Comparison of Teriparatide With Alendronate for Treating Glucocorticoid-Induced Osteoporosis - Study Results

Study Type:	Interventional
Study Design:	Randomized, Double Blind (Subject, Caregiver, Investigator, Outcomes Assessor), Active Control, Parallel Assignment
Condition:	Osteoporosis
Interventions:	Drug: Teriparatide Drug: Alendronate Sodium Drug: Placebo

Key information relevant to the recruitment process for the overall study, such as dates of the recruitment period and locations
No text entered.

Significant events and approaches for the overall study following participant enrollment, but prior to group assignment
No text entered.

	Teriparatide	Alendronate
STARTED	214	214
COMPLETED	123	118
NOT COMPLETED	91	96
Adverse Event	30	18
Withdrawal by Subject	29	42
Death	9	15
Lost to Follow-up	4	13
Protocol Violation	3	4
Protocol Entry Criteria Unmet	1	2
Sponsor Decision	6	1
Unspecified	2	1
Significant Lab Value	2	0
Physician Decision	5	0

	Teriparatide	Alendronate	Total
Number of Participants [units: participants]	214	214	428
Age, Customized [units: participants]
Age < 35 years	13	19	32
35 <= Age < 50 years	55	35	90
50 <= Age < 65 years	86	93	179
Age >= 65 years	60	67	127
Gender [units: participants]
Female	172	173	345
Male	42	41	83
Region of Enrollment [units: participants]
United States	62	64	126
Belgium	8	8	16
Denmark	4	3	7
Austria	1	2	3
Norway	1	1	2
Argentina	23	20	43
Brazil	72	73	145
Colombia	10	9	19
Venezuela	4	3	7
Mexico	14	15	29
Finland	1	2	3
Germany	14	14	28
Origin [units: participants]
Caucasian	153	148	301
Non-Caucasian	61	66	127
Prior Bisphosphonate User [units: participants]
Yes	20	20	40
No	194	194	388
Baseline Biochemical Markers of Bone Metabolism [units: ug/L] Mean ± Standard Deviation
Serum N-terminal propeptide of type 1 procollagen	46.161 ± 35.471	47.235 ± 29.606	46.704 ± 32.562
Procollagen I carboxy-terminal propeptide	166.402 ± 99.625	152.109 ± 56.637	158.845 ± 79.910
Bone-specific alkaline phosphatase	9.465 ± 4.522	9.711 ± 4.527	9.595 ± 4.513
Osteocalcin	17.130 ± 14.422	15.878 ± 10.286	16.492 ± 12.468
Baseline Bone Mineral Density [units: grams per square centimeters] Mean ± Standard Deviation
Lumbar Spine Bone Mineral Density	0.845 ± 0.131	0.844 ± 0.128	0.845 ± 0.129
Total Hip Bone Mineral Density	0.744 ± 0.113	0.757 ± 0.124	0.750 ± 0.119
Femoral Neck Bone Mineral Density	0.682 ± 0.114	0.697 ± 0.119	0.689 ± 0.116
C-terminal telopeptide of type I collagen [units: pmol/L] Mean ± Standard Deviation	3790.483 ± 2478.189	4543.156 ± 4838.919	4185.328 ± 3907.204

Measure Type	Primary
Measure Title	Change From Baseline at 18 Month Endpoint in Lumbar Spine Bone Mineral Density (BMD)
Measure Description	change from baseline at endpoint in bone mineral density of the lumbar spine as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	18 month endpoint
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	198	195
Change From Baseline at 18 Month Endpoint in Lumbar Spine Bone Mineral Density (BMD) [units: grams per square centimeters] Least Squares Mean ± Standard Error	0.059 ± 0.006	0.028 ± 0.006

Statistical Analysis 1 for Change From Baseline at 18 Month Endpoint in Lumbar Spine Bone Mineral Density (BMD)

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline lumbar spine bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use+gender. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	All comparisons were conducted using a 2-sided significance level of 0.05.

Measure Type	Secondary
Measure Title	Change From Baseline at 18 Month Endpoint in Lumbar Spine Bone Mineral Density (BMD), Female Subset
Measure Description	change from baseline at endpoint in bone mineral density of the lumbar spine as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	18 month endpoint
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients (female only subset).

