EVISTA

1
EVISTA Studies OverviewDaniel Thiebaud MD,
Medical Fellow,Global Osteoporosis Strategy,
Eli Lilly, Australia
2
Concept of a SERM
S elective E strogen R eceptor Modulator

• Not an estrogen and not a hormone
• Binds to estrogen receptors
• Has estrogen-like effects in some tissues
• Blocks estrogen effects in some tissues

3
Evista Update

• EVA Evista versus Alendronate
• MORE
• New non vertebral fractures
• Clinical vertebral fractures (3 and 6 months)
• CORE
• Invasive breast cancer and overall safety
• RUTH STAR timelines
• CHOOSE ASIA Observational study

4
Raloxifene versus Alendronate Comparison EVA
Trial

• First ever head-to-head fracture outcome trial
• Compare the osteoporotic fracture risk reduction
  efficacy of raloxifene and alendronate
• Double-blind, randomized, controlled, 5-year
  trial with raloxifene 60 mg/d vs alendronate 10
  mg/d
• Initially planned about 3000 postmenopausal women
  with osteoporosis. Enrollment terminated on Aug
  2004 with 1423 patients randomized, because too
  slow recruitment
• Calcium 500 mg vitamin D 400 IU to all patients
• Sites in US, Canada, and Puerto Rico

Adapted from Recker R et al, ASBMR 2005, JBMR
2005, 20,Suppl 1,S97
5
Baseline Characteristics
Characteristic Raloxifene (N707) Alendronate (N716) P-value
Age (years) 65.5 65.7 0.56
Caucasian () 86.7 86.9 0.83
BMI (kg/m2) 24.8 24.6 0.42
LS BMD (g/cm2) 0.82 0.82 0.79
T-score -2.32 -2.34 0.65
FN BMD (g/cm2) 0.61 0.61 0.98
T-score -2.39 -2.39 0.77
Total Hip BMD (g/cm2) 0.71 0.71 0.71
T-Score -1.99 -2.01 0.64
Adapted from Recker R et al, ASBMR 2005, JBMR
2005, 20,Suppl 1,S97
6
EVA Trial Incidence of VFx and Non-V Fx
Women with 1 new Fx, n() Type of
Fx ALN, 10mg/d RLX 60mg/d P
value N713 N699 Age, yrs
65.7 7.8 65.5 7.7 0.56 Vert or Non
Vert 22 (3.1) 20 (2.9) 0.84 Vertebrala
8 (3.1) 5 (1.9) 0.53
Moderate/Severe 4 (1.6) 0 0.04
Clinical Vertebral 3 (0.4) 0
0.1 NonVertebral 14 (2.0)
15 (2.2) 0.86
Nonvertebral-Sixb 11 (1.5) 10 (1.4)
0.89
b Includes the clavicule, humerus, wrist, pelvis,
hip and leg.
Recker R et al, ASBMR 2005, Abstract in JBMR
2005, 20,Suppl 1,S97
7
MOREMultiple Outcomes of Raloxifene Evaluation
Source
Review

• Multicenter, double-blind, placebo-controlled
  trial
• 25 countries, 180 centers, 3 years with 1 year
  extension
• 7705 postmenopausal women with osteoporosis
• Mean age 66 years
• Raloxifene 60 mg Evista, 120 mg, or placebo
• All patients given daily elemental calcium (500
  mg) and vitamin D (400-600 IU)
• Primary endpoints radiographic vertebral
  fracture, BMD, safety
• Secondary endpoints all osteoporotic fractures,
  cardiovascular health, breast cancer, cognitive
  function

