Discovery and development of
odanacatib: A selective cathepsin K
inhibitor for the treatment of
osteoporosis
Shawn J. Stachel
On behalf of the odanacatib discovery and product development teams
Merck Research Laboratories
8th RSC-SCI Symposium on Proteinase Inhibitor Design
April 16, 2013
1
Acknowledgements
Biology
Christine Brideau
Wanda Cromlish
Sylvie Desmarais
Jean-Pierre Falgueyret
Louis-Jacques Fortin
Sonia Lamontagne
France Landry
Frederic Masse
Dave Percival
Denis Riendeau
Paul Tawa
Jennifer Wang
Comp. Med.
Stephanie Alleyn
Liette Belair
Joel Bosquet
Carl Brière
Nancy Kelly
Gayle Lapointe
Mira Lavoie
Sonia Lesvesque
Denis Normandin
Karen Ortega
Roberta Rasori
Josianne Rozon
Wayne Sturkenboom
Simon Wong
Chemistry
Kevin Bateman
Christopher Bayly
Cameron Black
Michael Boyd
Nathalie Chauret
Sheldon Crane
Stephen Day
Jacques Yves Gauthier
Robert Houle
Elise Isabel
C.K. Lau
Serge Leger
Tammy LeRiche
Jean-Francois Levesque
Chun-Sing Li
Christophe Mellon
Renata Oballa
Joel Robichaud
Bruno Roy
John Scheigetz
Carmai Seto
Michel Therien
Laird Trimble
Jean-Francois Truchon
Vouy-Linh Truong
Qingping Wang
Bone Biology (WP)
Process Chemistry
Don Kimmel
Paul O’Shea
Sevgi Rodan
Dean Bender
Gideon Rodan
Mirlinda Biba
Le Duong
Jennifer Chilenski
Pat Masarachia
Jimmy DaSilva
Brenda Pennypacker
Rich Desmond
Maureen Pickarski
Paul Devine
Gregg Wesolowski
Pete Dormer
Ya Zhuo
Bruce Foster
Pharmaceutical R&D
Don Guathier
Cynthia Bazin
Danny Gauvreau
Sophie-Dorothee Clas
Francis Gosselin
Elizabeth Kwong
Derek Henderson
Pauline Luk
Greg Hughes
Rafik Naccache
John Leazer
Wayne Parent
Bill Leonard
Suzanne Spagnoli
John Limanto
Dendi Susanto
Christian Nadeau
Hongshi Yu
Thorsten Rosner
Safety Assessment
Kara Rubin
Laura Gumprecht
Ali Shafiee
Cindy Fishman
Veena Upadhyay
Pharmacology
Chris Welch
Sylvie Toulmond
Clinical Development
Aubrey Stoch Tony Lombardi
Deborah Miller Albert Leung Celera
2
Osteoporosis: A Continuing Disease Burden
 Osteoporosis: “porous bones”
 A progressive bone disease characterized by decreased bone density
and mass. Leads to an increased risk of fracture
Trabecular bone deteriorated
Normal
Osteoporosis
From the International Osteoporosis Foundation
 Estimated 200 million women worldwide with osteoporosis.1
 Only ~20% of osteoporosis patients are currently treated
 Hip fractures cause high morbidity and mortality2
− ~20% die in the 1st year
− ~33% are totally dependent or in a nursing home after one year
1 Kanis
JA (2007) WHO Technical Report, University of Sheffield, UK: 66
Osteoporosis Foundation statistics.
2 International
3
Bone Remodeling
Healthy: Bone resorption = formation
Osteoporosis: Bone resorption > formation
4
Osteoclast – The Target Cell for Cathepsin K Inhibition
Trans- endocytosis
Electron micrograph of an
osteoclast on bone
N
N
Lysosomes
Sealing
Zone
Cathepsin K
pH=5
Collagen
Bone resorption by osteoclasts is the initial step in remodeling
Rodan & Duong Bone Key 2008, 5, 6
5
Cathepsin K Deficiency
 Discovered in 19951
– Highly expressed in osteoclast
– Cleaves at multiple sites in triple helical region of
collagen type I and II
 Pycnodysostosis is a genetic disease associated with
cathepsin K deficiency2,3
– Rare autosomal recessive skeletal dysplasia
– ~150 cases reported worldwide
– Short stature with high bone mass and increased
fragility associated with high risk of fracture
– Normal intelligence, sexual development and
lifespan
 Cat K null
Henri de Toulouse-Lautrec
mice4
1
Inaoka, et al, BBRC 1995, 206, 89
Schilling, et al, Osteoporosis Int 2007, 18, 659
3
Saftig PNAS 1998, 95, 13453.
