1 Effect of Hormone Therapy: Estrogen Discontinuation EPIC study - - PDF document

SLIDE 1

1

OOC

Michael R. McClung, MD, FACP, FASBMR

Oregon Osteoporosis Center Portland, Oregon, USA Mary MacKillop Institute for Health Research Australian Catholic University, Melbourne, VIC mmcclung.ooc@gmail.com

Individualizing Osteoporosis Therapy

2019 NAMS Annual Meeting

Chicago, IL September 26, 2019 OOC

Disclosures

I am disclosing financial relationships as follows: Scientific Advisory Boards: Amgen, Myovant Honorarium for speaking: Amgen Michael McClung, MD 2019 OOC

Osteoporosis

Definition: A disorder due to bone loss that damages skeletal architecture, weakens the skeleton and predisposes a patient to fracture

• Several osteoporosis drugs effectively and

quickly reduce fracture risk in patients with

• steoporosis
• Osteoporosis is a chronic disease requiring

prolonged treatment

• It is important to develop a strategy for long-

term management

Images Courtesy of

• Drs. David Dempster & Roger Zebazi

Black DM and Rosen CJ. N Engl J Med 2016; 374:254-62

OOC

Case 1

• Healthy 65 year old woman
• menopause at age 51
• age 55: hot flashes persist
• weight 118#; BMI 21.6; BMD
• began transdermal estradiol

and micronized progesterone

• hot flashes abated
• age 65: insurance plan and PCP

recommend stopping estrogen

• no CV or fracture risk factors

Age Lumbar spine Total hip Femoral neck 55

• 2.1
• 1.6
• 1.5

60

• 1.4
• 1.2
• 1.0

65

• 1.5
• 1.1
• 1.1
• Management choices:
• discontinue estrogen
• continue estrogen
• discontinue estrogen and

begin raloxifene

• discontinue estrogen and

begin alendronate

SLIDE 2

2

OOC

Lumbar spine BMD Placebo Hormone Therapy/Placebo Years 1 2 3 4 5 6 Mean Percent Change from Baseline

• 6
• 3

3 6 9 Wasnich RD, McClung M et al, Menopause 2004;11:622-30

EPIC Prevention Trial--women between ages 45-59 Treated for 4 years; followed for 2 more years

Estrogen Discontinuation – EPIC study

Total hip BMD Years 1 2 3 4 5 6 Mean Percent Change from Baseline

• 4
• 2

2 4 6 OOC Annualized rates - %

CEE-MPA Treatment (N=8506) CEE-MPA Post-treatment (N-8052) PBO Post-treatment (N=7678) Placebo (N=8102)

Effect of Hormone Therapy:

Fractures in WHI

Heiss G et al. JAMA 2008;299:1036-45

0.5 1 1.5 2

Hip fracture Vertebral fracture Other fractures 34% 33% 24%

OOC Annualized rates - %

CEE-MPA Treatment (N=8506) CEE-MPA Post-treatment (N-8052) PBO Post-treatment (N=7678) Placebo (N=8102)

Effect of Withdrawing Hormone Therapy:

Fractures in WHI

Heiss G et al. JAMA 2008;299:1036-45

0.5 1 1.5 2

Hip fracture Vertebral fracture Other fractures

Loss of fracture protection during first 2 years off hormone therapy but no excess fracture risk after stopping estrogen OOC

Effect of Withdrawing Hormone Therapy:

Fractures in WHI

CEE-MPA Arm CEE Alone Arm

Cumulative hazard ratio of total fractures in WHI study participants followed for 5 years after stopping hormone therapy or placebo

Watts NB et al, J Clin Endocrinol Metab 2017;102:302–8 Small (15%) residual benefit to prior therapy observed N=10,134 N=5053