Reviewer Memo
Ettinger B et al. JAMA 282637-45, 1999 Cummings
SR et al. JAMA 2812189-97, 1999
Slide Modified
Memo
8
Incidence Rates for Vertebral, Wrist and Hip
Fractures in Women After Age 50
Source
40 30 20 10
Review
Vertebrae
Reviewer Memo
Annual incidence per 1000 women
Hip
Wrist
50 60 70 80
Age (Years)
Wasnich RD Primer on the Metabolic Bone Diseases
and Disorders of Mineral Metabolism. 4th edition,
1999,
Slide Modified
Memo
9
Risk of New Clinical Vertebral Fractures at 1
Year
Women with and without Prevalent Vertebral
Fractures
Women with Prevalent Vertebral Fractures
2.2
2.2
2.0
2.0
RR 0.34 (95 CI, 0.11 - 0.77)
1.8
1.8
1.6
1.6
1.4
1.4
66
of Women with New Clinical Vertebral Fracture
1.2
1.2
1.0
1.0
RR 0.32 (95 CI, 0.13 - 0.79)
0.8
0.8
0.6
0.6
68
0.4
0.4
0.2
0.2
0.0
0.0
Placebo
RLX 60
Placebo
RLX 60
Marici et al, Arch. Int med, 2002
10
Effect of Raloxifene on New Clinical Vertebral
Fractures at 6 Months
0.44 (n10)
Qu Y, et al. CMRO, 2005, 21 (12) 1955-59.
11
Cumulative Incidence of New Clinical Vertebral
Fractures in the First Year of MORE
P0.007 in the first 6 months for each
raloxifene group compared with placebo
Qu Y, et al. CMRO, 2005, 21 (12) 1955-59.
12
Effect of Raloxifene on New Clinical Vertebral
Fractures at 3 Months
0.22 (n5)
Qu Y, et al. CMRO, 2005, 21 (12) 1955-59.
13
Women with severe osteoporosis

• Does raloxifene prevent multiple new vertebral
  fractures ?
• Does raloxifene prevent the first severe
  vertebral fracture?
• Does raloxifene prevent subsequent fracture (also
  non vertebral) when a severe fracture is present?

14
Source
Effect of Raloxifene on the Risk of 2 or More
New Vertebral Fractures in Women MORE Trial -
3 Years
Review
RR 0.07(95 CI 0.01, 0.56)
1.6
Reviewer Memo
1.2
93
0.8
of Women with ?2 Incident Vertebral Fractures
0.4
0
Placebo N1457
Raloxifene 60 mg/day N1401
p 0.001
Lufkin E et al. North American Menopause Society
12th Annual Meeting Program and Abstract Book,
P21, p70, October 4-6, 2001
Slide Modified
Memo
15
Semiquantitative Evaluation of Vertebral Fracture
Severity
Source
Review
Reviewer Memo
Percent reduction in anterior, mid and/or
posterior vertebral height Adapted from Genant
HK et al. J Bone Miner Res 81137-1148, 1993
Slide Modified
Memo
16
Risk of At Least 1 New Moderate/Severe Vertebral
Fracture MORE Trial 3 Year
37
61
Siris E et al. Osteoporosis Int 13907, 2002
17
Reduction of 47 of at Least 1 New Nonvertebral
Fracture in Women With Baseline SQ Grade 3MORE
Trial - 3 Years
RH0.53 (95 CI 0.29, 0.99) P0.04
20
15
47
of Women with at Least 1 New Nonvertebral
Fracture
10
5
0
Placebo
Raloxifene 60 mg/d
Clavicle, humerus, wrist, pelvis, hip, leg
Delmas PD et al. Bone 2003334522-532
18
Raloxifene prevents Non Vertebral Fracture in
Women with 2 Prevalent Vertebral Fractures(n
1369, mean age 69y MORE Trial - 3 Years
pooled raloxifene
Nonvertebral Fracture
RR0.69 (95 CI 0.48, 0.99) Plt0.05
40
30
of Women With at Least 1 New non-Vertebral
Fracture
20
31
10
0
Placebo
Raloxifene
Clavicle, humerus, wrist, pelvis, hip, leg
Farrerons et al., CTI, 200372(4)391(P230)
19
Source
Randomized Studies of Antiresorptives in

