Transcript Presentation

Lebanese Society of Endocrinology Diabetes and Lipids
in Collaboration with OSTEOS
Sponsored by KFF
The Long Term Use of Bisphosphonates
in the Treatment of Osteoporosis
September 8, 2016
Ghada El-Hajj Fuleihan, MD, MPH, FRCP
Calcium Metabolism and Osteoporosis Program
WHO Collaborating Center for Metabolic Bone Diseases
American University of Beirut Medical Center
Beirut, Lebanon
Outline
 Background
 NOGG and AACE Guidance
 ASBMR Task Force Report on Long Term Use of BPs
 Applicability to GIOP and OP in men
 Limitations
 Case studies
Outline
 Background
 NOGG and AACE Guidance
 ASBMR Task Force Report on Long Term Use of BPs
 Applicability to GIOP and OP in men
 Limitations
 Case studies
Background
 Bisphosphonates (BPs) have dominated the landscape of
osteoporosis therapies for the last two decades
 Increasing evidence for an effect of prolonged BP use on rare
adverse events (AEs), namely Atypical Femoral Fractures (AFF)
and Osteonecrosis of the Jaw (ONJ), have raised serious
concerns, as addressed by the ASBMR1,2,3
 The long term retention of BPs in bone, albeit with a
differential temporal profile, and the AEs, led to the concept of
drug holiday, to maximize benefits and minimize harms
1. Shane E. et al. ASBMR AFF Task Force Report JBMR 2010;25 (11):22672. Khosla S. et al ASBMR ONJ Task Force Report JBMR 2007;22 (10):14793. Shane E. et al. ASBMR AFF 2nd Task Force Report JBMR 2013;
Why Stop Osteoporosis Rx?
 Rx not working (PINP @ 3/12, BMD @ 2-3y)
 Adverse effects (GI, flu-like illness)
 Rx no longer necessary:
– fracture risk ↓↓
– Persistent drug effect
 Concern re future adverse events (ONJ, atypical
subtrochanteric fractures)
Courtesy I Reid.
Background
 Fracture reduction for vertebral, non-vertebral and hip fractures has been
established for ALN, RIS, and ZLN, and hip fracture was a primary outcome only
for the RIS and ZLN trials.
 The long-term efficacy of these BPs in extension studies is primarily based on
trials conducted in a sub-set of trial participants and focused primarily on bone
density changes. Fracture reduction was evaluated as secondary or an exploratory
outcome.
 Ibandronate was not studied beyond 5 years, and the extension study for
risedronate had no placebo group and only included a small number of subjects
followed for up to 7 years (N=74)
Mellström DD, et al. Seven years of treatment with risedronate in women with postmenopausal osteoporosis. Calcif Tissue Int. 2004;75(6):462
Black DM, et al. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet. 1996;348:1535
Black DM,et al. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med. 2007;356:1809
Chesnut III CH, et al. Effects of oral IBN administered daily or intermittently on fracture risk in postmenopausal osteoporosis. JBMR 2004;19(8):1241-9.
Harris ST, et al. A randomized controlled trial Vertebral Efficacy With Risedronate Therapy (VERT) Study Group. JAMA. 1996;282(14):1344
Liberman UA, et al. Effect of oral ALN on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. NEJM 1995;333(22):1437
McClung MR, et al.Effect of risedronate on the risk of hip fracture in elderly women. N Engl J Med. 2001;344(5):333
Reginster J, et al. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. Osteoporos Int.2000;11(1):83-91.
 Based on BTMs, the beneficial effects of ALN persist for 2–3
years and possibly 1-2 years for ibandronate and risedronate.
 In the case of three years of ZOL therapy, it extends for at
least another three years.
 These findings are consistent with the relative binding
affinities of BPs for hydroxyapatite and human bone tissue.
Russel RG, Watts NB, Ebetino FH, Rogers MJ. Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on
clinical efficacy. Osteoporos Int. 2008;19(6):733-59.
ONJ and AFF

ONJ: First described over 160 years ago, but association with BP
intake was first reported by dentists/oral surgeons in 2003. It is much
more common in cancer patients receiving higher cumulative doses
of BPs than in patients with osteoporosis treated with lower doses.
ONJ incidence in OP is estimated to be between 1/10,000 and
1/100,000, and is only slightly higher than the ONJ incidence in the
general population

AFFs: First reported in 2005. AFFs can occur in patients not receiving
any anti-fracture medications; they account for about 1% of all
femoral fractures and about 3% of all femoral shaft fractures. AFF
incidence is estimated between 3.2-50 cases/100,000 person-years,
an estimate that doubles with prolonged duration of BP use (> 3
years, median duration 7 years), and seems to decline with
discontinuation.
ONJ and AFF

This led the FDA to request information from all BP drug manufacturers
regarding this potential safety signal and to assess long-term efficacy. On
October 13, 2010, it determined that new "Warnings and Precautions"
information regarding the risk of AFFs should be added to the labels of all BP
products approved for the prevention or treatment of osteoporosis.