SLIDE 3

3

OOC

Baseline 6 months 12 months`

Between group difference at 12 months Mean 3.5% (95% CI 2.6, 4.4) p<0.001

Preventing Bone Loss When Estrogen is Stopped

% Change from Baseline + SE

• 2
• 1

1 2 3

Alendronate 70 mg once weekly

2.3%*

• 1.4%*

124 PM women who discontinued HT or ET within 6 months randomized (1:1) to: ALN 70 mg once weekly

• or -

RLX 60 mg once daily Double dummy PBO Calcium 1000 mg and vitamin D 400 IU daily

McClung MR, Watts NB et al, J Bone Miner Res 2005;20:S397

Lumbar spine BMD

Raloxifene 60 mg daily

OOC

Case 1

• Healthy 65 year old woman
• menopause at age 51
• age 55: hot flashes persist
• weight 118#; BMI 21.6; BMD
• began transdermal estradiol

and micronized progesterone

• hot flashes abated
• age 65: insurance plan and PCP

recommend stopping estrogen

• no CV or fracture risk factors

Age Lumbar spine Total hip Femoral neck 55

• 2.1
• 1.6
• 1.5

60

• 1.4
• 1.2
• 1.0

65

• 1.5
• 1.1
• 1.1
• Management choices:
• discontinue estrogen
• continue estrogen
• discontinue estrogen and

begin raloxifene

• discontinue estrogen and

begin alendronate OOC

Case 1

• Healthy 65 year old woman
• menopause at age 51
• age 55: hot flashes persist
• weight 118#; BMI 21.6; BMD
• began transdermal estradiol

and micronized progesterone

• hot flashes abated
• age 65: insurance plan and PCP

recommend stopping estrogen

• no CV or fracture risk factors

Age Lumbar spine Total hip Femoral neck 55

• 2.1
• 1.6
• 1.5

60

• 1.4
• 1.2
• 1.0

65

• 1.5
• 1.1
• 1.1
• Management choices:
• discontinue estrogen
• continue estrogen
• discontinue estrogen and

begin raloxifene

• discontinue estrogen and

begin alendronate OOC

Case 1 – Key Points

• Estrogen therapy prevents bone loss in almost all postmenopausal

women

• In my opinion, estrogen should be considered to prevent bone loss in early

postmenopausal women at risk for osteoporosis, even without menopausal symptoms

• low peak BMD, thin, family history, history of fracture
• Discontinuation of estrogen results in loss of skeletal benefit (BMD

and fracture protection) very quickly - within one year

• Bisphosphonates but not raloxifene prevent the loss of BMD upon

stopping estrogen

• If/when estrogen is stopped, think of skeletal consequences and

consider switching to a bisphosphonate in patients at risk

SLIDE 4

4

OOC

Case 2

• Healthy 73 year old woman
• menopause at age 52
• age 62: wrist fracture
• age 68: humerus fracture; BMD
• begun on alendronate; well tolerated
• age 73: BMD
• Management choices:
• discontinue alendronate therapy
• continue alendronate therapy
• switch to IV zoledronic acid
• switch to denosumab
• switch to a bone-building agent

Age Lumbar spine Total hip Femoral neck 68

• 2.6
• 2.9
• 2.5

70

• 2.1
• 2.7
• 2.4

73

• 1.9
• 2.6
• 2.5

OOC

Criteria for Bisphosphonate “Holiday”

Adler R et al. J Bone Miner Res 2016; 31:16–35

• After 3-5 years of therapy
• Continue a treatment if patient still

meets criteria for therapy

• History of hip or spine fracture or

multiple other fragility fractures

• BMD consistent with osteoporosis
• Other risk factors resulting in high

fracture risk

• For patients not meeting these

criteria, consider a bisphosphonate holiday

OOC

Fractures with Long-term Zoledronate

Fracture protection persists but does not improve with long term therapy

Black DM et al. N Engl J Med 2007;356:1809–22 Black DM et al. J Bone Miner Res 2012;27:243-54 Black DM et al. J Bone Miner Res 2015;30:934-44 3.0% (14/469)

% Patients

Morphometric Vertebral Fractures

3.3% (92/2822) 10.9% (310/2853) 70% (62, 76)

ZOL PBO 5 10 15

P = <0.001

4-6 Years 1-3

4.4% (3/68)

7-9 Core study Extension study

82% (45/616)

Non-vertebral Fractures

8.0% (292/3861) 10.7% (388/3875) 25% (12, 36)

5 10 15

P = <0.001

4-6 Years 1-3

12.2% (10/95)

7-9 Core study Extension study

OOC

Transition from Alendronate to Zoledronate

• 2
• 1

1 2

Baseline Month 12

McClung M et al. Bone 2007;41:122–28

Lumbar spine BMD % change from baseline

Postmenopausal women with

• steoporosis who received

alendronate for at least 12 months (mean 36 months) Randomly assigned to continue alendronate or to zoledronate 5 mg IV

