B M k i O t i P di ti fBone Markers in Osteoporosis: Prediction of Fractures & Treatment Monitoring

Richard Eastell, MD FRCP FRCPath FMedSci,Professor of Bone Metabolism,

University of Sheffield,Sheffield, UKSheffield, UK

www.shef.ac.uk/bmg

Usefulness of Markers in the Individual PatientO tliOutline

• Bone turnoverBone turnover– Changes at the menopause and with

osteoporosisp• High bone turnover markers

– Association with bone loss and fracture– Association with bone loss and fracture• Effects of treatment on bone turnover

T t t it i• Treatment monitoring• Example

www.shef.ac.uk/bmg

The Effects of Age and Estrogen Status on Bone Remodelinge ects o ge a d st oge Status o o e e ode g

Activation frequency (#/yr)

1 2

1.4

1.6 A 2-fold increase across the menopause

A 3-fold increase by age 65 yr

0 8

1.0

1.2 A 3 fold increase by age 65 yr.

0 4

0.6

0.8

0 0

0.2

0.4

0.0

www.shef.ac.uk/bmg*Recker R, et al. J Bone Miner Res 2004;19:1628-33

BTMs Reflect Resorption and Formation– Matrix proteins

• osteocalcin (OC)

BTMs Reflect Resorption and Formation

Formation• procollagen type I propeptides

– C-terminal (PICP)

N-terminal (PINP)

Formation

– N-terminal (PINP)

– Enzyme

• bone isoform of alkaline phosphatase (bone ALP)

– Collagen degradation products

• pyridinium cross-links of collagen

(bone ALP) Commercially available assaysSerum: OC, PICP, PINP, bone ALP, NTX, CTX, TRACP 5bUrinary: CTX NTX free DPD– C- and N-telopeptides (CTX, NTX)

– Deoxypyridinoline (DPD)

– Enzyme

ResorptionUrinary: CTX, NTX, free DPD

• tartrate-resistant acid phosphatase (TRACP), type 5b

•Cathepsin K

www.shef.ac.uk/bmg

•Related factors

•OPG, RANK-L

Levels of Serum βCTX in Women C it b d t d OPUS ( 2780)Community-based study, OPUS (n=2780)

10

1

0 1(ng/

mL)

0.1

βC

TX

0.01

Pre menopausal Post menopausal

20 30 40 50 60 70 800.001

Pre-menopausalMean (SD) 0.16 (0.12) ng/mL

Post-menopausalMean (SD) 0.27 (0.19) ng/mL

www.shef.ac.uk/bmg

Age at visit

Data from Blumsohn A et al, J Bone Miner Res 2003;18:1274-81

High Bone Turnover Markers

www.shef.ac.uk/bmg

Bone Turnover Markers in Relation to Bone Loss and F t Ri kFracture Risk

• A high bone remodelling rate isA high bone remodelling rate is associated with– More rapid bone loss in postmenopausalMore rapid bone loss in postmenopausal

women1

– Increased risk of fractures2

• The association with fracture is variable2

– Hazard ratios of between 1 and 2– Hazard ratios of between 1 and 2– More consistently observed for bone

resorption than bone formation markersresorption than bone formation markers

www.shef.ac.uk/bmg

1. Stepan JJ. Osteoporos Int. 2000; 11 Suppl.6:S45-54;2. Garnero P. Osteoporos Int. 2000; 11 Suppl.6:S55-65.

Bone Turnover and Bone Loss from the SpineBone Turnover and Bone Loss from the Spine

5.0

yr) 5.0

r)

0.0

2.5

n B

MD

(%/y

0.0

2.5

BM

D(%

/yr

-5.0

-2.5

2=0 44

Cha

nge

in

-5.0

-2.5

r= 0 42Cha

nge

in

0 5 10 15 20 25-7.5

Years Since Menopause (YSM)

r2=0.44p=0.0076

4 8 16 32 64 128-7.5

uNTX (nmol BCE/mmol Cr)

r=-0.42p=0.0013

www.shef.ac.uk/bmgRogers A, et al. J Bone Miner Res. 2000;15:1398-1404.