FDA held a hearing Sept 2011 to review the long-term safety and efficacy of
BPs, and subsequently recommended that physicians re-assess the
indication for continued BP therapy beyond 3-5 years, but noted that in high
risk patients, drug discontinuation may not be advisable. Currently, all FDA
approvals of BPs for the treatment of osteoporosis contain the following
“Important Limitation of Use” statement: “The optimal duration of use has
not been determined. All patients on BP therapy should have the need for
continued therapy re-evaluated on a periodic basis.”
•
FDA Drug Safety Communication. Ongoing safety review of oral bisphosphonates and atypical sub trochanteric femur fractures. 2011 [Accessed May 1, 2014].
http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm203891.htm.
Whitaker M, Guo J, Kehoe T, Benson G. Bisphosphonates for osteoporosis-where do we go from here? N Engl J Med. 2012;366(22):2048-51.
FDA panel: Rubin R. Osteoporosis drugs need better labels. Time limits on the drugs are suggested, but how much time is yet to be determined.
http://www.webmd.com/osteoporosis/news/20110909/fda-panel-unclear-on-osteoporosis-drug-labels.
•
•
Shane et al. J Bone Miner Res. 2014 Jan;29(1):1-23.
Atypical Sub-Trochanteric Fractures
Kwek, Injury 39:224, 2008
ASBMR Task Force Case Definition of AFFs
Shane et al. J Bone Miner Res. 2014 Jan;29(1):1-23.
ONJ
Exposed bone in the mouth
present for >6 or 8 weeks
Outline
 Background
 AACE and NOGG
 ASBMR Task Force Report on Long Term Use of BPs
 Limitations
 Case studies
Guidelines for Clinical Practice for the Diagnosis and
Treatment of Postmenopausal Osteoporosis
 How long should patients be treated?
• For BPs if osteoporosis is mild, consider a “drug holiday” after 4
to 5 years of stability
• If fracture risk is high, consider a drug holiday of 1- 2 years
after 10 years of treatment
• Follow BMD and bone turnover markers during a drug holiday
period and reinitiate therapy if bone density declines
substantially, bone turnover markers increase, or a fracture
occurs
Watts et al. Endoc Pract 2010; 16(6):1016-9
Bisphosphonates: Algorithm for Long-term Treatment Monitoring
Advise 3-5 yrs* treatment
(Follow-up at 3/12 to
discuss treatment issues)
*3 yrs for zoledronic acid
5 yrs for other BPs
No fracture
Recurrent fracture(s)
Prevalent vertebral fracture(s)**
**In patients
taking oral BPs,
consider
continuation if:
• age > 75 yrs,
• previous hip
fracture
• current oral GC
therapy
≥ 7.5 mg/d
prednisolone
FRAX + BMD
after 3-5* years
Above NOGG intervention
threshold or hip BMD
T-score ≤-2.5
Below NOGG intervention
threshold and hip BMD
T-score >-2.5
Check adherence
Exclude 2°causes Re-evaluate
treatment Choice Continue
treatment
Consider drug holiday
Repeat FRAX + BMD
in 1.5-3 yrs
Compston et al Maturitas 2013
* Co-Primary authors
Nelson Watts Consultant to the Task Force
Adler R, El-Hajj Fuleihan G, et al JBMR 2016; 31(1):16-35
Copyright © 1999-2016 John Wiley & Sons, Inc. All Rights Reserved.
Task Force Charges
 Provide guidance on duration of BP therapy in patients with
postmenopausal osteoporosis, developing an algorithm that
incorporates risk assessment (efficacy).
 Determine how potential harms may affect duration of
therapy (safety), with a risk benefit perspective.
 Discuss how the algorithm and approach may apply to men
and to glucocorticoid-induced osteoporosis.
BP Long Term Randomized Trials
Extension Studies
FLEX
HORIZON EXTENSION
Design of Fracture Intervention Trial (FIT) LongTerm Extension (FLEX) Trial
6459 Participants
randomized
FIT (3-4.5 yrs)
3223 Assigned to receive PBO
Post-FIT open label (1-2 yrs)
3236 Assigned to receive ALN
2857 Eligible for FLEX screening
FLEX (5 yrs) PBO or ALN 5/10 mg
1099 FLEX
Participants
Randomized
437 Assigned to PBO
662 Assigned to ALN 5 or 10 mg
McNabb B et al JBMR 2013; 28:1319
Bauer D et al. JAMA Intern Med. 2014;174:1126
FLEX Extension Study
CHARACTERISTICS
N Core
6459
N extension
1099
Age
73 ± 5.7 yrs
Ethnicity
96% Caucasian
T-Score T hip -1.9
T-Score FN
-2.2
% Prevalent VFx 34%
% Clinical Fxs 60%
Ca/D
500 mg/250 IU
OUTCOMES
• 1 Outcome
• 2 Outcomes
• Exploratory
FN BMD
BMD other sites
BTMs
Fractures
Design of HORIZON EXTENSION
7736 Participants
randomized to Tmt
ZLN or PBO for 3 yrs
3876 Received PBO
One YR Post-Horizon
3867 received ZLN 5mg (0,12,24 months)
ZLN ARM eligible for screening
EXTENSION ZLN or PBO for 3 years
1233 Participants
Randomized
617 to PBO- Z3P3
616 ZLN 5 mg/YR for 3 YRS Z6
Black D et al. NEJM 2007; 356:1809
Black D et al .JAMA 2012;174(7):1126
Cosman F et al JCEM 2014 in press
HORIZON Extension Study
CHARACTERISTICS
N original
7736
N extension:
1233/1040
Age
75 ± 5.5 yrs
Ethnicity
80% white/16 % South Am, 4% Asia
T-score FN
-2.55
Prevalent VFx 60%
Ca/D
1000-1200 mg /400-1200IU
OUTCOMES
• 1 Outcome
• 2 Outcomes
FN BMD
BMD other sites
BTMs
Fractures
Safety
Incidence of Fracture by Treatment Group
FLEX Extension. N= 1099
Fractures
Placebo, No. (%)
(n = 437)
Pooled ALN, No. (%)
(n = 662)
Relative Risk (95% CI)*
Clinical
23 (5.3)
16 (2.4)
0.45 (0.24-0.85)
Morphometric
46 (11.3)
60 (9.8)
0.86 (0.60-1.22)
Any
93 (21.3)
132 (19.9)
0.93 (0.71-1.21)
Nonspine
83 (19.0)
125 (18.9)
1.00 (0.76-1.32)
Hip
13 (3.0)
20 (3.0)
1.02 (0.51-2.10)
Forearm
19 (4.3)
31 (4.7)
1.09 (0.62-1.96)
Vertebral
Clinical
*Adjusted for clinic and stratum.
Black et al. JAMA 2006; 296(24):2927-2938
15-
Morphometric Vertebral Fracture
Core PBO
10.9 %
10-
Odds Ratio= 0.51 [0.26, 0.95)
Z3P3
6.2 %
5-
Z6
3.0 %
Core ZOL
3.3 %
0
Extension Study (Year 3-6)
n= Fractures/Total
30/486
14/469
Patients with Non-Vertebral Fracture (%)
Patients with New Vertebral Fracture (%)
Incidence of Fractures by Treatment in the
Horizon Extension
15-
Non-Vertebral Fracture
Hazard Ratio= 0.99 [0.7, 1.5)
Core PBO
10.7 %
10-Core ZOL
8.0 %
Z3P3
7.6 %
Z6
8.2%
50
Extension Study (Year 3-6)
47/617
45/616
Patients with Hip Fracture (%)
15Hip Fracture
Relative Hazard= 0.90 [0.3, 2.5)
105Core PBO
2.5 %
0
Core ZOL
3.3 %
n= Fractures/Total
Z3P3
1.4 %
Z6
1.3 %
Extension Study (Year 3-6)
8/617
7/616
Black et al. J Bone Miner Res. 2012 Feb;27(2):243-54
Summary of FLEX and Horizon Extension Studies
LONG TERM BP vs SWITCH to PBO
 FLEX demonstrates that ALN 10 yrs
– Maintained BMD at all sites versus loss in PBO, p<0.001
– Reduced risk of clinical vertebral fractures: RR=0.45 (0.24-0.85)
 HORIZON extension demonstrates that ZLN for 6 yrs
– Maintained BMD at all sites versus loss in PBO, p<0.001
– Reduced risk of morphometric vertebral fractures: RR=0.51
(0.26,0.95).
Black et al. JAMA 2006; 296:2927
Black et al. JBMR 2012; 27:240
Reductions (RR) for Fractures for Continuing
BPs ALN and ZOL
3
Relative Hazard (± 95% CI)
Favors Bisphosphonate
Favors Placebo
Courtesy
of Dennis
Black JAMA
2006;Black
et a. Black,
JBMR Ph.D.
2012
BP Long Term Randomized Trials
Extension Studies
Identify high risk subjects when therapy is stopped
PBO ARM
Fracture Incidence by Baseline FN T-Score
and Prevalent Vertebral Fracture*
Non-vertebral Fractures
Subgroup
No.
Placebo
No. (%)
Clinical Vertebral Fractures
Placebo
No. (%)
Baseline BMD T-score at FN
> −2.0
-2≤ x <2.5
≤−2.5
461
311
18 (10.1)
26 (20.6)
3 (1.7)
9 (7.1)
322
39 (29.5)
11 (8.3)
Prevalent vertebral fracture
No
723
48 (16.7)
11 (3.8)
Yes
376
35 (23.3)
12 (8.0)
Black et al. JAMA 2006; 296(24):2927-2938
FLEX Age-Adjusted Risk of Any Clinical Fracture
After Discontinuing ALN
Variables Measured at FLEX Baselinea
Risk of Fracture, Relative Hazard Ratio (95% CI)b
Demographic and clinical characteristics
Age, per 5-y increase, y
1.54 (1.26-1.85)