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	160	159
Change From Baseline at 18 Month Endpoint in Lumbar Spine Bone Mineral Density (BMD), Female Subset [units: grams per square centimeters] Least Squares Mean ± Standard Error	0.056 ± 0.006	0.023 ± 0.006

Statistical Analysis 1 for Change From Baseline at 18 Month Endpoint in Lumbar Spine Bone Mineral Density (BMD), Female Subset

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline lumbar spine bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	All comparisons were conducted using a 2-sided significance level of 0.05.

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined
Measure Description	change from baseline in bone mineral density of the lumbar spine as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	3, 6, 12, 18, 24, 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. No missing data were imputed.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	198	195
Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change from baseline at Month 36 (N=123,N=112)	0.090 ± 0.006	0.044 ± 0.007
Change from baseline at Month 24 (N=136,N=131)	0.081 ± 0.006	0.043 ± 0.006
Change from baseline at Month 18 (N=156,N=148)	0.066 ± 0.005	0.031 ± 0.005
Change from baseline at Month 12 (N=170,N=159)	0.054 ± 0.004	0.028 ± 0.004
Change from baseline at Month 6 (N=178,N=173)	0.034 ± 0.004	0.018 ± 0.004
Change from baseline at Month 3 (N=183,N=184)	0.019 ± 0.004	0.012 ± 0.004

Statistical Analysis 1 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.058

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 3

Statistical Analysis 2 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 6 months

Statistical Analysis 3 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 12 months

Statistical Analysis 4 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 months

Statistical Analysis 5 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 24 months

Statistical Analysis 6 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36 months

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Female Subset
Measure Description	change from baseline in bone mineral density of the lumbar spine as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	3, 6, 12, and 18 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. No missing data were imputed.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	172	173
Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Female Subset [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change from baseline at Month 18 (N=127,N=119)	0.062 ± 0.005	0.027 ± 0.005
Change from baseline at Month 12 (N=139,N=129)	0.052 ± 0.004	0.024 ± 0.004
Change from baseline at Month 6 (N=147,N=139)	0.031 ± 0.004	0.015 ± 0.004
Change from baseline at Month 3 (N=149,N=150)	0.017 ± 0.004	0.012 ± 0.004

Statistical Analysis 1 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Female Subset

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.120

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 3 Months

Statistical Analysis 2 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Female Subset

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 6 months

Statistical Analysis 3 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Female Subset

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 12 months

Statistical Analysis 4 for Time Course of Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Female Subset

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 months

Measure Type	Secondary
Measure Title	Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined
Measure Description	change from baseline in bone mineral density of the lumbar spine as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	24 and 36 months and Endpoint at 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. For 24 and 36 month time points, no missing data were imputed. However, at the 36 month endpoint, last observation carried forward analysis was applied.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	198	195
Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change, baseline to 36 month endpoint(N=198,N=195)	0.073 ± 0.008	0.034 ± 0.008
Change from baseline at 36 months (N=123, N=112)	0.090 ± 0.006	0.044 ± 0.007
Change from baseline at 24 months (N=136, N=131)	0.081 ± 0.006	0.043 ± 0.006

Statistical Analysis 1 for Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline lumbar spine bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use+gender. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for 36 month endpoint

Statistical Analysis 2 for Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for 36 months

Statistical Analysis 3 for Change From Baseline in Lumbar Spine Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 24 months

Measure Type	Secondary
Measure Title	Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined
Measure Description	change from baseline in bone mineral density of the femoral neck as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	18, 24, 36 months, and 18 and 36 month endpoints
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. For 18, 24 and 36 month time points, no missing data were imputed. However at the 36 month endpoint, last observation carried forward analysis was applied.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	185	177
Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change, baseline to 36 month endpoint(N=185,N=177)	0.033 ± 0.006	0.017 ± 0.006
Change from baseline at 36 months(N=120,N=113)	0.041 ± 0.005	0.021 ± 0.005
Change from baseline at 24 months(N=135,N=131)	0.030 ± 0.005	0.015 ± 0.005
Change, baseline to 18 month endpoint(N=185,N=176)	0.024 ± 0.005	0.014 ± 0.005
Change from baseline at 18 months (N=156,N=145)	0.028 ± 0.005	0.017 ± 0.005