High Bone Resorption Markers May Predict Hip Fracture Ri k i Eld l W th EPIDOS St dRisk in Elderly Women – the EPIDOS Study

EPIDOS = Epidemiologie de l'Osteoporose

5

4

3ratio

3

2

Odd

s r

2

11Low hip

BMDHigh

U-CTXHigh

U-DPDLow BMD Low BMD

+ +High CTX High DPD

www.shef.ac.uk/bmgGarnero P, et al. J Bone Miner Res. 1996;11:1531-38.

BMD = bone mineral density; U = urinary.

10-year Risk of Hip Fracture10 year Risk of Hip Fracture

• This approach hasThis approach has now been implemented asimplemented as the WHO FRAXTM

scorescore– Doesn’t include

bone markersJohnell O, et al. Osteoporosis Int.

2002;13:523-6.

www.shef.ac.uk/bmg

Treatment Effects

www.shef.ac.uk/bmg

Postmenopausal OsteoporosisEffects of Treatments

Quiescence NegativeRemodelling Imbalance

Remodelling Balanceg

Anti-catabolic Drugs•Decrease activation

www.shef.ac.uk/bmg

Activation Resorptionfrequency•Restore remodelling balance

Changes in Bone Resorption Marker NTX With Alendronate

Time (weeks)

Therapy in Osteoporosis – Mean Decrease 75%

4 8 12 240

Time (weeks)

NTX

iFDPD

OC –25

Bone ALPhang

e

–50% c

h

–75

www.shef.ac.uk/bmgMachado A, et al. J Bone Miner Res. 1999;14:602-8.

ALP = alkaline phosphatase; iFDPD = free deoxypyridinoline; OC = osteocalcin; NTX = N-terminal telopeptide.

Treatment Effects on Telopeptide MarkersCTX d NTXCTX and NTX

• Different responses to treatmentsDifferent responses to treatments– Strontium, calcium, 10 to 20%– Raloxifene 30 to 40%Raloxifene, 30 to 40%– HRT and risedronate, 50 to 60%– Zoledronic acid ibandronate alendronate 70– Zoledronic acid, ibandronate, alendronate, 70

to 80%• All reduce vertebral fractures• All reduce vertebral fractures

– Mechanism may involve effects other than a decrease in bone resorptiondecrease in bone resorption

• The telopeptide response helpful for any particular treatment

www.shef.ac.uk/bmg

particular treatment

Postmenopausal OsteoporosisEffects of Treatments

Quiescence NegativeRemodelling ImbalancePositiveRemodelling ImbalancegRemodelling Imbalance

Anabolic Drugs•Increase activation frequencyI d iti d lli•Induce positive remodelling

balance

www.shef.ac.uk/bmg

Activation Resorption

Randomized, Controlled Trial of Alendronate or Teriparatide i P t l O t i

% Change from baseline (Mean ± SE)

in Postmenopausal Osteoporosis

250Teriparatide

250Alendronate

150

200

150

200

0

50

100

0

50

100

-100

-50

0

-100

-50

0

MonthsNTX (Resorption)PINP (Formation)

0 1 3 6 12 0 1 3 6 12

P<0.001bet een gro p difference

www.shef.ac.uk/bmg

NTX (Resorption) between-group difference teriparatide vs. alendronate

2004Arlot M, et al. J Bone Miner Res. 2005;20:1244-53.