BMI, per SD increase
1.10 (0.87-1.38)
Prevalent Vertebral fracture
1.11 (0.71-1.75)
Previous non-spine fracture
1.24 (0.64-2.40)
BMDT-score, lowest tertile vs other 2
Total hip
1.87 (1.20-2.92)
Femoral neck
2.17 (1.38-3.41)
BTMs, highest tertile vs other 2
NTX/Cr, nmol/mmol
1.33 (0.84-2.10)
BAP, ng/mL
1.39 (0.89-2.17)
Abbreviations: BAP, bone-specific alkaline phosphatase; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared);
FLEX, Fracture Intervention Trial Long-term Extension; NTX/Cr, type 1 collagen cross-linked N-telopeptide to creatinine concentration ratio.
a With the exception of age, each variable was examined in a separate age-adjusted model.
b Any non-spine or clinical vertebral fracture occurring after the 1-year visit.
Bauer et al. JAMA Intern Med. 2014;174(7):1126-1134
Proportion of Women With Any Clinical Fracture after
ALN Discontinuation
P for trend<0.01
P for trend NS
P for trend NS
P for trend NS
Bauer et al. JAMA Intern Med. 2014;174(7):1126-1134
Predictors of Morphometric Vertebral Fracture in
Discontinuation Group (Z3P3) During Extension Study
(univariable models)
Factors at Extension Baseline
Odd Ratio (95% CI)
P-Value
≥ 75 years
1.33 (0.62, 2.86)
0.461
FN T-score as categorical (~55% of
population)
≤ -2.5
3.32 (1.37, 8.05)
0.008
TH T-score as categorical (~25% of
population)
≤ -2.5
3.99 (1.79, 8.92)
0.0007
Yes
4.75 (1.35,
16.77)
0.015
Age
Factors During Core Trial
Incident Vertebral Fracture
Cosman F et al. JCEM 2014. in press
Predictors of Increased Fracture Risk
OFF Therapy
 Age: for any clinical fracture, RHR 1.53 (1.26-1.85)/ 5 years
increments, in FLEX, but not HORIZON ( age cut off 75 yr)
 FN or Hip T-score ≤ -2.5
 Clinical VFx risk: FLEX 2 fold increase
 Morphometric VFx risk: HORIZON 3-4 fold increase
 Incident Fractures during Core study in HORIZON
 Morph VFx increased risk of incident morph VFx 5 fold
 Non- VFx increased risk of incident Non-VFx 2.5 fold
 Risk Factors: in HORIZON increased number of risk factors (low
T-score , prevalent fractures, incident fractures) increased
fracture risk.
BP long Term Extension Trials
FLEX
HORIZON EXTENSION
Risk Stratification BP vs PBO
Who are patients to benefit from continued therapy
Effect of ALN on Fracture Risk by Sub-bgroups of
Baseline FN T-Score and Prevalent Vertebral Fracture*
Non-vertebral Fractures
Subgroup
No.
Placebo
No. (%)
Clinical Vertebral Fractures
Alendronate
No. (%)
RR (95%CI)
Placebo
No. (%)
Alendronate
No. (%)
RR
(95% CI)
Baseline BMD T-score at FN
> −2.0
461
18(10.1)
42 (14.9)
1.5 (0.86-2.6)
3 (1.7)
4 (1.4)
0.84 (0.18-4.2)
>−2.5 to ≤−2.0
311
26 (20.6)
38 (20.5)
1.0 (0.63-1.7)
9 (7.1)
3 (1.6)
0.22 (0.05-0.74)
≤−2.5
322
39 (29.5)
43 (22.6)
0.77 (0.50-1.2)
11 (8.3)
9 (4.7)
0.57 (0.23-1.40)
P value for interaction†
.40
.72
Prevalent vertebral fracture
No
723
48 (16.7)
61 (14.1)
0.86 (0.59-1.3)
11 (3.8)
7 (1.6)
0.42 (0.16-1.1)
Yes
376
35 (23.3)
62 (27.7)
1.20 (0.80-1.8)
12 (8.0)
9 (4.0)
0.47 (0.19-1.1)
P value for interaction‡
.23
.86
Abbreviations: BMD, bone mineral density; CI, confidence interval; RR, relative risk.
*Analyses of RR and assessment of interaction were done with unadjusted proportional hazards models. Parallel analyses of morphometricvertebral fracture did not
show any significant trends for alendronate efficacy among subgroups.
†Interaction between BMD as a continuous variable and treatment.
‡Interaction between prevalent vertebral fracture status and treatment.
Black et al. JAMA 2006; 296(24):2927-2938
Effect of Continued ALN Treatment and Risk of Fracture
Stratified by Baseline Presence of Vertebral Fracture and FN T-Score
Femoral neck T-score
at FLEX baseline
Non-vertebral
No.
Morphometric vertebral
PBO
No. (%)a
ALN
No. (%)b
RRc
(95%CI)
PBO
No. (%)a
ALN
No. (%)b
RRd
(95% CI)
No VF at FLEX
baseline
-2< FLEX FN T-score
333
14 (10.8)
30 (14.8)
1.41 (0.75–2.66)
7 (5.7)
9 (4.8)
0.84 (0.30–2.31)
-2.5<FLEX FN T-score ≤ -2
203
13 (15.9)
15 (12.4)
0.79 (0.37–1.66)
6 (7.6)
6 (5.4)
0.69 (0.21–2.22)
FLEX FN T-score ≤2.5
184
21 (28.0)
16 (14.7)
0.50 (0.26–0.96)
8 (11.0)
8 (7.7)
0.68 (0.24–1.90)
P value for interactione
.019
.92
Vertebral fracture at FLEX baseline
-2< FLEX FN T-score
128
4 (8.2)
12 (15.2)
1.68 (0.54–5.21)
2 (4.7)
9 (11.5)
2.67 (0.55–12.98)
2.5<FLEX FN T-score2
108
13 (29.5)
23 (35.9)
1.32 (0.67–2.61)
9 (23.1)
11 (17.7)
0.72 (0.27–1.93)
FLEX FN T-score ≤2.5
138
18 (31.6)
27 (33.3)
1.11 (0.61–2.02)
14 (27.5)
17 (25.4)
0.90 (0.39–2.05)
P value for interactione
.60
.96
aNumber
of participants with at least one fracture in the FLEX placebo group.
of participants with at least one fracture in the FLEX ALN group (5 and 10 mg/day combined).
cRelative hazard estimated with Cox proportional hazard models for time to first fracture.
dOdds ratio estimated with logistic regression models.
ep Values for tests of interaction. Relative risks were tested for multiplicative interaction with FN T-score as a continuous variable.Appendix IV-B
bNumber
Schwartz et al. JBMR 2010; 25(5):976–982
Morphometric Vertebral Fracture Incidence and Odds Ratio by
Treatment in Risk-Stratified Subgroups
Stratification by TH BMD: Incident Morphometric Vertebral Fracture Rates
Z3P3
Z6
Treatment subgroup interactions not significant
14.3 %
(16/112)
Proportion of Patients (%)
10-
5-
OR:0.26
OR:0.68
(95% CI: 0.08, 0.69)
P= 0.0113
(95% CI: 0.28, 1.58)
P= 0.3793
NNT= 9.9
NNT= 86.2
4.2 %
(5/120)
3.8 %
(14/373)
2.6 %
(9/347)
0TH BMD ≤ -2.5
TH BMD > -2.5
BMD, bone mineral density; CI, confidence interval; TH, total hip; NNT, number needed to treat; OR, odds ratio
Cosman F et al. JCEM 2014. in press
Stratification by FN BMD: Incident Morphometric Vertebral Fracture Rates
Z3P3
Z6
Treatment subgroup interactions not significant
15-
Proportion of Patients (%)
10-
9.2 %
(23/250)
OR:0.36
(95% CI: 0.15, 0.77)
P= 0.01
NNT= 18
5-
3.5 %
(9/257)
OR:0.79
(95% CI: 0.23, 2.53)
P= 0.70
NNT= 167
3.0 %
(7/235)
2.4 %
(5/210)
0
FN BMD ≤ -2.5
FN BMD > -2.5
BMD, bone mineral density; CI, confidence interval; FN, femoral neck; NNT, number needed to treat; OR, odds ratio
Cosman F et al. JCEM 2014. in press
Stratification by Incident Vertebral Fracture During Core Study:
Incident Morphometric Vertebral Fracture Rates
Z3P3
25-
25%
(4/16)
Treatment subgroup interactions not significant
Proportion of Patients (%)
20P= 0.12
15-
10-
Z6
OR:0.46
(95% CI: 0.22, 0.90)
P= 0.03
NNT= 34
NNT= 4
NA
(0/11)
5.6 %
(26/467)
5-
2.6 %
(12/454)
0
Incident vertebral fracture During Core
Non Incident vertebral fracture During Core
CI, confidence interval; NNT, number needed to treat; OR, odds ratio
Differences between Z6 and Z3P3 groups were analyzed during Fisher’s exact test for categorical variables
Cosman F et al. JCEM 2014. in press
Change in Femoral Neck BMD at Year 6
Relative to Core Baseline (ITT)
Placebo
Z6
Z3P3
Change from Baseline (%)
5
1.36%
+4.5%
4
0.47%NS
(-0.15, 1.10)
3
(0.58, 2.15)
P = 0.0007
+3.1%
2
1
Start of extension
0
–1
–2
Core Study
–3
0
1
Extension Study
2
3
4
5
6
Time (Years from Core Study Baseline)
ITT, intention to treat; NS, not significant
Black DM, et al. 3rd joint ECTS/IBMS Meeting. Athens, Greece: May 2011; Poster
Incidence of New Morphometric Vertebral
Fractures with 6 vs. 3 Years Zoledronic Acid (ITT)
Placebo
Proportion of Patients (%)
15
10.9%
(310/2853)
ZOL Core study (Yr 0–3)
Z3P3
70%†
(95% CI: 62%, 76%)
P < 0.001
10
6.2%
(30/486)
5
Z6
3.3%
(92/2822)
52%*
(10%, 74%)
P = 0.0348
3.0%
(14/469)
0
Core Study1
Extension Study2
*RRR vs. Z3P3;†RRR vs. placebo; CI, confidence interval; ITT, intention to treat; RRR, relative risk reduction