Alendronate Zoledronate

In the absence of non-compliance

• r malabsorption, there appears to

be no BMD advantage to switching

SLIDE 5

5

OOC

Effects of Long-term Therapy on Total Hip BMD

1. Black DM et al. J Bone Miner Res 2015;30:934-44

Long-term Denosumab

FREEDOM Extension

Percentage Change From Baseline Study Year

1 2 3 4 5 6 7 8 9 10

4.6% Zoledronate 5 mg/y1 10 - 8 - 2 - 6 - 0 - 4 -

• 2 -

OOC

Effects of Long-term Therapy on Total Hip BMD

Progressive increase with denosumab vs plateau after 5 years with bisphosphonates

1. Black DM et al. J Bone Miner Res 2015;30:934-44 2. Bone HG et al. Lancet Diabetes Endocrinol 2017 2017;5:513-23

Long-term Denosumab

FREEDOM Extension

• 2

2 4 6 8 10 9.2%

Percentage Change From Baseline Study Year

1 2 3 4 5 6 7 8 9 10

4.6% Zoledronate 5 mg/y1 Denosumab2

OOC

Switching From Bisphosphonates to Denosumab

Data are least-squares means and 95% confidence intervals. *p < 0.0001 denosumab vs BP.

1Recknor C et al. ASBMR Poster FR0388. 2Kendler DL et al. J Bone Miner Res. 2010;25:72-81. 3Miller PD et al. J Clin Endo Metab. 2016;101:3163-70.

IBN ALN ZOL RIS

1.6%* 1.4%* 0.9%* 1.3%*

1 2 3

Total Hip Percent Change From Baseline

0.5% 0.9% 1.1% 0.6% 2.0% 2.2% 1.9% 1.9%

0.0% 1.0% 2.0% 3.0% 4.0% vs RIS vs IBN vs ALN vs ZOL Patients who had previously been treated with bisphosphonates randomly assigned to a bisphosphonate or denosumab. Denosumab

OOC

Teriparatide After Bisphosphonate Therapy

Boonen S et al. J Clin Endocrinol Metab 2008;93:852–60

% change in BMD from baseline Non- bisphosphonate

245 women treatment-naïve or previously treated with antiresorptive drugs began teriparatide therapy for 24 months (EUROFORS)

Alendronate Risedronate

Lumbar spine Total hip

Substantial increase in spine and hip BMD

10 4 6 8 2

SLIDE 6

6

OOC

Romosozumab after Bisphosphonate Therapy

Integral vBMD and FEA at the Hip by QCT

Integral vBMD

Mean Percent Change from Baseline (95% CI) Months

5 4 3 1

• 1
• 2
• 3
• 4

12 6 2

Teriparatide (n = 178) Romosozumab (n = 176) –0.8%* –0.2% 3.4%†‡ 2.3%†‡ Langdahl B et al. Lancet 2017;390:1585–94

• In patients previous treated with alendronate, vBMD and estimated bone

strength by finite element analysis (FEA) at the hip increased significantly with romosozumab but not with teriparatide

Teriparatide (n = 178) Romosozumab (n = 176) Mean Percent Change from Baseline (95% CI)

5 3 2

• 1
• 2

12 6 1 4

Months 2.5%*† 2.1%*† –1.0%* –0.7%

Estimated bone strength (FEA)

OOC

FNIH Meta-regression

Change in BMD vs Reduction in Fracture Risk

Bouxsein M et al. J Bone Miner Res 2019 Jan 23. doi: 10.1002/jbmr.3641

Larger increases in total hip BMD were associated with greater reduction in vertebral and hip fracture risk

Not significant

Study level comparison of changes in total hip BMD vs fracture risk 38 placebo-controlled trials 19 therapeutic agents 111,183 subjects 4557 vertebral fractures 882 hip fractures

r2=0.56 [p=0.0002] r2=0.48 [p=0.013]

OOC

Expected 1-year Nonvertebral Fracture Incidence in Open-label Period (%) Total Hip T-score at Month 12

During Open-label Period In the ARCH Study1

All subjects for both treatment groups (N = 3,342) 95% CI 1 2 3 4 5 6 7 8 9 –3.50 –3.25 –3.00 –2.50 –2.75 –1.50 –1.75 –2.25 –2.00 Total Hip T-score Throughout 10 Years of Denosumab Denosumab (N = 3,612) 95% CI Expected 1-Year Nonvertebral Fracture Incidence (%)