4
Gelb, et al, Science 1996, 273, 1236
– Osteopetrotic phenotype
– Increased bone density
– Otherwise healthy and fertile
2
6
Cathepsin K as a Therapeutic Target

Data suggestive that cathepsin K represents a promising
target for treatment of osteoporosis
– Human and mouse genetic data
– Cathepsin K collagenase activity
– Restricted cellular distribution

Goal: To develop an orally active, selective and
reversible cathepsin K inhibitor for treatment of osteoporosis
7
Substrate to Inhibitor: Warhead
Use knowledge of mechanism to convert substrate into inhibitor:

HS-Enz
O
R
HO S-Enz
N
H
R'
R
N
H
R'
Tetrahedral
intermediate

O
O
H2O
- H2N-R'
R
S-Enz
+ Enz-SH
R
OH
Acyl enzyme
intermediate
Replace scissile bond with an electrophile (warhead) that reacts
with active site cysteine:
Palmer, J. T.; Rasnick, D.; Klaus, J. L.; Brömme, D. J. Med.
8 Chem. 1995, 38, 3193
Substrate to Inhibitor: Recognition Sequence
 Cathepsin K cleaves Type I collagen at multiple positions
 Recognition sequence1: Gly-Leu-Lys—Gly-His
P3 P2
P1
P1’ P2’
P2
P2
P1'
H
N
O
N
H
H
N
O
O
N
H
P3
O
H
N
N
HN
O
N
P1
NH2
N
P2'
N
H
H
N
O O
S
O
P3
P1
APC3328
 Recognition sequence combined with electrophilic warhead gives a mechanism
based-inhibitor of cathepsin K2,3
1Atley,
M.; Mort, J. S.; Lalumiere, M.; Eyre, D. R. Bone 2000, 26, 241
M. E.; Klaus, J. L; Barnes, M. G.; Brömme, D. Nat. Struct. Biol. 1997, 4, 105
3Brömme, D.; Klaus, J. L.; Okamoto, K.; Rasnick, D.; Palmer, J. T Biochem. J. 1996, 315, 85
9
2McGrath,
Opportunities for Inhibitor Design
• Co-crystal reveals key interactions with cathepsin K and opportunities
for obtaining selectivity
– S1 pocket is a narrow groove on enzyme
– S2 pocket suggests accommodation of alternative hydrophobic groups
– S3 sub-site contains aspartic acid (Asp61) that may be exploitable for
selectivity
– P2-P1 amide forms key interaction with enzyme backbone (Gly66, Asn158)
– P2-P3 amide: NH hydrogen-bond to Gly66, oxygen points towards solvent
S2
O
S3
N
HN
N
H
H
N
O O
S
O
S1
McGrath, M. E.; Klaus, J. L; Barnes, M. G.; Brömme, D Nat. Struct. Biol. 1997, 4, 105
10
Reversible Inhibitors
 Irreversible inhibitors have potential liabilities
 Vinyl sulfone can be replaced by nitrile (electrophilic)
– Adds cysteine reversibly to form thioimidate
O
N
HN
N
H
H
N
O
O
Enz-SH
S
O
Nitrile-based inhibitor