1. Black DM, et al. N Engl J Med. 2007;356:1809–1822; 2. Black DM, et al. 3rd joint ECTS/IBMS Meeting. Athens, Greece: May 2011; Poster
Absolute Risk of Morphometric Vertebral Fracture in Subgroups Defined by
Combining Risk Predictors by Increasing Level of Risk in Z3P3 Subgroups and in
Same Z6 Subgroups with NNT to Prevent 1 Morphometric Vertebral Fracture
At Extension Baseline
During Core
Incident
Vertebral
Fracture
Incident
Nonvertebral
Fracture
FN T-score
≤-2.5
TH T-score
≤ -2.5
Prevalent
Vertebral
Fracture
Yes
No
Yes
No
Yes
Yes
No
Yes
Yes
Yes
Yes
%
(N)
Z3P3 % fx
(n/N)
Z6 % fx
(n/N)
NNT
No
14.7 %
(139)
7.7 %
(5/65)
2.7 %
(2/74)
20
No
No
7.1 %
(67)
11.5 %
(3/26)
2.4 %
(1/41)
11
Yes
No
No
12.2 %
(115)
12.9 %
(8/62)
5.7 %
(3/53)
14
Yes
Yes
No
Yes
1.2 %
(11)
16.7 %
(1/6)
0%
(0/5)
6
Yes
Yes
Yes
No
1.2 %
(11)
66.7 %
(4/6)
0%
(0/5)
2
Cosman F et al. JCEM 2014. in press
Summary of FLEX and HORIZON Extension
Studies Continued BP vs PBO
• Continued ALN for 10 yrs reduced risk of clinical fractures1
 If FN BMD T-score ˃ -2.5 to ≤-2 RR=0.22 [0.05-0.74]
• Continued ZLN for 6 yrs reduced risk of morphometric VFx2
 If Total hip T-score ≤ -2.5, RR=0.26 [0.08-0.69]
 If FN BMD T-score ≤ -2.5, RR=0.36 [0.15-0.77]
• In HORIZON subjects with incident morphometric VFx during core
study benefited most from continued therapy2
• In low risk subjects, vertebral fracture rates were low and there was
no increased risk of non vertebral fractures after therapy
discontinuation: for up to 5 years FU (after 5 yrs of ALN), and for
up to 3 years FU (after 3 yrs ZLN) 2
1. Black et al. JBMR 2006; 296: 2927
2. Cosman F et al JCEM 2014 in press
Risks Associated with Chronic Bisphosphonate Use and Other Health Outcomes
4500
4200
Incidence per 100,000 person-years
4000
3500
3000
2700 2730
2500
2000
1500
1755
1075
1000
500
5
0
21 9
3
113
16
500
325
26
260
169
420
438
284
273
Treated with BPs (10 years)
Treated with BPs (5 years)
Treated with BPs
Risk
GI Bleed
CNS Bleed
Likelihood of suffering fractures and other adverse events in adult patients. The baseline risk of fractures, and for fractures, ONJ, and AFF on BP therapy are presented 1-3. The incidence rates for fractures on
BP therapy (green bars) are based on a conservative 35% risk reduction applied to all baseline fracture rates. The baseline risk for stroke is depicted for in men and is reduced by 20% on aspirin therapy 4. The
risk for GI and CNS bleed on aspirin therapy are presented for men, the risks for women are 52-66% those in men4. For ONJ5 there may be a risk with increased BP duration of use but the quality of the
evidence is poor. For AFF2 the risks represent those reported while on BP for 5 and up to 10 years. Some, but not all studies, suggest the BP-associated risk of AFF may increase substantially, even
quadruple, with up to 10 years of BP use. Rates for Motor Vehicle Accidents (MVA) 5 and Homicides6 are derived from Emergency Department visits and expressed per 100,000.
Osteonecrosis of the jaw (ONJ); atypical femoral fracture (AFF); BP= bisphosphonates; GI = Gastro-intestinal; CNS= Central Nervous System; MVA= Motor Vehicle Accident.
1. Cauley JA, Chalhoub D, Kassem AM, El-Hajj Fuleihan G. Geographic and ethnic disparities in osteoporotic fractures. Nature Reviews. Endocrinology. 2014;10(6):338-351.
2. Dell RM, Adams AL, Greene DF, et al. Incidence of atypical nontraumatic diaphyseal fractures of the femur. Journal of Bone and Mineral Research: 2012;27(12):2544-2550.
3. Tennis P, Rothman KJ, Bohn RL, et al. Incidence of osteonecrosis of the jaw among users of bisphosphonates with selected cancers or osteoporosis. Pharmacoepidemiology and Drug
Safety. 2012;21(8):810-817.
4. Stegeman I, Bossuyt PM, Yu T, Boyd C, Puhan MA. Aspirin for Primary Prevention of Cardiovascular Disease and Cancer. A Benefit and Harm Analysis. PloS one. 2015;10(7):e0127194.
5. Bergen G, Peterson C, Ederer D, et al. Vital signs: health burden and medical costs of nonfatal injuries to motor vehicle occupants - United States, 2012. MMWR Morb Mortal Wkly
Rep. 2014;63(40):894-900.
Adler R, El-Hajj Fuleihan G, et al JBMR 2016; 31(1):16-35
6. Center for Disease Control. Assault or Homicide. 2013; http://www.cdc.gov/nchs/fastats/homicide.htm.
Copyright © 1999-2016 John Wiley & Sons, Inc. All Rights Reserved.
Benefits & Risks of Bisphosphonates
NNT for 3 years:
 Vertebral #
14
 Non-vertebral # 35
 Hip #
90
NNH
 ONJ
>10,000
 Sub-trochanteric >3,000
Courtesy I Reid
Based on data in Black, NEJM, 2010
Benefits & Risks of Bisphosphonates
NNT for 3 years:
 Vertebral #
14
 Non-vertebral # 35
 Hip #
90
Risks/Benefits of 5 yrs of BPS
Fractures averted/100,000:
175 hip, 1470 vertebral,
945 wrist fractures
TOTAL 2590 fractures averted
AFF associated /possibly incurred:
16/100,000
NNH
 ONJ
>10,000
For 162 fractures averted/AFF
 Sub-trochanteric >3,000
possibly incurred
Courtesy I Reid
Based on data in Black, NEJM, 2010
ASBMR TF on long term BP
JBMR 2016
Approach for
Management of
Postmenopausal
Women on
Long Term
Bisphosphonate
Therapy
Post-menopausal women treated with oral (≥ 5yrs) or IV (≥ 3 yrs) BPs
≥ 3-5 years but <10 years
Hip, spine or multiple other osteoporotic fractures before or during therapy
Yes
Reassess benefits/risks
Consider continue BP (1) or change
to alternative therapy (2)
Reassess every 2-3 years
No
Hip BMD T-Score ≤ -2.5 (3)
OR
high fracture risk (4)
Yes
Reassess benefits/risks
Consider continue BP for up to 10 yrs (1) or
change to alternative therapy (2)
Reassess every 2-3 years
No
Consider drug holiday
Reassess every 2-3 years (5)
(1) From the registration trials, the benefits of 5 years of therapy clearly outweigh the risks. For treatment up to 10 years with oral bisphosphonates (FLEX extension) and 6 years with intravenous
bisphosphonates (HORIZON extension), estimates of benefits and risks are based on much weaker data. For patients who fracture on therapy, assess adherence and rule out secondary causes of
osteoporosis. Management of high risk patients is discussed in the text.
(2)The benefits of switching to an alternative anti-fracture therapy after prolonged bisphosphonate treatment have not been adequately studied.
(3) Based on FLEX and Horizon extension study (Caucasian women), may not apply to other populations.
(4) High fracture risk: defined by older age (70-75 yrs), other strong risk factors for fracture, or FRAX fracture risk score that is above country specific thresholds. The use of FRAX in patients on
therapy was only assessed in the Manitoba observational cohort. (1)
(5) Reassessment includes clinical evaluation, risk assessment including risk factors, and may include bone density measurement by DXA. The monitoring interval with DXA should be based upon
changes that are detectable and clinically significant. Reassessment may be necessary at less than 2 years in patients with a new fracture, or in light of anticipated accelerated bone loss (e.g.
institution of aromatase inhibitor or glucocorticoid therapy).
Adler R, El-Hajj Fuleihan G, et al JBMR 2016; 31(1):16-35
Summary
 After 3 years of intravenous ZOL and 5 years of oral ALN
treatment, high risk postmenopausal Caucasian women,
such as those with recent incident or prevalent vertebral
fractures in the HORIZON extension, or with hip T-scores of
≤-2.5 appeared to benefit the most from continued BP
treatment.
 The evidence for this benefit is limited to reducing the risk of
vertebral fractures, and data for other BPs are lacking.
 Furthermore, tools to identify subjects who will fracture
when therapy is discontinued are limited.