Postmenopausal Osteoporotic Women
Risk of Vertebral Fractures
Review
LS BMD
Relative Risk (95 CI)
Raloxifene
Preexisting vertebral
2.2
1
60 mg/d
fracture (VFx)
No preexisting VFx
2.9
1
Reviewer Memo
Alendronate
Preexisting VFx
6.2
2
5/10 mg/d
No preexisting VFx
6.8
3
Risedronate
Preexisting VFx
4.3
4
5 mg/d
Preexisting VFx
5.9
5
Calcitonin
Preexisting VFx
0.7
6
200 IU/d
0
0.5
1.0
Not head
-
to
-
head comparison, vs placebo
Slide Modified
Memo
20
Differences in Trial Design Baseline
fractures Ca and Vitamin D Supplementation /
Ethical rules

• Baseline fractures and age quite different
  between trials
• Differences in calcium and vitamin D
  supplementation, a regimen that has been shown to
  reduce the risk of hip fractures, may have also
  contributed to the different results in hip.
• The estimated number of patients who received
  calcium and vitamin D supplementation in the FIT
  and WHI HRT trial was 30 -40.
• All patients in the MORE trial were supplemented
  in the trial.
• MORE had stringent ethical rules patients
  having a fracture or losing too much BMD could
  discontinue ¾ more patients discontinued in
  placebo, strong bias against raloxifene
• Also ¾ more patients took additional bone active
  drug in 4th year

21
Number Needed to Treat (NNT) to Prevent 1
Vertebral Fracture
Study Duration (Years)
NNT
Without Preexisting Vertebral Fracture
Raloxifene 60 mg/d 4.0 34 Alendronate 5/10
mg/d 4.2 60
With Preexisting Vertebral Fracture
Raloxifene 60 mg/d 3.0 16 Alendronate 5/10
mg/d 2.9 14 Risedronate 5 mg/d 3.0 20
Not head-to-head comparisons
Delmas PD, et al. J Clin Endocrinol Metab. 2002
873609-3617. Marcus R, et al. Endocrine Rev.
20022316-37.
22

Japan-China trials Any New Clinical
Fractures Asian Women with Osteoporosis - One
Year
10
9
RR 0.11 (0.03-0.51)
8
7
RR 0.17 (0.04-0.75)
6
5
6.0 (n12)
4
3
1.0 (n2)
2
0.7 (n2)
1
0
Raloxifene 60 mg/day
Raloxifene Pooled
Placebo
Nakamura et a, IBMS-ECTS Geneva, June 2005, and
Bone 36, Suppl 2
23
Source
Raloxifene Bone Efficacy - Summary
Review

• Significant reduction in risk of vertebral
  fractures
• 66 ? in risk of clinical vertebral fractures
  during the first year
• 55 ? in risk of women without preexisting
  vertebral fractures at 3 years of therapy.
• Efficacy sustained in the 4th year (40-50
  reduction).
• 93 ? in risk of multiple vertebral fractures at
  3 yr in those without preexisting vertebral
  fractures
• 47 ? in risk of non vertebral fractures in women
  with severe vertebral fractures or 31 in women
  with 2 pre-existing fractures
• Improves properties of bone quality
• 34 reduction of non vertebral fractures in
  MORECORE 8 y in women with pre-existing SQ3
  fracture
• Easy to use and good tolerability

Reviewer Memo
Slide Modified
Memo
24
Source
Completed and Ongoing Large-Scale Raloxifene
Clinical Trials
Review
19,365
20000
15000
10,101
Reviewer Memo
Number of Enrolled Women
7,705
10000
4,011
5000
1,400 1y
1,764
0
Osteoporosis Prevention
MORE
CORE
RUTH
STAR
EVA
MORE, Multiple Outcomes of Raloxifene Evaluation
CORE, Continuing Outcomes Relevant to EVISTA
RUTH, Raloxifene Use for The Heart STAR, Study
of Tamoxifen and Raloxifene EVA,
EVISTA-Alendronate Comparison
2004
Slide Modified
Memo
25
Breast Cancer
26
Effect of Raloxifene on All Breast Cancer MORE
Trial - 4 Years
Source
Review
Reviewer Memo
Placebo
RLX (pooled)

