treated Population including all randomized participants who entered the second extension study and received at least one dose of open-label treatment was used for analysis.

Osteoporotic vertebral clinical fractures were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Vertebral fractures were assessed for all thoracic vertebral levels (T1 to T12). Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence of participants in the second extension study with osteoporotic vertebral clinical fractures is provided. Due to early termination of the study, the cumulative incidence using a time-to-event methodology was not assessed across base and extension studies.

Time frame: Up to approximately 34 months of observation

Population: The All-Participants-as-Treated Population including all randomized participants who entered the second extension study and received at least one dose of open-label treatment was used for analysis.

Osteoporotic clinical fractures of any type were confirmed by central adjudication using radiographic evidence on all symptomatic fractures reported as adverse experiences (excluding fractures of the fingers, toes, face, and skull). Non-vertebral fractures were assessed across multiple anatomical sites including clavicle, distal femur or shaft, fibula, hip, humerus, pelvis, radius, ribs, sacrum, tibia, ulna, and wrist; Vertebral fractures assessed across all vertebral levels (C7, T1 to T12, L1 to L5) were included in the analysis. Osteoporotic (fragility) fractures did not include fractures with traumatic or pathological etiologies. A clinical fracture was defined as a fracture with clinical fracture-related symptoms (e.g., pain). A time-to-event methodology was used to evaluate results: the incidence proportion (cumulative incidence) of participants in the second extension study with at least one osteoporotic clinical fracture of any type is provided.

Time frame: Up to approximately 34 months of observation

Population: The The All-Participants-as-Treated Population including all randomized participants who entered the second extension study and received at least one dose of open-label treatment was used for analysis.

Atypical subtrochanteric/diaphyseal femoral fractures (AFF) are an uncommon type of low-energy (eg, osteoporotic) femoral shaft fracture of unclear causation infrequently reported in osteoporotic persons treated with long-term anti-resorptive therapy (eg, bisphosphonates). All femoral (femur, femur-distal, femur-shaft) fractures in the base study + first extension were adjudicated against both ASBMR 2010 & 2013 criteria. All 5 major features (ie, location along femoral shaft, no/minimal trauma, transverse/short oblique fracture, non-comminuted, complete/incomplete fracture) were required for ASBMR 2010 AFF case definition; while 4 of 5 major features (ie, no/minimal trauma, substantially transverse orientation at cortical origin with possible oblique orientation medially, non-comminuted/minimally comminuted, complete/incomplete fracture, localized periosteal reaction of lateral cortex) with requirement for location along femoral shaft were required for ASBMR 2013 AFF case definition.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population, including all randomized participants who took at least one dose of study medication, was used for analysis (base study + first extension).

The incidence rate of femoral shaft fracture events (including both atypical and non-atypical; of any etiology including traumatic) confirmed by adjudication was determined for the base study + first extension. Results are expressed as number of participants with an event per 100 person-years of follow-up.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated 3-point MACE (a composite outcome measure of 1. cardiovascular death, 2. non-fatal definite MI, or 3. non-fatal definite stroke) was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated 4-point MACE (a composite outcome measure of 1. cardiovascular death, 2. non-fatal definite MI, 3. non-fatal definite stroke, or 4. hospitalization for unstable angina) was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated all-cause death was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. The analysis includes data obtained from vital status data collections of participants no longer followed in the base study. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated any reported episode of atrial fibrillation or atrial flutter was determined for the base study using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Participants with known history of atrial fibrillation or atrial flutter were included and electrocardiogram confirmation was not required. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated cardiovascular death was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. The analysis includes data obtained from vital status data collections of participants no longer followed in the base study. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated fatal definite myocardial infarction was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated fatal or non-fatal definite myocardial infarction was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated fatal or non-fatal definite stroke was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated fatal definite stroke was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated hospitalization for unstable angina was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated new or presumed new onset atrial fibrillation or atrial flutter was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Participants with known history of atrial fibrillation or atrial flutter were excluded and electrocardiogram confirmation was not required. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

The time to first TIMI-adjudicated new or presumed new onset atrial fibrillation or atrial flutter was determined for the base study + first extension using a Cox proportional hazards model with terms for treatment, stratum and geographic region. Participants with known history of atrial fibrillation or atrial flutter were excluded and electrocardiogram confirmation was required. Results are expressed as number of participants with an event per 100 person-years of follow-up. Results from MK-0822-083, a follow up study to the MK-088-018 base study and the double-blind period of the first extension study for participants who discontinued prior to 5 years follow-up, are not included.

Time frame: Up to approximately 74 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study + first extension).

Atypical subtrochanteric/diaphyseal femoral fractures (AFF) are an uncommon type of low-energy (eg, osteoporotic) femoral shaft fracture of unclear causation infrequently reported in osteoporotic persons treated with long-term anti-resorptive therapy (eg, bisphosphonates). All femoral (femur, femur-distal, femur-shaft) fractures in the base study were adjudicated against both ASBMR 2010 and 2013 criteria. All 5 major features (ie, location along femoral shaft, no/minimal trauma, transverse/short oblique fracture, non-comminuted, complete/incomplete fracture) were required for ASBMR 2010 AFF case definition; while 4 of 5 major features (ie, no/minimal trauma, substantially transverse orientation at cortical origin with possible oblique orientation medially, non-comminuted/minimally comminuted, complete/incomplete fracture, localized periosteal reaction of lateral cortex) with requirement for location along femoral shaft were required for ASBMR 2013 AFF case definition.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).

The incidence rate of femoral shaft fracture events (including both atypical and non-atypical; of any etiology including traumatic) confirmed by adjudication was determined for the base study. Results are expressed as number of participants with an event per 100 person-years of follow-up.

Time frame: Up to approximately 60 months of observation

Population: The All-Participants-as-Treated population including all randomized participants who took at least one dose of study medication was used for analysis (base study).