Patients treated with risedronate may need earlier reassessment
because of the shorter biologic half-life of this BP.

Repeat DXA or BTM measurements may be considered during this
‘holiday,’ but there are no data to guide the clinician regarding reinstitution of therapy, because neither 1 year change in BMD nor 1 year
change in BTMs predicted fractures post-BP discontinuation.

It would be reasonable to consider withholding therapy as long as
BMD is stable, and to re-start BP therapy (or an alternate osteoporosis
medication) if the BMD T-score is ≤ -2.5, or if other new/additional risk
factors for fractures emerge. However, this approach is based on expert
opinion.

Furthermore, the use of a T-score cut-off of -2.5 for risk stratification
and decision-making regarding therapy discontinuation is based on
studies conducted almost exclusively in community-dwelling,
postmenopausal Caucasian women
0
Tunisia
Philippines
China
Colombia
Australia
Spain
Lebanon
US Black
Jordan
Poland
New Zealand
France
Romania
Netherlands
Turkey
Mexico
Finland
Germany
Italy
Hong Kong
Hungary
Argentina
S' Indian
S Korea
S'pore Chinese
Belgium
Japan
Malta
Taiwan
Canada
UK
Austria
18
US Caucasian
Switzerland
Sweden
Denmark
FRAX Calculated 10 year Overall Fracture Risk
20
Female age 65 years, T-score -2.5, BMI=25 kg/m2
16
14
12
10
8
6
4
2

There are few data estimating the risk of AFF after stopping BPs. Of the 3
large cohort studies, only the Swedish study by Schilcher included
information about the risk of AFF after stopping treatment.

The risk fell by 70%/year since last BP use (odds ratio 0.28, 95% CI: 0.210.38), and the most dramatic reduction in risk occurred after the first year
of discontinuation. ONLY 46 AFF on FU over 3 years after D/c

Effect of Bisphosphonate Discontinuation on ONJ Risk: Because of the longterminal half-life of BPs, the American Dental Association,(75) and the
American Association of Oral and Maxillofacial Surgeons(76) do not
recommend routine discontinuation of BP treatment for osteoporosis in
most patients about to undergo invasive dental procedures. There are no
studies of the incidence of ONJ in patients at different times after
discontinuation of BP treatment for osteoporosis.
Outline
 Background
 NOGG and AACE Guidance
 ASBMR Task Force Report on Long Term Use of BPs
 Limitations
 Applicability to GIOP and OP in men
 Case studies
Limitations of Evidence and Algorithm
 Methodological Issues:
 Studies are underpowered
 Fractures were secondary or exploratory, post-hoc exploratory
nature for many analyses
 Both FLEX and HORIZON extension demonstrate a reduction in
vertebral fractures with continued BP therapy: Morph vs Clinical VFx
 No data for other BPs
 No adequate data on non-vertebral fracture reduction
 Data from FLEX and HORIZON: do not provide evidence to advocate
use of markers today.
Limitations of Evidence and Algorithm
 Algorithm does not cover management of patients on BPs
beyond 10 yrs
 The T-score cut-off of -2.5 is applicable to Caucasian subjects
and may not be applicable to other ethnic groups or
populations
 No adequate data for use of FRAX in patients on therapy
 No trials demonstrating that re-starting BP or switching to
alternate therapies results in fewer fractures
 No fracture data with prolonged BP therapy in men or GIOP
 T-score cut-off of -2.5 needs to be adjusted in GIOP
Bone Turnover Markers
 “Based on the limited evidence available from FLEX and
HOROZON extension studies there is no evidence to support the
measurements of BTMs to assess fracture risk after long term BPs
therapy or in offset periods
 Some experts use BTMs to determine whether a discontinued BP
stiff exerts is still exerting its effect, and resume therapy when
they exceed the lower half of the premenopausal range.
 This approach is based on the evidence that maintenance of
BTMs in the lower half of the premenopausal range is associated
with a lower fracture risk, and that such observations may be
extended to patients who discontinue BPs after long term
therapy.” Vasikaran, Eastell, Bruyere; OI 2011
Outline
 Background
 NOGG and AACE Guidance
 ASBMR Task Force Report on Long Term Use of BPs
 Applicability to GIOP and OP in men
 Limitations
 Case studies
Applicability of Approach to Men and GIOP
 The approach was deemed by TF members to be applicable to
men, although the evidence is much scarcer than in women
 Safety considerations are anticipated to be similar as in women
 The approach may also be applicable to patients with GIOP with
some adaptations ( for eg. raising the T-score cut-off of -2.5)
Knowledge Gaps/Research Agenda
 Validate FRAX in patients on BP therapy
 To identify patients at high fracture risk after therapy discontinuation
 To identify patients who respond best to continued therapy
 Investigate additional predictors to identify high risk
individuals who benefit most from continued BP therapy
 Identify monitoring parameters for patients off therapy
Outline
 Background
 NOGG and AACE Guidance
 ASBMR Task Force Report on Long Term Use of BPs
 Applicability to GIOP and OP in men
 Limitations
 Case studies
Case 1
Hip Fracture after D/C of Long-term BPs
A 71 year old Lebanese woman with history of vertebral fracture,
parental history of hip fracture, has been on BPs for 11 years.
At her last visit 4 months prior, a drug holiday was recommended,
as she had been fracture free for 7 years, with a plan for a clinical
evaluation in one year and BMD assessment in 2-3 years.
At that time, her BMI was 21.5 kg/m2, and her calculated 10-year
risk for major osteoporotic fracture (MOF), using FRAX Lebanon,
was 12%, and for hip fracture 5%.
Case 1
Hip Fracture after D/C of Long-term BPs
She presented to the ER, having sustained an inter-trochanteric
fracture after a fall by tripping on a folded carpet.
BMD scans at the time of therapy discontinuation had revealed a
T-score of -2.3 at the FN, -1.9 at the Thip (NHANES database), 3%
increase at the Thip compared with baseline, and no change
compared with the previous year, on same machine. Evaluation
for causes of secondary osteoporosis was negative.
Case 1
Hip Fracture after D/C of Long-term BPs
Management Approach and Rationale
The patient was discharged on calcium and vitamin D
with the plan to start denosumab therapy on follow-up
in the clinics after evidence of fracture healing
Questions
 Should the patient have been advised drug holiday?
 What therapy should she be switched to and why?
Case 1
Hip Fracture after D/C of Long-term BPs
This case illustrates the limitations of the algorithm given the lack of any data on BP
treatment beyond 10 years in high risk subjects.
 Sequential therapy with another antiresorptive drug, changing from one oral BP
to another, or switching from oral to IV BP, or changing to denosumab has no
supportive evidence base (in terms of fracture reduction).
 Fracture risk is highest within first few years of a MOF, estimated at 14% for a
second hip fracture, 10% for other non-vertebral fractures, and 4% for clinical
vertebral fractures within 2 years1.