Optimization to dipeptide nitrile
L-006235
Cat K IC50 = 0.2 nM
J. Palmer. et. al. J. Med. Chem. 2005, 48, 7520
11
Selective Inhibition of Cathepsin K is Important
to Avoid Potential Off
-target Activity
Off-target
Cathepsins
Cysteine Cathepsins
B, C, F, H, K, L, O, S, V, W, Z
Aspartyl Cathepsins
D, E
Serine Cathepsins
A, G
 Fifteen human cathepsins are present inside cells within acidic
compartments called lysosomes
 Potential consequences of off-target inhibition
− Cat L: cardiomyopathy, impaired neovascularization
− Cat F: neuronal ceroid lipofuscinosis
− Cat S: impaired immune response
 High selectivity for cathepsin K desired to avoid potential off-target
effects
F. Lecaille Chem. Rev. 2002, 102, 4459.
12
L-006235 : A Potent, Selective, Reversible Dipeptidic
Nitrile Inhibitor of Cathepsin K
Cat K
0.2
IC50 (nM)
Cat B
Cat L
Cat S
Cat F
1100
47000
3000
6300
– >5000-fold selective in purified enzyme assays
– Active in rabbit and monkey models of osteoporosis
J. Palmer. et. al. J. Med. Chem. 2005, 48, 7520
13
LL-006235
-006235 in Cathepsin K
Asp61
2.9
2.9
Asn166
Cys25
3.0
Gly66
14
Loss of Specificity of Cathepsin K Inhibition:
Whole Cell Assay
Inhibition of Cathepsins, IC50 (nM)
Cathepsin K
Enzyme Cell
L-006235
0.2
5
Cathepsin B
Cathepsin L
Enzyme
Cell
Enzyme
Cell
1100
17
6300
340
Cathepsin S
Enzyme
47000
L-006235
– High selectivity profile of L-006235 is lost in whole cell assays
J. Palmer. et. al. J. Med. Chem. 2005, 48, 7520
15
Cell
790
Hypothesis: Basic Inhibitors Accumulate in
Acidic Lysosomes and Inhibit Off-Target
Cathepsins
 Lysosomotropism: accumulation of lipophilic weak bases in acidic
sub-cellular compartments
Lysosome
Plasma Membrane
pH ~7
pH ~5
BH+
L-006235
Merged
B
B
pH ~7
DNDDND-99 (lyso)
L-006,235
Co-localization in live HUVECs
weak basic
amine
J. Falgueyret et. al. J. Med. Chem 2005, 48, 7535
16
Cat K Inhibitors in Whole Cell Assays
Inhibition of Cathepsins, IC50 (nM)
Cathepsin K
Enzyme
Enzyme
Cell
0.2
2.3
1100
4200
17
2900
L-006235
Cmpd 2
O
Basic
N
N
N
S
Cathepsin L
Cathepsin B
N
H
H
N
Enzyme
6300
24000
O
O
L-006235
Cell
Enzyme
N
N
S
N
H
H
N
O
Cmpd 2
– Non-basic analogs show reduced activity in cell-based assays
– Non-basic P3 groups are generally less active
J. Falgueyret et. al. J. Med. Chem 2005, 48, 7535
17
Cell
340
47000
>10000 27000
O
Neutral
N
Cathepsin S
N
790
>10000
Why have a Basic Substituent?
 10- to 100-fold increase in intrinsic potency
 Improved enzyme selectivity
 Enhanced pharmacokinetics (bioavailability, half-life)
 Liability of basic group:
– Lysosomotropism leading to loss of functional selectivity
18
SAR Optimization Efforts
N
N
N
N
O
O
O
N
N
N
N
H
N
H
N
CN
O
O
S
Ar
N
R
S
R
O O
S
R
R
H
N
O
– Extensive SAR studies on dipeptide framework resulted in loss of
selectivity and/or potency
– Abandoned dipeptide series of inhibitors
19
Amide Replacement

P2-P1 amide forms key interaction with enzyme backbone (Gly66, Asn158)
but, P2-P3 amide NH hydrogen-bond to Gly66, oxygen points towards solvent
O
Loss of
N
H
selectivity
Cat K =
Cat L =
Cat S =
Cat B =
73 nM
3666 nM (50x)
3401 nM (46x)
>10000 nM (137x)
CN
O
Cat K = 51 nM
Cat L = 42 nM (0.8x)
Cat S = 41 nM (0.8x)
Cat B = 1903 nM (37x)
Cat K = 56 nM
Cat L = 498 nM (9x)
Cat S = 1578 nM (28x)
Cat B = >10000 nM (>180)
Increase in
selectivity
 Amide bond can be replaced but loss in Cat K potency
 Selectivity can still be achieved without P2-P3 amide
J. Robichaud et. al. BMCL 2004, 14, 4291
H
N
20
Further Optimization of Amide Replacement
 High selectivity can still be achieved without P2-P3 amide with addition of
basic P3 group
HN
N
H
N
Potency
O
selectivity
(R)-2
Cat K = 56 nM
Cat L = 498 nM (9x)
Cat S = 1578 nM (28x)
Cat B = >10000 nM (>180)
J. Robichaud et. al. BMCL 2004, 14, 4291
J. Robichaud et. al. J. Med. Chem. 2003, 46, 3709
Cat K = 11 nM
Cat L = 3950 nM (359x)
Cat S = 2010 nM (183x)
Cat B = 3725 nM (339x)
21
CN
Re-Addition of a Hydrogen-Bond Donor
 Additional of hydrogen bond donor is favorable for cat K potency.