Total Cases 77
Years since Randomization
Arrow denotes annual mammogram (optional)
2004
Sourced from Cauley J et al. Breast Cancer Res
Treatment 65125-34, 2001
Slide Modified
Memo
27
Source
MORE plus CORE Study Design
Review
Gap MORE Conclusion CORE Screening
MORE (N7705)
CORE (n4011)
Three Treatment
Two Treatment
Groups
Groups
Placebo
Placebo
Reviewer Memo
Raloxifene HCl 60 mg/day
Raloxifene HCl 60 mg/day
Raloxifene HCl 120 mg/day
0
1
2
3
4
5
6
7
8
Year
8 Years Total Follow-up
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
28
CORE Study Objectives
Source
Review

• Determine the effect of raloxifene on incidence
  of invasive breast cancer over an additional 4
  years of therapy (8 years total for MORE CORE)
• Determine the effect of raloxifene on incidence
  of invasive, ER() breast cancer over the same
  time period
• Assess the tolerability of raloxifene over 8
  years

Reviewer Memo
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
29
Breast Cancer Assessment
Source
Review

• Clinical breast exams
• MORE every 6 months
• CORE every 12 months
• Mammograms
• MORE baseline and after 2, 3, and 4 years of
  treatment
• CORE baseline and after years 2 and 4 of
  treatment
• Breast cancer cases adjudicated by an independent
  committee of physicians blinded to treatment
  assignment and not affiliated with study sponsor

Reviewer Memo
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
30
Source
CORE Demographics at MORE Baseline
CORE Enrollees (N 4011)
MORE Participants (N 7705)
CORE Primary Analysis Dataset (N 5213)
Review
Characteristic
65.8
Mean age, yr 66.5 66.2 Age ? 60 years
() 81.5 81.2 Mean BMI, kg/m2 25.2 25.3 Caucasian,
() 95.7 95.5 Current smoker, (
yes) 16.7 16.2 Mean time postmenopause,
yr 18.7 18.4 Family history of breast
cancer, () 12.6 12.0 Hysterectomy, (
yes) 22.7 21.4 Previous hormone therapy,
() 29.1 26.5
80.1
Reviewer Memo
25.2
96.2
16.0
17.9
11.9
20.4
25.6
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
31
Source
Incidence of Invasive Breast Cancer 4 Years of
CORE
Review
4.0
Placebo 5.2 per 1000 Women-Yrs
HR 0.41 (95 CI 0.24-0.71)
3.0
Reviewer Memo
N5213
p lt0.001
59
Cumulative Incidence ()
2.0
1.0
Raloxifene 2.1 per 1000 Women-Yrs
0.0
0 1 2 3 4
Year
Jan 1, 1999
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
32
Source
Incidence of Invasive Breast Cancer 8 Years of
MORE plus CORE (N7705)
Review
4.0
Placebo 4.2 per 1000 Women-Yrs
HR 0.34 (95 CI 0.22-0.50)
3.0
Reviewer Memo
p lt0.001
66
Cumulative Incidence ()
2.0
1.0
Raloxifene 1.4 per 1000 Women-Yrs
0.0
0
1
2
3
4
5
6
7
8
Years in Study
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
33
Incidence of Invasive ER and ER- Breast
Cancer8 Years of MORE Plus CORE Trials
HR 0.24 (95 CI 0.15 to 0.40) P lt.001
4
3.5
Placebo (N2576)
Raloxifene (N5129)
3
2.5
Incidence/1000 woman-years
2
HR 1.06 (95 CI 0.43 to 2.59) P .90
1.5
1
0.5
0
n44
n22
n7
n15
ER breast cancer
ER- breast cancer
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
34
Summary of Adverse Outcomes over the 8 Years of
MORE-CORE (N4011)
Percentage of participants who experienced event (n) Percentage of participants who experienced event (n) P-value
Placebo (N1286) Raloxifene (N2725) P-value
Mortality 2.3 (29) 1.7 (47) 0.27
All cancers 8.6 (110) 5.7 (156) 0.001
All cancers excluding breast cancer 6.3 (81) 4.6 (126) 0.027
Hospitalization 40.9 (526) 38.8 (1057) 0.21
Treatment-emergent AEs 99.0 (1273) 98.6 (2688) 0.45
Treatment-emergent serious AEs 45.5 (585) 42.3 (1154) 0.07
Study discontinuation CORE due to AE 2.4 (31) 1.9 (53) 0.35
Excluding non-melanoma skin cancers
Martino S et al. Curr Med Res Opin 2005
35
Summary of Gynecologic AE Data over the 8 Years
of MORE-CORE (N4011)
Percentage of participants who experienced event (n) Percentage of participants who experienced event (n) P-value
Placebo (N1286) Raloxifene (N2725) P-value
Uterine cancer 0.39 (4) 0.32 (7) 0.75
Endometrial hyperplasia 0.29 (3) 0.37 (8) gt0.99
Ovarian cancer 0.16 (2) 0.11 (3) 0.66
Postmenopausal bleeding 5.4 (55) 5.5 (120) 0.87
Vulvovaginal signs and symptoms 5.8 (75) 5.0 (135) 0.26