One RCT demonstrated that when PM ♀ were on ALN for at least 6 months,
those switched to Dmab had significantly greater increments in bone density at
the hip (+0.85%) and other skeletal sites, compared to patients continued on
alendronate2.
1. Lyles KW et al. N Engl J Med. 2007;357(18):1799-809.
2. Kendler DL et al. J Bone Miner Res. 2010;25(1):72-81.
504 PM women, > 55 years, on ALN for at least 6 months, received
open label ALN then for one months and then
Randomized to continued weekly ALN versus DMAB sc 60 q 6 months
Kendler et al. JBMR 2010;25(1):72-81
Percentage Change from Baseline in Bone Mineral Density in Subjects
Transitioning to Denosumab or Continuing on Alendronate Therapy
Kendler et al. JBMR 2010;25(1):72-81
Percentage Change in Bone Mineral Density at the Total
Hip Lumbar Spine to Prior Alendronate Use Strata
Kendler et al. JBMR 2010;25(1):72-81
Case 1
Hip Fracture after D/C of Long-term BPs
No treatment eliminates fracture occurrence completely.
Decision to switch to a non-bisphosphonate drug in a
patient who had received BPs for 11 years is clearly based
on expert opinion, and aimed at minimizing the risk of AFF
in the future, a risk shown to increase with long term use
of BPs (but also with Dmab).
Dell RM et al. J Bone Miner Res. 2012;27(12):2544-50.
Case 2
Patient on Long-term Glucocorticoid Therapy
 A 62 year old woman who received a left lung transplant for COPD 5 years
prior was seen in the clinic for bone evaluation.
 On cyclosporine, prednisone 5 mg/daily, vitamin D2 50,000 IU every other
week, and calcium 1200 mg daily.
 Started ALN a few weeks after lung transplantation and continues on the
medication without adverse events. She has no hx of fractures.
 DXA scan showed a lumbar spine T score of -2.3 and femoral neck T score
of -2.2. This has remained stable in the past few years. 25 (OH)D is 45
ng/ml and PTH is 56 pg/ml (10-65 pg/ml).
Case 3
Patient with Decreasing BMD Following BP D/C
 75 year old Caucasian woman was treated with ALN for postmenopausal
osteoporosis for ten years, without recognized fractures during treatment. To
minimize her risk of ONJ and AFFs, her ALN was discontinued.
 Her lowest BMD value at the time of discontinuation of therapy was -2.2 at the left
femoral neck site. Follow-up DXA bone density testing after 3 years of drug
holiday showed that the lowest T-score at the femoral neck was now -2.7, with
7.5% loss at this site.
 Similar losses were reported at the LS and right FN. Even though bone loss after
discontinuation of oral BPs1 is not a predictor of fractures off therapy and she had
not fractured, her physician decided to resume therapy with ZOL 5 mg IV once a
year because of her bone loss and low T-score.
 For the last two years on ZOL treatment, she has had no fractures, no new side
effects, and has gained 5.0% at her left femoral neck site.
1. McNabb BL et al. J Bone Miner Res. 2013;28(6):1319-27.
Eastell et al. JBMR 2015;30(3):570-574
Annual Change in Total Hip BMD
Eastell et al. JBMR 2015;30(3):570-574
Case 3
Patient with Decreasing BMD Following BP D/C
Management Approach and Rationale
With this improvement in BMD and with little evidence that
further treatment would be of benefit to decrease fracture
risk1, a second drug holiday would be suggested for this
patient with clinical assessment yearly and a repeat DXA in
2 years.
1. Black DM et al. J Bone Miner Res. 2012;7( 2):243-54.
Case 3
Patient with Decreasing BMD Following BP D/C
This case illustrates the importance of measuring BMD
correctly so that decisions can be based on center
specific least significant change.
Drug holidays in patients who have received long term
BP therapy and do not have fractures, should be
considered but their duration and the best monitoring
parameters off therapy are not clearly defined.
Case 4
Woman with Multiple Risk Factors
 77 year old Caucasian woman with a parental history of hip fracture was
noted to have low BMD (LS T-score of -1.9) on her first DXA scan done at the
onset of menopause. Soon after the initial BMD, an L3 compression fracture
with 50% loss of height was discovered.
 Treatment with ALN was initiated and continued for 9 years. ALN was then
discontinued to minimize the risk for ONJ or AFF.
 A follow up DXA scan showed the lumbar spine T score remained at -1.9; the
mean total hip T-score was -2.3.
 She was referred to an osteoporosis specialist for further management in view
of the fact she remained at high fracture risk. Indeed, she had a BMI of 20
kg/M2, and her 10 year FRAX would be estimated at 35% for MOF and 23 %
for hip fracture, if she were drug naive.
Case 4
Woman with Multiple Risk Factors
 While the spine T-score was unchanged over nine years of BP therapy and
no fracture had occurred, the osteoporosis expert recommended
switching anti-fracture therapy using a non-BP, after discussing the limited
evidence and pros and cons of such an approach with the patient.
 She was begun on teriparatide 20 mcg SC daily, which increased her
lumbar spine BMD by 13.4% and total hip BMD by 6.8% by 24 months.

To preserve the increases in spine and hip BMDs, she received a series of
four ZOL infusions 5.0 mg IV every 12 months.

One year after the fourth infusion, her spine BMD had increased by 3.1%
while her total hip BMD was stable. No new clinical fractures occurred
since her original treatment.
Case 4
Woman with Multiple Risk Factors
Management Approach and Rationale
The risk of subsequent fractures is highest within the first 3-5 years
following fracture incidence, therapy was continued because of her high
fracture risk as assessed above.
The evidence for fracture reduction is only available for vertebral fracture
reduction with oral ALN for up to 10 years and IV ZOL for up to 6 years,
and there is no evidence for non-vertebral fracture risk reduction, the
expert decided to extend treatment beyond the 10 years (13 years BP 2
years TPT) depicted in algorithm (Figure 2), in view of the patient’s
presumed high risk based on low BMI, positive family history, and
calculated FRAX. However, the use of FRAX in patients on therapy has not
been well validated1.
1. Leslie WD et al. J Bone Miner Res.2012;27(6):1243-51.
Case 4
Woman with Multiple Risk Factors
Family history of hip fracture clearly plays an important
role in the fracture risk calculator, FRAX. This patient has
responded well to sequential therapy, although her
management was based on expert opinion because of the
lack of long term studies.
Case 5
Woman on Aromatase Inhibitor Therapy
 63 year old Caucasian woman with a history of osteoporosis was
diagnosed with Stage IIA breast cancer.
 She received breast-conserving surgery with lumpectomy and
sentinel lymph node biopsy.
 T2N0 and strongly expressed both estrogen receptor and
progesterone receptor but was HER2 negative.
 Therapy with an aromatase inhibitor (AI) was recommended for 5
years starting upon completion of adjuvant radiation therapy.
Case 5
Woman on Aromatase Inhibitor Therapy
 3 years prior to the diagnosis of breast cancer the patient had been
diagnosed with osteoporosis and had started weekly ALN.
 Plans were made to continue the ALN while initiating the AI. Calcium,
vitamin D and weight bearing exercise counseling were provided.
 FN T-score was -2.7 at baseline, -2.6 at 3 years of ALN and starts of AI, and
-2.2 at 5 years ALN and 2 years AI.
 Comparison of the change in bone density in g/cm2 between baseline and
5 years of ALN confirmed that it exceeded the least significant change and
was statistically significant. Is more than 5 years of alendronate therapy
appropriate for this patient who is planning to continue an AI for another 3
years?
Case 5
Woman on Aromatase Inhibitor Therapy
Management Approach and Rationale
 AIs are standard endocrine therapy for PM ♀. However AIs cause
significant bone loss.
 In the pivotal ATAC trial, subjects randomized to the AI
anastrazole experienced a BMD decrease of 6% in the spine and
7% in the hip after 5 years of treatment1,2 .
 Similar trials that studied letrazole and exemestane, two other
AIs, reported similar results3. As expected, clinical trials
comparing AIs to tamoxifen have reported an increased fracture
risk with AIs4-6.
1 Howell
A et al. Lancet 2005;365(9453):60-2.
Eastell R et al. J Clin Oncol 2008;26(7):1051-7.
3 Coleman RE et al. Lancet Oncol 2007;8(2):119-27.
4 Rabaglio M et al. Ann Oncol. 2009;20(9):1489-98.
5 Forbes JF et al. Lancet Oncol 2008;9(1):45-53.
6 Coates AS et al. J Clin Oncol.2007;25(3):320-3.
2
Case 5
Woman on Aromatase Inhibitor Therapy
 In the 36-month ZO-FAST study, postmenopausal women with early stage
hormone receptor positive breast cancer receiving letrazole were
randomized to either immediate ZOL 4 mg intravenously every 6 months or
delayed ZOL that was administered if T-scores fell below -2.0 or if a fracture
occurred during the study1,2.
 Subjects in the immediate group had an average 1.6% increase in BMD at
the total hip while subjects randomized to delayed treatment had a 4.2%
loss of BMD at the total hip. A small number (0.4%) of subjects in the
immediate ZOL group and none in delayed group were diagnosed with ONJ.
However, this increase was not statistically significant.
 The SABRE trial studied the effects of risedronate on bone mass in patients
receiving anastrazole. Subjects randomized to risedronate experienced a
1.8% increase in total hip BMD, while control subjects had a 1.1% decrease
after 24 months2.
Eidtmann Het al. Ann Oncol.2010;21(11):2188-94.
1
2
Van Poznak C et al. J Clin Oncol. 2010;28(6):967-75.
Case 5
Woman on Aromatase Inhibitor Therapy
 A UK Expert Group was convened in 2008 to provide
recommendations on the treatment of AI-related bone loss1.
 This panel recommended BP treatment in the following patients:
• Age equal to or greater than 75
• LS or hip T-score of < -2.0 at the initiation of AI therapy
• known vertebral fracture at the initiation of AI therapy
• Annual rate of bone loss greater than 4% at the lumbar spine
or total hip
• T-score equal to or less than -2.0 after 24 months of AI
therapy.
1Reid
DM et al. Cancer Treat Rev. 2008;34(Suppl 1):S3-18.
Case 5
Woman on Aromatase Inhibitor Therapy
 A Task Force convened by the National Comprehensive Cancer
Network recommended using the FRAX calculation tool, and
that AI treatment to be considered a risk factor for “secondary
osteoporosis,” a selection choice in the FRAX calculator1.
 The panel recommended osteoporosis treatment for a hip or LS
T-score less than -2.0 or when the FRAX-calculated 10-year risk
is greater than 3% for the hip or greater than 20%, using the US
FRAX Calculator, for any major osteoporotic fracture. The
consultants also suggested consideration of osteoporosis
treatment for patients with a hip or LS T-score between -1.5 and
-2.0.
1Gralow
JR et al. J Natl Compr Canc Netw. 2013; Aug (Suppl 3):S1-50; quiz S1.
Case 5
Woman on Aromatase Inhibitor Therapy
 In the case presented here, the patient’s FN T-score at the completion of 5 years
of ALN increased from -2.7 to -2.3.
 The current FN T-score is better than -2.5, below the threshold for treatment
recommended by this Task Force if one just considered the T-score threshold.
 However, another criterion considered by the Task Force is high fracture risk that
incorporates strong risk factors for fractures (including AI use).
 Thus, given the anticipated high risk for bone loss in this patent, the ample
evidence that AIs have adverse effects on bone, and the recommendations from
other panels, a BP would be recommended for an additional three years in this
patient while she remains on an AI.
Case 5
Woman on Aromatase Inhibitor Therapy
Intervention thresholds differ between organizations, and at
each assessment, fracture risk must be determined by the
current status of the patient plus the potential effect of
further AI treatment.
Case 6
Male Case