Strontium increases bone formation markers and decreases b ti k b t th ff t ll (<10%)

EE

bone resorption markers, but the effects are small (<10%)

0.8

1.2 ***

* **** **

0.4

0 8 * Bone ALP (ng/mL)

+8%

-300

0

S-CTX (pmol/L)

-600

300

*** *** *** * **

S CTX (pmol/L)

-12%

0 3 6 12 24 36

E E ti t f diff b t t ti d l b i l i

Months***

www.shef.ac.uk/bmg

E= Estimate of difference between strontium group and placebo group, covariance analysis, baseline adjusted *** p<0.001; ** p<0.01; * p<0.05

Meunier PJ et al, N Engl J Med. 2004 Jan 29;350(5):459-68.

Treatment Monitoring

www.shef.ac.uk/bmg

The average response to risedronate 5 mg exceeds LSC li f BTM (3 th ) th BMD (18 th )LSC earlier for BTM (3 months) than BMD (18 months)

Spine BMD, % Δ from baseline

Risedronate 5mg

NTX, % Δ from baseline

0

ControlRisedronate 5mg

-25

LSC

-50LSC

-750 6 36

ControlRisedronate 5mg

Time, months Time, months

www.shef.ac.uk/bmg

LSC, least significant change

Using BMD and Bone Turnover Markers in Clinical P ti A ti C t b li Th iPractice: Anti-Catabolic Therapies

100LSC (LSBMD)

10% 17%

0

Urinary NTX, % changeat 6 months

10% 17%

-100

at 6 monthsLSC (NTX)

-20065%8%

-10 0 10 20 30 40

Spine BMD, % change

www.shef.ac.uk/bmg

LSC, least significant changeBMD, bone mineral density

Fracture Incidence Over 3.6 Years of Alendronate, vs 12-Month Bone ALP: % Change From Baseline

• One year change in bone ALP in alendronate-treated

• FIT study3105 in alendronate arm

women– 3105 in alendronate arm– One-year change in bone

ALP (insufficient samples ip

ip

ure

ure

100100101011 for PINP or CTX)

– 3.6 years of follow-up for fracture

OR

Hi

OR

Hi

Frac

tuFr

actu 11

0.10.10.010.01

0.0010.001 fracture

Change in Bone ALP (%)Change in Bone ALP (%)--6060 --4040 --2020 00 2020

10101010

11

Vert

ebra

l Ve

rteb

ral

ract

ure

ract

ure

0.10.1

Ch i B ALP (%)Ch i B ALP (%)--6060 --4040 --2020 00 2020

OR

VO

R V FrFr

www.shef.ac.uk/bmg

OR=Odds ratio.Bauer DC et al. J Bone Mineral Res. 2004;19:1250-1258.

Change in Bone ALP (%)Change in Bone ALP (%)

Surrogates for Fracture Risk ReductionSurrogates for Fracture Risk Reduction

Proportion of 3 year fragility fracture treatment effect explained by surrogatesProportion of 3 year fragility fracture treatment effect explained by surrogates

Surrogate Proportion 95% CISurrogate Proportion explained

95% CI

Femoral neck 14% 7 21%Femoral neck BMD

14% 7, 21%

U i NTX 52% 29 75%Urinary NTX 52% 29, 75%

NTX d 67% 39 100%NTX and FNBMD

67% 39, 100%

www.shef.ac.uk/bmgDelmas PD, et al. ECCEO 2005; abstract P360

We Would Like to Suppress Bone Turnover to Levels Found inH lth Y W (Wh H A T S f 0)

Treatment: Calcium Calcium and risedronate 5 mg

Healthy Young Women (Who Have an Average T-Score of 0)l s

25

Treatment: Calcium Calcium and risedronate 5 mg

Vert

ebra

lo

3 ye

ars

15

20 Reduce absolute vertebral fracture risk by 10%

ence

of V

ures

, 0 to

10

15

Reduce bone resorption by 2 SD

Inci

deFr

actu

0

5

00.0 0.5 1.0 1.5 2.0

(20) (35)-2.0 -1.5 -1.0 -0.5

www.shef.ac.uk/bmg

Urinary NTX, T-score

Eastell R, et al. J Bone Miner Res. 2003;18:1051-1056.