Statistical Analysis 1 for Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline femoral neck bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use+gender. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36-month endpoint

Statistical Analysis 2 for Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36 months

Statistical Analysis 3 for Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.002

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 24 months

Statistical Analysis 4 for Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	0.011

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline femoral neck bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use+gender. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 month endpoint

Statistical Analysis 5 for Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.015

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 months

Measure Type	Secondary
Measure Title	Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined
Measure Description	change from baseline in bone mineral density of the total hip as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	18, 24, 36 months, and 18 and 36 month endpoints
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. For 18, 24 and 36 month time points, no missing data were imputed. However, at 18 and 36 month endpoints, last observation carried forward analyses were applied.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	185	177
Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change, baseline to 36 month endpoint(N=185,N=177)	0.032 ± 0.005	0.017 ± 0.005
Change from baseline at 36 months(N=120,N=113)	0.037 ± 0.005	0.020 ± 0.005
Change from baseline at 24 months(N=135,N=131)	0.034 ± 0.004	0.018 ± 0.004
Change, baseline to 18 month endpoint(N=185,N=176)	0.026 ± 0.004	0.017 ± 0.004
Change from baseline at 18 months (N=156,N=144)	0.027 ± 0.004	0.018 ± 0.004

Statistical Analysis 1 for Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline total hip bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use+gender. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36 month endpoint

Statistical Analysis 2 for Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36 months

Statistical Analysis 3 for Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 24 months

Statistical Analysis 4 for Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	ANOVA
P Value ^[3]	0.006

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	This is a last observation carried forward analysis that included all patients with a baseline and at least one postbaseline total hip bone mineral density (BMD) measurement.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	ANOVA Model: Change in BMD=treatment+region+prior bisphosponate use+gender. Least Squares Means from treatment.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 month endpoint

Statistical Analysis 5 for Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.009

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 months

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined
Measure Description	change from baseline in bone mineral density of the femoral neck as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	12, 18, 24, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. No missing data were imputed.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	185	177
Time Course of Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change from baseline at Month 36 (N=120,N=113)	0.041 ± 0.005	0.021 ± 0.005
Change from baseline at Month 24 (N=135,N=131)	0.030 ± 0.005	0.015 ± 0.005
Change from baseline at Month 18 (N=156,N=145)	0.028 ± 0.005	0.017 ± 0.005
Change from baseline at Month 12 (N=167,N=158)	0.026 ± 0.005	0.012 ± 0.005

Statistical Analysis 1 for Time Course of Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 12 months

Statistical Analysis 2 for Time Course of Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.015

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 months

Statistical Analysis 3 for Time Course of Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.002

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 24 months

Statistical Analysis 4 for Time Course of Change From Baseline in Femoral Neck Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36 months

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined
Measure Description	change from baseline in bone mineral density of the total hip as assessed by dual energy X-ray absorptiometry (DXA)
Time Frame	12, 18, 24, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
The intention-to-treat analysis was performed using all randomized and treated patients. No missing data was imputed.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	185	177
Time Course of Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined [units: grams per square centimeters] Least Squares Mean ± Standard Error
Change from baseline at 36 Months (N=120, N=113)	0.037 ± 0.005	0.020 ± 0.005
Change from baseline at 24 Months (N=135, N=131)	0.034 ± 0.004	0.018 ± 0.004
Change from baseline at 18 Months (N=156, N=145)	0.027 ± 0.004	0.018 ± 0.004
Change from baseline at 12 Months (N=167, N=158)	0.022 ± 0.004	0.014 ± 0.004

Statistical Analysis 1 for Time Course of Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.011

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 12 months

Statistical Analysis 2 for Time Course of Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	0.009

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 18 months

Statistical Analysis 3 for Time Course of Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 24 months

Statistical Analysis 4 for Time Course of Change From Baseline in Total Hip Bone Mineral Density (BMD), Women and Men Combined

Groups ^[1]	All groups
Method ^[2]	Mixed Models Analysis
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	Mixed Model Repeated Measures: Change in BMD=treatment+region+prior bisphosphonate use+gender+baseline BMD+visit+treatment*visit.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Least Squares Means obtained from the treatment*visit interaction term from the above mixed model repeated measures model.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for 36 months