A 79 year old Caucasian man was referred for evaluation of osteoporosis after he suffered
an ankle fracture. Evaluation revealed he had fractured his left hip 3 years prior after
stepping off a curb; he had received no evaluation or treatment at that time.

Comorbidities included hypertension, chronic kidney disease stage 3, neuropathy that
affected his balance and gastro-esophageal reflux disease that was controlled with
omeprazole. BMD at that time showed T-scores at the spine of -1.4, total hip of -2.0, FN of
-3.0, and distal 1/3 radius of -3.3, all compared with a male normative database. His
estimated GFR was 44 mL/min, and he was started on risedronate.

Follow-up BMD performed 2.5 years later revealed stable readings. A subsequent DXA
performed at age 84 showed essentially the same BMD (done on the same machine by the
same technologist with adequate quality control).

His estimated creatinine clearance had dropped to 28 mL/min (Cockcroft-Gault equation),
and he remained at risk for falling because of his neuropathy. His FN T-score was -2.9 by
the male normative database and -2.7 by the white female normative database. If he were
an untreated patient, his FRAX 10 year MOF risk would be 18% and 10 year hip fracture
risk 8.6%. His daughter accompanied him to clinic and raised the question of whether he
would still require treatment for his skeletal health.
Case 6
Male Case
 This is a patient who remains at high risk for fracture based on his prior
fracture, FN T-score, and poor balance due to neuropathy.
 There has been no improvement in this patient’s neuropathy, so he
remains a fall risk. He has had a minimal response to risedronate based on
the fact that his BMD has not decreased in 5 years. If he continues
risedronate, he would be below the recommended estimated creatinine
clearance specified in the package insert.
 Thus, in light of the fact that he does not have recurrent or latent
infections, the decision was made change to Dmab 60 mg subcutaneously
every 6 months. He will be briefly evaluated at every injection visit with a
planned repeat DXA in 2 years.
Case 6
Male Case
Men Fracture
Regardless of the normative database used in this case, the
patient was at high risk for fracture.
He had decreased renal function, thus he was switched to a
non-renally excreted drug.
The Effect of 6 vs 9 Years of Zoledronic Acid Treatment in
Osteoporosis: A Randomized Extension to the
HORIZON-Pivotal Fracture Trial (PFT)
Results
•
•
•
•
•
Mean age 78 years, with 28% over 80 years of age
63% from Europe, 32% from N/S America, 6% from Asia
No differences in the biochemical markers Z6P3 vs Z9
No difference in BMD (THip :-1.31% in Z6P3 vs.-0.54% in Z9, p=n.s.) Z6P3 vs Z9
Incidence of fractures was low and similar in both groups (morphometric vertebral
fractures: 5.2% in Z6P3 vs 3.2% in Z9)
BMD: Bone mineral density; N/S: North/South; n.s., not significant; Z6P3, placebo; Z9, zoledronic acid.
Black D, et al. San Francisco, USA. The Effect of 6 versus 9 Years of Zoledronic Acid Treatment in Osteoporosis: A Randomized Extension to the HORIZONPivotal Fracture Trial (PFT). Abstract [SA0389]. Presented at the 2013 Annual Meeting of The American Society for Bone and Mineral Research. October 4, 2013.
Stratification of FN BMD:
Incident Morphometric Vertebral Fracture Rates
Z3P3
Proportion of Patients (%)
15
10
5
Z6
Treatment subgroup interactions not significant
9.2%
(23/250)
OR: 0.36
(95% CI: 0.15, 0.77)
P = 0.01
OR: 0.79
(95% CI: 0.23, 2.53)
P = 0.70
NNT = 18
NNT = 167
3.5%
(9/257)
3.0%
(7/235)
2.4%
(5/210)
0
FN BMD ≤ -2.5
FN BMD > -2.5
BMD, bone mineral density; CI, confidence interval; FN, femoral neck; NNT, number needed to treat; OR, odds ratio
Effects of Long-Term Risedronate
Treatment Unblinded at Year 3
Continuous Risedronate: Antifracture Efficacy for
Radiographic Vertebral Fractures Sustained Over
7 Years1
*#
*
7
Placebo
Risedronate
0
*
-30
#*
#*
-50
#*
-40
*
*
-20
CA + D
*
-10
Off
treatment
*
Median Change From Baseline
in Urinary NTX (%)
Risedronate-induced Reduction in Bone
Resorption Recovered After Discontinuation2
On
treatment
-60
03 6
Month
36
Placebo (n=77)
48
60
72
Annualized Incidence of
New Vertebral Fractures
Placebo 5 yrs/Ris 2 yrs (n=81)
13
Ris 7 yrs (n=83)
12
11
*#
10
*#
*# *#
9
8
*#
7
6
*
*#
5
4
3
*
2
1
0
-1
BLN 1
2
3
4
5
6
Years
84
Ris 5 mg (n=89)
407
407
130
130
81
83
Discontinued Risedronate: Greater Antifracture Efficacy
for New Vertebral Fractures in Patients Previously on
Risedronate Than Those on Placebo at 4 Years2
Incidence of New
Vertebral Fractures (%)
Mean Change in LS
BMD (%)
LS BMD Increased Over 7-Year
Treatment Period1
20
Control (n=361)
Previously on Ris 5 mg daily (n=390)
15
46%
(95% CI: 14–66%)
P=0.009
10
5
0
Year 4
* P<0.05 vs. baseline, # P<0.05 vs. placebo; BLN = baseline, D = vitamin D, Ris = risedronate
1. Mellström DD, et al. Calcif Tissue Int. 2004;75:462–468. 2. Watts NB, et al. Osteoporos Int. 2008;19:365–732.
Ibandronate Discontinuation
Ravn Bone 22:559, 1998
Denosumab Re-treatment and Changes to
Serum CTx and BSAP Levels
Phase 2 Study in Women With Low BMD
Placebo
30 mg Q3M
Serum CTx
BSAP
Discontinued Re-treatment
Treatment
60 mg Q6M
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
0
6
12
18
24
Months
30
36
42
48
25
Median mcg/L (Q1, Q3)
Median ng/mL (Q1, Q3)
1.6
Discontinued Re-treatment
Treatment
60 mg Q6M
20
15
10
5
0
0
6
12
18
24
30
36
42
48
Months
Miller PD, et al. Bone. 2008;43:222-229.
Denosumab Re-treatment and Changes in
Lumbar Spine and Total Hip BMD
Phase 2 Study in Women With Low BMD
Placebo
30 mg Q3M
Lumbar Spine
14
Discontinued Re-treatment
Treatment
60 mg Q6M
6
Percent Change
(LS Mean ± SE)
10
8
6
4
2
0
4
2
0
-2
-4
-2
-4
8
Discontinued Re-treatment
Treatment
60 mg Q6M
12
Percent Change
(LS Mean ± SE)
Total Hip
0
6
12
18
24
Months
36
48
-6
0
6
12
18
24
36
48
Months
Adapted from Miller PD, et al. Bone. 2008;43:222-229.
Cosman F. BoneKey 2014;(3):518
Mean Change at 1 Yr (%)
Mean Percent Changes in Areal Bone Mineral Density of the Spine,
Hip and Radius by DXA in Patients on PTH Plus Alendronate vs
PTH Monotherapy and Alendronate Monotherapy
Lumbar Spine
Total Hip
Femoral Neck
Distal one third
of radius
Areal Bone Mineral Density
Cosman F. BoneKey 2014;(3):518
Adapted from Black et al
Mean (SE) Percentage Changes in Bone Mineral Density
Total Hip
Change in BMD (%)