Martino S et al. Curr Med Res Opin 2005
36
Source
Adverse Events Reported During MORE Plus CORE
8 Years
Review
Number ()
Placebo Raloxifene p-value (n1286) (n2725)
Reviewer Memo
Flushing (hot flushes) 89 (6.9) 342
(12.6) lt0.001 Leg cramps 152 (11.8) 407
(14.9) 0.008 Peripheral edema 120 (9.3) 288
(10.6) 0.240
Martino S, et al. J. Natl. Cancer Inst.
200496(23)1751-1761
Slide Modified
Memo
37
Raloxifene and Non-Vertebral Fx at 8 yrs Poisson
analyses
CORE
CORE/MORE
0.86,1.17
0.83, 1.07
0.63, 0.96
1.0
0.43,1.02
0.94
Incidence RR (95CI) at 6 sites
0.44, 0.92
Incidence RR (95CI) at 6 sites
0.78
0.66
0.64
N 277 1425 615 Adapted from Siris
ES et al J Bone Miner Res 2005 201514-1522
38
Cardiovascular Effects
39
Pathophysiology of Atherothrombosis
Foam Cells
Fatty Streak
Intermediate Lesion
Atheroma
Fibrous Plaque
Complicated Lesion
Plaque Rupture
Clinical Events
Endothelial injury
Lipid accumulation
Inflammation
40
Effect of Raloxifene 60 mg/d on Cardiovascular
Risk Factors
0
-4
-10
-7
-12
Total Chol
LDL-C
HDL-C
TG
Fibrinogen
Plt0.05 vs. placebo
Adapted from Walsh BW et al., JAMA
19982791445-51
41
Cumulative Incidence of Cardiovascular
Events MORE Trial 4 Years
Source
All Enrolled Women N 7705
High-Risk Women n 1035
Review
Total number of events 97
Total number of events 272
14
Placebo
Placebo RLX 60 mg/d
RLX 60 mg/d
40
12
Reviewer Memo
10
8
6
4
4
2
RR0.60 (95 CI0.38-0.95)
RR0.86 (95 CI0.64-1.15)
0
0
12
24
36
0
12
24
36
48
0 12 24 36 48
Months Since Randomization
Adapted from Barrett-Connor E et al. JAMA
287847-57, 2002
Slide Modified
Memo
42
RUTH StudyRaloxifene Use for The Heart