Vertebral Fractures over 3 years and Bone Resorption C bi d T t d d Pl b GCombined Treated and Placebo Groups

Decile Analysis, about 66/decile

Th h ld i t20

25No. women withvertebral fractures

Threshold is atmean for young women (T=0)

15

20

10

0

5

1 2 3 4 5 6 7 8 9 100

Urinary CTX (T-score) at 3 to 6 months, decileT=-1 T=+1

www.shef.ac.uk/bmgEastell R, Hannon RA, Garnero P, Campbell MJ, Delmas PD. JBMR 2007 (letter online 17th September 2007)

Develop Reference Range for Women Based on t l t 150 H lth Y W

Reference Range Graphical Display of the DataNTx/Cr Premenopausal Reference Range (Healthy drug-free subset)

at least 150 Healthy Young Women

LocationMean 47.066Median 42.4079Mode No Mode 0 02

0.025

0.03

AdjR

elativeMode No Mode

SpreadStandard Deviation 24.4041Variance 595.5589Parametric 95% LRL -1 7422

0.01

0.015

0.02 justed e Frequency

Parametric 95% LRL 1.7422Parametric 95% URL 95.8742Non Parametric 2.5th centile 16.3759Non Parametric 97.5th centile 103.5667

-0.005

0

0.005

-50 0 50 100 150 200 250

Minimum 8.03217427Maximum 211.0338478Interquartile Range 25.2433Range 203.0016735

-0.015

-0.01

MiscellaneousN 370CV 0.5185

-0.025

-0.02

www.shef.ac.uk/bmg

Effect of Risedronate Therapy on Bone Resorption

Lower half ofreference interval

Upper half ofreference interval

O O

80

Osteoporosis, Untreated Osteoporosis, Risedronate

70

400 11 62 27% 7 63 28 2%

atien

ts

50

60

70

tient

s

30

Num

ber o

f Pa

20

30

40

Num

ber o

f Pa

20

0

10

20

0

10

NTX/Cr, nmol/mmol Creatinine NTX/Cr, nmol/mmol Creatinine10010 3210010 32

www.shef.ac.uk/bmgEastell R and Delmas PD. J Bone Miner Res. 2005;20:1261-2

Does monitoring affect adherence?

www.shef.ac.uk/bmg

Does Monitoring affect Adherence?Does Monitoring affect Adherence?

• 75 women with osteopenia75 women with osteopenia• Treated with Raloxifene and randomised

Repeat prescriptions (no medical contact)– Repeat prescriptions (no medical contact)– Visit nurse every 3 months

Visit nurse and monitor BTM every 3 months– Visit nurse and monitor BTM every 3 months• Measured compliance using MEMS caps

www.shef.ac.uk/bmgClowes JA et al, J Clin Endocrinol Metab 2004;89:1117-23

Monitoring improves adherence to treatment with R l ifRaloxifene

Proportion of patients who remained adherent

Cumulative adherence >75 %

Proportion of patients who remained adherent1.2

1 0 %– Proportion of adherent

subjects increased by 57%

1.0

0.8

in those monitored– P = 0.04

0.6

0 40.4

0.2

Monitored

Not monitored

Days4003002001000

0.0

www.shef.ac.uk/bmgClowes JA et al, J Clin Endocrinol Metab 2004;89:1117-23

Days

Greater adherence is associated with greater biological response

% Change uNTX% Change Hip BMD40

20

0

6

4 0

- 20

- 402

- 60

- 80

0

- 2

120100806040200

- 100

- 120120100806040200

- 4

r = - 0.46P 0 0001

r = 0.34P 0 003

% Adherence% Adherence

www.shef.ac.uk/bmg

P = 0.0001P = 0.003Clowes JA et al, J Clin Endocrinol Metab 2004;89:1117-23

Case PresentationCase Presentation

• Woman (71) develops ( ) pacute onset back pain

• There are no clear risk factors for osteoporosis from the questionnaire or the -3 2 -2 8questionnaire or the biochemical workup

• Bone turnover marker

3.2 2.8T-score

• Bone turnover marker– NTX T-score +4 (150

nmol BCE/mmol Cr)– Bone alkaline

phosphatase T-score +3 (90 ug/L)

www.shef.ac.uk/bmg

+3 (90 ug/L)

What does a high bone turnover marker mean in t i ?osteoporosis?