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum N-terminal Propeptide of Type 1 Procollagen
Measure Description	No text entered.
Time Frame	1, 6, 18, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
Subset of patients with measures of biochemical markers

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	99	100
Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum N-terminal Propeptide of Type 1 Procollagen [units: percent] Mean ± Standard Error
Percent change from baseline at Month 1(N=98,N=99)	107.0 ± 13.744	-14.8 ± 2.861
Percent change from baseline at Month 6(N=86,N=85)	130.8 ± 20.752	-43.2 ± 3.481
Percent change from baseline at Month18(N=77,N=76)	86.3 ± 20.231	-35.8 ± 4.477
Percent change from baseline at Month36(N=59,N=57)	61.7 ± 16.156	-20.2 ± 10.587

Statistical Analysis 1 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum N-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 1

Statistical Analysis 2 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum N-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 6

Statistical Analysis 3 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum N-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 18

Statistical Analysis 4 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum N-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 36

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum C-terminal Propeptide of Type 1 Procollagen
Measure Description	No text entered.
Time Frame	1, 6, 18, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
Subset of patients with measures of biochemical markers.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	78	89
Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum C-terminal Propeptide of Type 1 Procollagen [units: percent] Mean ± Standard Error
Percent change from baseline at Month 1(N=66,N=78)	48.2 ± 7.024	-8.8 ± 2.912
Percent change from baseline at Month 6(N=67,N=76)	10.5 ± 5.318	-22.3 ± 3.694
Percent change from baseline at Month18(N=55,N=65)	5.1 ± 8.604	-21.3 ± 4.207
Percent change from baseline at Month36(N=45,N=49)	-1.3 ± 7.082	-15.4 ± 7.394

Statistical Analysis 1 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum C-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 1

Statistical Analysis 2 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum C-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 6

Statistical Analysis 3 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum C-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	0.004

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 18

Statistical Analysis 4 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum C-terminal Propeptide of Type 1 Procollagen

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	0.013

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 36

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Bone-Specific Alkaline Phosphatase
Measure Description	No text entered.
Time Frame	1, 6, 18, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
Subset of patients with measures of biochemical markers

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	77	89
Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Bone-Specific Alkaline Phosphatase [units: percent] Mean ± Standard Error
Percent change from baseline at Month 1(N=65,N=78)	35.0 ± 7.518	0.5 ± 4.897
Percent change from baseline at Month 6(N=66,N=76)	52.3 ± 10.995	-7.2 ± 12.384
Percent change from baseline at Month18(N=54,N=65)	34.4 ± 9.548	-4.1 ± 10.501
Percent change from baseline at Month36(N=44,N=49)	44.7 ± 14.357	20.6 ± 19.912

Statistical Analysis 1 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Bone-Specific Alkaline Phosphatase

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 1

Statistical Analysis 2 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Bone-Specific Alkaline Phosphatase

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 6

Statistical Analysis 3 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Bone-Specific Alkaline Phosphatase

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 18

Statistical Analysis 4 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Bone-Specific Alkaline Phosphatase

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 36

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum Type 1 Collagen Degradation Fragments
Measure Description	No text entered.
Time Frame	1, 6, 18, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
Subset of patients with measures of biochemical markers

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	84	91
Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum Type 1 Collagen Degradation Fragments [units: percent] Mean ± Standard Error
Percent change from baseline at Month 1(N=70,N=79)	29.0 ± 8.192	-38.6 ± 4.110
Percent change from baseline at Month 6(N=66,N=75)	66.8 ± 17.143	-42.4 ± 5.050
Percent change from baseline at Month18(N=64,N=71)	29.1 ± 11.714	-47.5 ± 6.305
Percent change from baseline at Month36(N=49,N=48)	30.7 ± 14.205	-16.2 ± 23.178

Statistical Analysis 1 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum Type 1 Collagen Degradation Fragments

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 1

Statistical Analysis 2 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum Type 1 Collagen Degradation Fragments

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 6

Statistical Analysis 3 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum Type 1 Collagen Degradation Fragments

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 18

Statistical Analysis 4 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Serum Type 1 Collagen Degradation Fragments

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 36

Measure Type	Secondary
Measure Title	Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Osteocalcin
Measure Description	No text entered.
Time Frame	1, 6, 18, and 36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
Subset of patients with measures of biochemical markers