• 10,101 patients, DBRCT, placebo vs raloxifene (60
  mg/d)
• Entry postmenopausal women at high risk for,
  orsuffering from, heart disease
• Primary endpoints
• Coronary CHD death, non-fatal MI, or
  hospitalized acute coronary syndrome other than
  MI
• Invasive breast cancer
• Length of trial up to 7.5 years with anticipated
  completion in 2006

DBRCT double-blind, randomized, controlled
trial CHDcoronary heart disease MImyocardial
infarction
Wenger NK, et al. Am J Cardiol.2002901204-1210.
43
NSABP-P2 (STAR) StudyStudy of Tamoxifen And
Raloxifene

• 19,747 patients, double-blind, randomized
• Tamoxifen (20 mg/d) vs raloxifene (60 mg/d)
• Entry postmenopausal, high risk for invasive
  breast cancer (lobular carcinoma in situ or
  5-year risk of gt1.67 by the Gail model)
• Primary endpoint
• Invasive breast cancer
• Secondary endpoints
• Uterine safety, nonvertebral fracture,
  cardiovascular, overall toxicity and side effects
• Started 1999 with final analyses when 327 cases
  have occurred but women will continue to be
  followed results anticipated in 2006

Vogel V, et al. Clin Breast Cancer.
20023153-159.
44
WHI Estrogen-Progestin TrialGlobal Index
Assessment of Risk-Benefit

• Defined to summarize important aspects of health
  benefits vs risks
• Defined for each woman as the earliest
  occurrence of
• Coronary heart disease (CHD)
• Pulmonary embolism
• Invasive breast cancer
• Stroke

• Endometrial cancer
• Colorectal cancer
• Hip fracture
• Death due to other causes

Writing Group for the Womens Health Initiative.
JAMA. 2002288321-333.
45
Global Safety Index Assessing Risk and Benefit
1.3
NNH 88
Increased Harm
1.2
Relative Risk
HR 1.15 95 CI 1.03-1.28
MORE2 Raloxifene 60 mg/d
1.1
1
1
WHI1 CEE/MPA
0.9
Relative Risk
0.8
HR 0.75 95 CI 0.6-0.96
Not head-to-head trials.
NNT 69
0.7
1. Adapted from Writing Group for the Womens
Health Initiative. JAMA. 2002288321-333.
Increased Benefit
2. Barrett-Connor E. et al., J Bone Minral Res,
2004Aug 19,1270-1275 .
0.6
46
Effect of Raloxifene on WHI Global Risk-Benefit
Index
HR, 0.75 95 CI, 0.59-0.95
HR, 0.75 95 CI, 0.60-0.96
Event Rate Annualized
Placebo
Raloxifene 60 mg/d
Raloxifene 120 mg/d
Barrett-Connor E. et al., J Bone Minral Res,
2004Aug 19,1270-1275
47
Raloxifene Administration and Tolerability

• Single daily dose (60 mg tablets)
• Liver metabolism (glucuronidation)
• May be given without regard to meals or time of
  day
• No adjustment needed for most commonly used
  concomitant medications
• Can be used with Calcium and Vit D (recommended
  in patients with fractures)
• No GI side effects

48
Tolerability With Raloxifene Vs Alendronate In
the Clinical Practice at 12 Months
Plt0.001
30
(n476)
Raloxifene
25.8
(n426)
Alendronate
25
20
16.4
Plt0.001
Plt0.001
Patients
15
11
9.9
10
4.8
3.4
5
0
Total Discontinuation Discontinuation Discontinu
ation due to Aes due to GI disorders
Turbí C et al. Clin Ther 26245-256, 2004.
2004
49
Comparison of Raloxifene and Bisphosphonates on
Adherence, Health Outcomes and Treatment
Satisfaction in Post-Menopausal Asian
Women CHOOSE Asia Observational Study
Objectives

• Primary To demonstrate that raloxifene is
  associated with better adherence compared with
  daily dosing bisphosphonates in Asian
  postmenopausal women at increased risk of
  osteoporotic fractures.
• Secondary To demonstrate that raloxifene therapy
  is associated with improved
• treatment satisfaction
• quality of life
• compared with bisphosphonates.