• Could it just be a spurious result?Ho can e se the information?• How can we use the information?• Is the patients more likely than average to have

rapid bone loss or an increased risk ofrapid bone loss or an increased risk of fracture?

• Does this high result mean she might have• Does this high result mean she might have secondary osteoporosis?

• What do we expect to happen to the resultWhat do we expect to happen to the result when we treat her? What is the goal of our treatment?

www.shef.ac.uk/bmg

Could it just be a spurious result?

www.shef.ac.uk/bmg

Bone resorption markers do vary from day to day, b t it i lik l th t h i i hi h t t tbut it is very likely that she is in a high turnover state

Data from Clowes JA, et al. J Bone. 2002;30:886-90.

Fasting NTx

13

g

58

1012

ject

12141618

Subj

1820

0 40 80 120 160 200 240

www.shef.ac.uk/bmg

If we know why bone turnover markers vary, ll f itwe can allow for it

• Controllable source– Circadian

• Uncontrollable source– GrowthCircadian

– Day to day– Diet

Growth– Age and gender– FracturesDiet

– ExerciseFractures

– Diseases and drugs

• Consequence– Morning, fasting

• Consequence– Establish age and g, g

samples– Take average of more

ggender-specific reference range

www.shef.ac.uk/bmg

than one measurementHannon RA, Eastell R. Osteoporosis Int. 2000;11 (Suppl.6):S30-44.

Circadian Rhythm of Serum CTXCircadian Rhythm of Serum CTX

Eagleton A, PhD thesis, 2003.

sCTX0.75

)

0.50

TX (n

g/m

l

0.25

sCT

08:00 12:00 16:00 20:000.00

Time

www.shef.ac.uk/bmg

TimeArrows indicate meal timesUntreated osteoporosis, n = 15Osteoporosis treated with risedronate for 3 months, n = 15

Biochemical Markers following Ankle FractureBiochemical Markers following Ankle Fracture

Formation Resorption

220% baseline % baseline

Formation Resorption

220

160

180

200 TRAcPifDpdNTx

PINPI-Bone ALP

WG-Bone ALPOC

160

180

200

100

120

140

100

120

140

137 14 2860

80

100

42 90 180 360137 14 2860

80

100

42 90 180 360

Time, days Time, days137 14 28 42 90 180 360

www.shef.ac.uk/bmgIngle BM et al. Osteoporosis Int. 1999;10:408-15.

I h i lik l h hIs the patients more likely than average to have rapid bone loss or an increased risk of fracture?

www.shef.ac.uk/bmg

She is likely to be a ‘fast bone loser’W ld h ll l 5% ?Would she really lose 5% per year?

5.0

yr) 5.0

r)

0.0

2.5

n B

MD

(%/y

0.0

2.5

BM

D(%

/yr

-5.0

-2.5

2=0 44

Cha

nge

in

-5.0

-2.5

r= 0 42Cha

nge

in

*0 5 10 15 20 25

-7.5

Years Since Menopause (YSM)

r2=0.44p=0.0076

4 8 16 32 64 128-7.5

uNTX (nmol BCE/mmol Cr)

r=-0.42p=0.0013 *

www.shef.ac.uk/bmg

Rogers A, et al. J Bone Miner Res. 2000;15:1398-1404.