Throughout 10 Years of Treatment in the FREEDOM Study2

–3.50 –3.25 –3.00 –2.50 –2.75 –1.50 –1.75 –2.25 –2.00 1 2 3 4 5 6 7 8 9

On-treatment Total Hip BMD Correlates with Current Fracture Risk

N indicates the number of subjects who had a month 12 total hip BMD T-score and had ≥ 1 open-label alendronate dose for the ARCH study and number of subjects randomized to denosumab in FREEDOM who had an observed total hip BMD T-score at FREEDOM baseline and ≥ 1

• bserved total hip BMD T-score during FREEDOM or the Extension. Dashed lines indicate upper and lower 95% CI.

BMD=bone mineral density; CI=confidence interval

• 1. Cosman F, et al. Presented at: ASBMR Annual Meeting. September 28-October 1, 2018. Montréal, Canada. Oral 1074
• 2. Ferrari S, et al. J Bone Miner Res 2019 Mar 28

Denosumab Alendronate and romosozumab

OOC

Case 2

• Healthy 73 year old woman
• menopause at age 52
• age 62: wrist fracture
• age 68: humerus fracture; BMD
• begun on alendronate; well tolerated
• age 72: BMD
• Management choices:
• discontinue alendronate therapy
• continue alendronate therapy
• switch to IV zoledronic acid
• switch to denosumab
• switch to a bone-building agent

Age Lumbar spine Total hip Femoral neck 68

• 2.6
• 2.9
• 2.5

70

• 2.1
• 2.7
• 2.4

73

• 1.9
• 2.6
• 2.5

SLIDE 7

7

OOC

Case 2

• Healthy 73 year old woman
• menopause at age 52
• age 62: wrist fracture
• age 68: humerus fracture; BMD
• begun on alendronate; well tolerated
• age 72: BMD
• Management choices:
• discontinue alendronate therapy
• continue alendronate therapy
• switch to IV zoledronic acid
• switch to denosumab
• switch to a bone-building agent

Age Lumbar spine Total hip Femoral neck 68

• 2.6
• 2.9
• 2.5

70

• 2.1
• 2.7
• 2.4

73

• 1.9
• 2.6
• 2.5

I think that bisphosphonates should not be used for more than 5 years consecutively

• no additional benefit
• increased risk of atypical fracture
• switching leads to added BMD gain
• then follow that therapy with bisphosphonate

OOC

Case 3

• Healthy 77 year old woman
• menopause age 54
• estrogen x 6 years until WHI
• age 66: breast cancer; AI therapy x 5 years
• age 71: wrist fracture after hard fall; BMD
• age 77: acute back pain while gardening;

compression fractures T12 and L1; BMD

• evaluation for secondary causes was

negative

• Management choices:
• bisphosphonate
• denosumab
• bone forming agent
• teriparatide
• abaloparatide
• romosozumab

Age Lumbar spine Total hip Femoral neck 71

• 2.4
• 2.3
• 2.0

77

• 2.9
• 2.7
• 2.4

OOC

Teriparatide vs Risedronate

VERO Trial

• 1360 postmenopausal women with
• steoporosis and previous vertebral

fractures

• 72% had been treated previously
• Randomly assigned to teriparatide 20

ugm SQ daily or risedronate 35 mg PO

• nce weekly
• Followed for 2 years
• More than 50% reduction in vertebral

and non-vertebral fractures with teriparatide vs risedronate

Kendler DL et al. Lancet 2017 Nov 9. pii:S0140-6736(17)32137-2

OOC

Romosozumab vs Alendronate

Phase 3: ARCH: Vertebral Fracture Risk Reduction

Saag K et al. N Engl J Med. 2017;377:1417-27

Number of vertebral fractures in Year 2 115 45 61%

• Romosozumab was superior to alendronate – BMD and fracture risk
• Alendronate maintained the benefit of previous romosozumab therapy
• 61% reduction during year 2 in patients who received room followed by

alendronate compared to those who received 2 years of alendronate

SLIDE 8

8

OOC

Romosozumab vs Alendronate

Phase 3: ARCH: Fracture Risk Reduction

• Clinical and non-vertebral fracture risk reduction was alsosperior with

romosozumab was maintained upon transitioning from romosozumab to alendronate

Saag K et al. N Engl J Med. 2017;377:1417-27

OOC

Bone-building Agent vs Placebo, Followed by Anti-remodeling Drug

Cosman F et al. N Engl J Med 2016;375:1532-43

Year 1 Year 2

Romosozumab Placebo All patients on denosumab 60 mg Q6M

Fracture risk lower while on anti-remodeling agent after treatment with a bone-building agent vs treatment with placebo