Change in BMD (%)
Posterior-Anterior Lumbar Spine
Change in BMD (%)
Femoral Neck
Cosman F. BoneKey 2014;(3):518
A supplement to JNCCN—Journal of the National Comprehensive Cancer Network
Algorithm for the
Management of Bone
Health in Cancer Patients
in the United States
A supplement to JNCCN—Journal of the National Comprehensive Cancer Network
BMD, bone mineral density;
DXA, dual-energy x-ray absorptiometry;
FRAX, Fracture Risk Assessment tool
Rates of Bone Loss with Cancer Therapies
The rates of bone loss associated with various cancer therapies are substantially greater than those seen with normal aging in men and women.
ADT, androgen deprivation therapy;
AI, aromatase inhibitor;
GnRH, gonadotropin-releasing hormone
A supplement to JNCCN—Journal of the National Comprehensive Cancer Network
Algorithm for Imaging for Cancer Patients in
The United States
Modified from Hamaoka T, Madewell JE, Podoloff DA, et al. Bone imaging in metastatic breast cancer. J Clin Oncol 2004;22:2942
A supplement to JNCCN—Journal of the National Comprehensive Cancer Network
The Dose and Frequency of Administration of Zoledronic Acid and
Denosumab for Treating Osteoporosis, Preventing Bone Loss From Endocrine
Therapies, and Preventing Skeletal-Related Events in Patients With Cancer
ADT, androgen deprivation therapy; AI, aromatase inhibitor.
aThis dose is not included in the FDA label for endocrine therapy–induced bone loss.
A supplement to JNCCN—Journal of the National Comprehensive Cancer Network
Safety Considerations and Recommendations for Patients
Undergoing Therapy With an Antiresorptive Agent
Cont.
A supplement to JNCCN—Journal of the National Comprehensive Cancer Network
Van Poznak et al. J Clin Oncol 2010; 28:967-975
Patient Demographic and Baseline Clinical Characteristics
A, anastrozole; P, placebo; R, risedronate; BMI, body mass index; HRT, hormone replacement therapy.
Van Poznak et al. J Clin Oncol 2010; 28:967-975
Estimated Percentage
Change at Each Visit up to 24
Months for (A) Lumbar Spine
and (B) Total Hip Bone
Mineral Density
(*) From simple geometric means
Van Poznak et al. J Clin Oncol 2010; 28:967-975
Eidtmann et al. Annals of Oncology 2010;21:2188-2194
Changes in Bone Mineral Density and T-score Categories
throughout the Zometa-Femara Adjuvant Synergy Trial Study
Eidtmann et al. Annals of Oncology 2010;21:2188-2194
Rizzoli et al. Nat. Rev. Rheumatol 2015;11:98-109
Physiopathology of GIOP in Inflammatory Diseases
Rizzoli et al. Nat. Rev. Rheumatol 2015;11:98-109
Intervention
Thresholds in
Various GIOP
Guidelines
Cont.
Rizzoli et al. Nat. Rev. Rheumatol 2015;11:98-109
Treatment Options According to Various GIOP Guidelines
Cont.
Rizzoli et al. Nat. Rev. Rheumatol 2015;11:98-109
Black et al. JBMR 2015; 30(5):934-944
Mean Changes in
BMD over 9 Years of
Treatment
BMD, bone mineral density;
ITT, intention-totreat;
LS, least square.
Black et al. JBMR 2015; 30(5):934-944
Mean Changes in
Bone Turnover
Markers over 9
Years of
Treatment
b-CTX, beta C-terminal type 1 collagen
telopeptide; BSAP, bone-specific alkaline
phosphatase; PINP, procollagen type I Nterminal propeptide.
Black et al. JBMR 2015; 30(5):934-944
The Incidence of
Morphometric
Vertebral Fractures and
Clinical Fractures
CI, confidence interval; PBO, placebo; ZOL, zoledronic acid.
Black et al. JBMR 2015; 30(5):934-944
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Approach to Postmenopausal Women and Men age 50
Years Initiating or Receiving Glucocorticoid Therapy
Counsel and assess risk factors of those
starting or on prevalent glucocorticoid therapy
Determine Patient Risk Category


Low Risk*
If glucocorticoids <7.5
mg/day: no pharmacologic
treatment recommended
If glucocorticoids ≥ 7.5
mg/day: alendronate,
residronate, or zoledronic
acid


Medium Risk*
If glucocorticoids <7.5
mg/day: alendronate or
residronate
If glucocorticoids ≥ 7.5
mg/day: alendronate,
residronate, or zoledronic
acid


High Risk*
If glucocorticoids <5 mg/day for ≤ 1
month: alendronate, residronate, or
zoledronic acid
If glucocorticoids ≥5 mg/day for ≤ 1
month or any dose of
glucocorticoids used for >1 month :
alendronate, residronate, zoledronic
acid, or teriparatide
Monitor Patients or prevalent corticoid therapy
* = for low- and medium-risk patients, recommendations are for an anticipated or prevalent duration
Of ≥ 3 months of glucocorticoids.
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Approach to Premenopausal Women and Men age 50 Years
Initiating or Receiving Glucocorticoid Therapy
No prevalent
fragility fracture
Counsel and assess
risk factors of those
starting or on
prevalent
glucocorticoid
therapy
Inadequate data for
recommendation
Women
(nonchildbearing
potential) or
mean age <50 years
Glucocorticoids 1---3
months
•
If pred≥ 5 mg/day:
alendronate or
residronate
OR
•
If pred ≥ 7.5 mg/day:
zoledronic acid
Glucocorticoids > 3 months
• alendronate OR
• residronate OR
• zoledronic acid OR
• teriparatide
Monitor
patients on
prevalent
glucocorticoid
therapy
Glucocorticoids 1---3
months
No consensus
Prevalent fragility
fracture
Women
(childbearing potential)
pred= prednisone
Glucocorticoids > 3 months
•
alendronate if If pred
≥ 7.5 mg/day OR
•
residronate If pred ≥
7.5 mg/day OR
•
teriparatide If pred ≥
7.5 mg/day
•
pred <7.5 mg daily: no
consensus
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Clinical Factors that may Shift an Individual to a Greater
Risk Category for Glucocortcoid-induced Osteoporosis
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Recommendations on
Counseling for Lifestyle
Modification and
Assessment of Patients
Starting
Glucocorticoids at any
Dose with an
Anticipated Duration
≥3 Months
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Recommended monitoring for patients receiving prevalent
glucocorticoid therapy for a duration of ≥ 3 months
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Pharmacologic Recommendations
for Postmenopausal Women and
Men age ≥ 50 Years Starting
Glucocorticoid Therapy with an
Anticipated Duration of
≥ 3 Months, or Prevalent
Glucocorticoid Therapy of a
Duration of at Least 3 Months
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526
Recommendations for
Premenopausal Women and
Men under Age 50 Years with a
History of Fragility Fracture
Grossman et al. Arthritis Care & Research 2010;62(11):1515-1526