10-year Risk of Hip Fracture10 year Risk of Hip Fracture

With l BMD i f t dWith low BMD, prior fracture and high bone turnover marker shehas a 30% 10-year risk of fracturing, better do something soon!better do something soon!

Johnell O, et al. Osteoporosis Int.2002;13:523-6.

www.shef.ac.uk/bmg

Use in Decision-MakingUse in Decision Making

• Does this help us diagnose osteoporosis?Does this help us diagnose osteoporosis?• Does this help us select the best

treatment?treatment?• Does the result make us suspect

secondary osteoporosis?secondary osteoporosis?

www.shef.ac.uk/bmg

The high turnover doesn’t allow us to diagnose osteoporosis, or choose the best t t t it i ht i t t d t itreatment; it might point us to secondary osteoporosis

• Bone turnover markers are increased in b t 25% f ith iabout 25% of women with primary

osteoporosisB t k h t t• Bone turnover markers have not yet proven useful in the selection of treatment

Bisphosphonates are effective at all levels of– Bisphosphonates are effective at all levels of bone turnover1 or only with high turnover2

– Teriparatide most effective if bone turnover isTeriparatide most effective if bone turnover is high!3

• A high level of bone turnover may indicate g ythe presence of secondary osteoporosis

www.shef.ac.uk/bmg

1. Seibel MJ, et al. J Bone Miner Res. 2004;19:323-3292. Bauer DC, et al. J Bone Miner Res. 2006;21:292-2993. Ettinger B, et al. J Bone Miner Res. 2004;19:745-751

What are the causes of high bone turnover? Watch outgparticularly for endocrine or other bone diseases

HIGH LOW• Metabolic bone diseases

– Paget’s diseaseSecondary

• Cushing’s syndrome– Secondary

hyperparathyroidism • osteomalacia

– low osteocalcin• Osteogenesis imperfecta

l PICP• renal osteodystrophy• malabsorption syndrome

• Endocrine diseases

– low PICP• Hypophosphatasia

low alkaline phosphatase• Endocrine diseases– Primary

hyperparathyroidism

– low alkaline phosphatase

yp p y– Thyrotoxicosis– Hypogonadism

www.shef.ac.uk/bmg

• Malignancy, e.g. myeloma• Recent fracture

Wh t d t t h t th lt hWhat do we expect to happen to the result when we treat her? What is the goal of our treatment?

www.shef.ac.uk/bmg

After starting antiresorptive therapy, should we make follow-up measurements of bone turnover?

• Identify poor response that can be causedIdentify poor response that can be caused by– Poor compliancePoor compliance– Secondary osteoporosis– Ineffective therapiesIneffective therapies

• Take action to improve response Advise on correct dosing instructions– Advise on correct dosing instructions

– Investigate for secondary osteoporosis– Consider larger dose or different treatmentg

• Encourage compliance– No immediate symptomatic benefit to patient

www.shef.ac.uk/bmg

No immediate symptomatic benefit to patient

Our patient was treated with alendronate 70 mg once k d t b th f t ta week, and met both of our targets

NTX/Cr, nmol BCE/mmol Goals

150Decrease to levels in the

50

100 lower half of the reference range for young women, <35 nmol/mmol Cr

B1 B2 mB1B2 V1 V2 V30

NTX, % change

100

200

Decrease by more than the least

200

-100

0 significant change, 50%

www.shef.ac.uk/bmg

B2 mB1B2 V1 V2 V3-200

BMD increase of 5.3% at lumbar spine at 1 year

Usefulness of Markers in the Individual Patient SSummary

• Bone turnover– Increases at the menopause and

increases further in osteoporosis• High bone turnover markers

– Associated with higher risk of bone l d f tloss and fractures

• Treatments for osteoporosis– Result in changes in bone turnover

markers• Treatment monitoring• Treatment monitoring

– Goal is to decrease bone turnover markers beyond least significant

www.shef.ac.uk/bmg

markers beyond least significant change and into pre-menopausal range