Romosozumab Placebo Romosozumab To DMab Placebo to DMab Year 2: N=25 Year 2: N=5

Bone H et al. J Clin Endo Metab 2018;103:2949–57

Months 1-18 Months 19-43

Abaloparatide Placebo All patients on Alendronate 70 mg QW

12 months 24 months

OOC

Case 3

• Healthy 77 year old woman
• menopause age 54
• estrogen x 6 years until WHI
• age 66: breast cancer; AI therapy x 5 years
• age 71: wrist fracture after hard fall; BMD
• age 77: acute back pain while gardening;

compression fractures T12 and L1; BMD

• evaluation for secondary causes was

negative

• Management choices:
• bisphosphonate
• denosumab
• bone forming agent
• teriparatide
• abaloparatide
• romosozumab

Age Lumbar spine Total hip Femoral neck 71

• 2.4
• 2.3
• 2.0

77

• 2.9
• 2.7
• 2.4

OOC

Case 3

• Healthy 77 year old woman
• menopause age 54
• estrogen x 6 years until WHI
• age 66: breast cancer; AI therapy x 5 years
• age 71: wrist fracture after hard fall; BMD
• age 77: acute back pain while gardening;

compression fractures T12 and L1; BMD

• evaluation for secondary causes was

negative

• Management choices:
• bisphosphonate
• denosumab
• bone forming agent
• teriparatide
• abaloparatide
• romosozumab

Age Lumbar spine Total hip Femoral neck 71

• 2.4
• 2.3
• 2.0

77

• 2.9
• 2.7
• 2.4

SLIDE 9

9

OOC

Case 3 – Key Points

• Recent studies consistently show that the sequence of a bone-

forming agent followed by an anti-remodeling drug induces greater gains in BMD and greater reduction in fracture risk than the anti- remodeling drug alone

• The treatment benefit of a bone-building agent persists for at least 2

years after transitioning to an anti-remodeling drug

OOC

Individualizing Osteoporosis Therapy

Estrogen, raloxifene Bisphosphonate Teriparatide, abaloparatide, romosozumab Denosumab

When concerned about hip fracture After 18-24 months After 18-24 months 3-5 years

Low risk High risk Consider drug holiday Re-treat

Denosumab Bisphosphonate

2 years ALN 1-2 doses ZOL If discontinued For very high risk patients McClung M. Personal opinion, 2019

OOC

Individualizing Osteoporosis Therapy

Summary

• Skeletal effects of osteoporosis drugs are lost, often quickly,

when treatment is stopped

• drug holidays are never justified except for patients at low

fracture risk on bisphosphonates

• Lifelong management of osteoporosis is required
• treatment for only 5 years is not appropriate
• BMD achieved on therapy correlates with current fracture risk
• influences initial therapy and sequence of therapy
• Bone-forming therapy followed by an anti-remodeling drug is

superior to beginning treatment with an anti-remodeling drug

McClung M. Personal opinion, 2019

OOC

Individualizing Osteoporosis Therapy

Summary

No one approach to treatment is right for all patients

McClung M. Personal opinion, 2019

Photo courtesy of Betsy Love McClung, RN, MN

We do not treat osteoporosis -

• we treat patients with osteoporosis

SLIDE 10

10

OOC

Thank you

Photo by Anonymous

OOC OOC

Teriparatide vs Risedronate

VERO Trial

• 1360 postmenopausal women with
• steoporosis and previous vertebral

fractures

• 72% had been treated previously
• Randomly assigned to teriparatide 20

ugm SQ daily or risedronate 35 mg PO

• nce weekly
• Followed for 2 years
• More than 50% reduction in vertebral

and non-vertebral fractures with teriparatide vs risedronate

Kendler DL et al. Lancet 2017 Nov 9. pii:S0140-6736(17)32137-2