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	97	101
Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Osteocalcin [units: percent] Mean ± Standard Error
Percent change from baseline at Month1(N=94,N=100)	147.1 ± 26.408	-5.2 ± 3.447
Percent change from baseline at Month 6(N=84,N=88)	120.0 ± 16.713	-33.6 ± 3.656
Percent change from baseline at Month18(N=77,N=77)	108.9 ± 22.750	-25.0 ± 5.118
Percent change from baseline at Month36(N=56,N=58)	62.1 ± 20.929	-19.6 ± 12.980

Statistical Analysis 1 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Osteocalcin

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value for Month 1

Statistical Analysis 2 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Osteocalcin

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value at Month 6

Statistical Analysis 3 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Osteocalcin

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value at Month 18

Statistical Analysis 4 for Time Course of Change From Baseline in Bone Turnover Markers in Subset of Patients - Osteocalcin

Groups ^[1]	All groups
Method ^[2]	Wilcoxon (Mann-Whitney)
P Value ^[3]	<0.001

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	Wilcoxon Rank Sum Test using the aligned ranks for scores.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	P-value at Month 36

Measure Type	Secondary
Measure Title	Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures
Measure Description	Clinical vertebral fracture was defined as a radiographically confirmed fracture that was associated with symptoms such as back pain.
Time Frame	36 months
Safety Issue	No

Population Description

Explanation of how the number of participants for analysis was determined. Includes whether analysis was per protocol, intention to treat, or another method. Also provides relevant details such as imputation technique, as appropriate.
For vertebral fractures, only those patients with baseline and postbaseline spinal radiographs were included in the analysis.

Reporting Groups

	Description
Teriparatide	Teriparatide 20 micrograms/day injection plus oral placebo, 36 months
Alendronate	Alendronate 10 mg/day oral plus injection placebo, 36 months

Measured Values

	Teriparatide	Alendronate
Number of Participants Analyzed [units: participants]	214	214
Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures [units: participants]
Any Fracture	19	27
Nonvertebral Fracture	16	15
Vertebral Fracture	3	13
Clinical Vertebral Fracture	0	4
Nonvertebral Fragility Fracture	9	5
Severity-Radiographic Vertebral Fracture: Mild	1	7
Severity-Radiographic Vertebral Fracture: Moderate	2	2
Severity-Radiographic Vertebral Fracture: Severe	0	4

Statistical Analysis 1 for Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures

Groups ^[1]	All groups
Method ^[2]	Cochran-Mantel-Haenszel
P Value ^[3]	0.212

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	No text entered.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for Any Fracture

Statistical Analysis 2 for Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures

Groups ^[1]	All groups
Method ^[2]	Cochran-Mantel-Haenszel
P Value ^[3]	0.843

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	No text entered.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for Nonvertebral Fracture

Statistical Analysis 3 for Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures

Groups ^[1]	All groups
Method ^[2]	Cochran-Mantel-Haenszel
P Value ^[3]	0.007

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	No text entered.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for Vertebral Fracture

Statistical Analysis 4 for Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures

Groups ^[1]	All groups
Method ^[2]	Cochran-Mantel-Haenszel
P Value ^[3]	0.037

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	No text entered.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for Clinical Vertebral Fracture

Statistical Analysis 5 for Any Fracture, Nonvertebral Fractures, Vertebral Fractures, Clinical Vertebral Fractures, and Severity Fractures

Groups ^[1]	All groups
Method ^[2]	Cochran-Mantel-Haenszel
P Value ^[3]	0.256

[1]	Additional details about the analysis, such as null hypothesis and power calculation:
	No text entered.
[2]	Other relevant information, such as adjustments or degrees of freedom:
	No text entered.
[3]	Additional information, such as whether or not the p-value is adjusted for multiple comparisons and the a priori threshold for statistical significance:
	p-value for Nonvertebral Fragility Fracture

Responsible Party:	Eli Lilly ( Chief Medical Officer )
Study ID Numbers:	6484, B3D-US-GHBZ
Study First Received:	January 13, 2003
Results First Received:	January 14, 2009
Last Updated:	February 13, 2009
ClinicalTrials.gov Identifier:	NCT00051558 History of Changes
Health Authority:	United States: Food and Drug Administration