Poster presented at IBMS-ECTS Geneva , June 2005,
Bone 36, Suppl 2, 2005
50
Methods - 1

• Study Design
• One-year, open-label, observational study
  conducted in
• Hong Kong
• Malaysia
• Pakistan
• Philippines
• Singapore
• Taiwan.
• Postmenopausal women aged 55 years or older and
  at increased risk of osteoporosis.
• Study treatments administered by a physician
  during the normal course of care

Raloxifene
or
Bisphosphonates (alendronate, risedronate,
etidronate)
VISIT 1 Baseline
VISIT 2 6 months
VISIT 3 12 months
51
Patient Baseline Characteristics
Characteristic Raloxifene N 707 Bisphosphonates N 277
Age (years), Mean SD (range) 66.9 8.5 (55-97) 67.7 9.2 (55-91)
Race/ethnicity, n () Chinese Malay Filipino Indian sub-continent Asian other 364 (51.5) 6 (0.8) 186 (26.3) 144 (20.4) 7 (1.0) 120 (43.3) 3 (1.1) 79 (28.5) 69 (24.9) 6 (2.2)
Menopause, n () Natural Surgical 613 (86.7) 94 (13.3) 237 (85.6) 40 (14.4)
No. years since menopause, Mean SD 17.6 9.9 19.2 10.3
Baseline fractures, n () 307 (43.4) 116 (41.9)
plt0.05 ANOVA
52
Adherence Patient Discontinuations
Enrolled N 984
Raloxifene treatment n 707
Bisphosphonates treatment n 277
Alendronate n 206
Risedronate n 71
33.1 6.8 5.7 2.8 1.4
37.5 2.9 10.1 9.7 2.2
Lost to follow up Chose to leave Stopped
treatment Switched treatment Reason missing
plt0.05 plt0.01 Fishers Exact Test
Completed study 50.2
Completed study 37.5
Poster presented at IBMS-ECTS Geneva , June 2005,
Bone 36, Suppl 2, 2005
53
Adherence Treatment Duration (days)
Raloxifene
Bisphosphonates
plt0.05
0
100
200
300
400
500
Mean Period of Exposure to Medication (days)
Wilcoxon rank sum test
Poster presented at IBMS-ECTS Geneva , June 2005,
Bone 36, Suppl 2, 2005
54
Treatment Satisfaction at 12 months
100
100
plt0.01
plt0.01
80
80
60
60
Percent of patients
Percent of patients
40
40
20
20
no effect
satisfied
feel better
feel worse
no opinion
dissatisfied
Raloxifene
Bisphosphonates
Fishers Exact Test
Poster presented at IBMS-ECTS Geneva , June 2005,
Bone 36, Suppl 2, 2005
55
Quality of Life Mean Change in Health State
(VAS) Score
plt0.01
Mean Change from Baseline to Endpoint
Raloxifene
Bisphosphonates
Health State score was out of 100
ANCOVA
Poster presented at IBMS-ECTS Geneva , June 2005,
Bone 36, Suppl 2, 2005
56
Incidence of New Fractures
6
p0.058
Raloxifene
5
Bisphosphonates
plt0.01
4
Fishers Exact Test
Percentage of of new, self-reported fractures
3
2
1
0
total
wrist
spine
humerus
other
Fracture site
There were no new fractures of the clavicle and
pelvis for raloxifene or bisphosphonates