X = Cat K (IC50)
NH
54 nM
O
1147 nM
CH2 2353 nM
J. Robichaud et. al. BMCL 2004, 14, 4291
X = Cat K (IC50)
NH
72 nM
O
1475 nM
CH2 3394 nM
22
X = Cat K (IC50)
NH
208 nM
O
2999 nM
CH2 2349 nM
Potency and Selectivity Optimization
selectivity
Cat K =
Cat L =
Cat S =
Cat B =
2 nM
108 nM (54x)
24 nM (12x)
26 nM (13x)
Cat K =
Cat L =
Cat S =
Cat B =
3 nM
2101 nM (700x)
158 nM (53x)
1812 nM (604x)
 Selectivity and potency can still be achieved without P2-P3 amide with
addition of basic P3 group
HN
N
Cat K = 11 nM
H
Cat L = 3950 nM (359x)
N CN
Cat S = 2010 nM (183x)
O
Cat B = 3725 nM (339x)
(R)-2
J. Robichaud et. al. BMCL 2004, 14, 4291
23
Another Peptidomimetic
O
N
N
N
N
H
H
N
CN
O
S
How can P2-amide be replaced while maintaining H-bond capability?
 H-bond donor requires non-basic nitrogen
– Non-basic amine (pKa = 1.5) is not protonated at physiological pH
 Retains the H-bond donating properties of an amide bond
O
O
N
H
CF3 O
O
N
H
M. Zanda et. al ChemMedChem 2007, 2,1693
O
24
N
H
N
H
CF3 Amide Replacement: P2 Leucine vs Cyclohexyl
50
50
O
O
N
H
H
N
CN
N
H
O
H
N
CN
O
 Trifluoroethylamine is an effective amide isostere
 Cyclohexyl ring in P2 is much less potent than the isobutyl side-chain
 Modeling suggests loss of potency in cyclohexyl analog is a result of
steric interactions between P2 and the CF3 group
W. C. Black et. al. BMCL 2005, 15, 4741
Black, W. C.; Percival, M. D. ChemBioChem 2006, 7, 1525
25
CF3 Amide Replacement
IC50 (nM)
O
L-006235
N
N
N
S
3
N
N
N
N
H
H
N
Cat B
Cat L
Cat S
0.2
1100
6300
47000
0.03
3315
26
848
0.005
1110
47
451
CN
O
CF3
N
H
Cat K
H
N
CN
O
S
CF3
N
H
4
H
N
O
CN
N
HN
 Phenyl piperazine has greater potency and selectivity than thiazole piperazine
 Can this exquiste potency/selectivity advantage be extended to non-basic
inhibitors?
W. C. Black et. al. BMCL 2005, 15, 4741
26
“Non-basic” P3 Substituents
L-873724
 Non-basic derivatives in this series are potent and selective
 L-873724 stands out as best compound in this series
C. S. Li et al. BMCL 1985,16, 1985
27
X-ray structure of
Why
are
Trifluoroethylamines
so
Potent?
Cat K -Complex
CF3
H2NO2S
28
N
H
H
N
O
CN
L-873724 has Similar Potency in Whole Cells and
Purified Cathepsins
 Trifluoroethylamine isostere gave enough potency to remove the basic P3
 Selectivity profile of L-873724 is maintained in whole cell assays
O
N
N
N
H
N
S
H
N
CF3
H
N
CN
N
H
O
MeO2S
L-006235
CN
O
L-873724
Inhibition of Cathepsins, IC50 (nM)
Cathepsin B
Cathepsin L
Cathepsin S
Enzyme
Cell
Enzyme
Cell
Enzyme
Cell
L-006235
1100
17
6300
340
47000
790
L-873724
5240
4800
264
1220
178
94
C. S. Li et al. BMCL 1985,16, 1985
29
L-873724 Metabolism
 L-872724 is extensively metabolized in human hepatocytes (56%
metabolism)
CF3
H
N
N
H
NC 70141-2225 hum heps L-873,724 t=2
25Oct02-014
100
18.92
Human hepatocytes
2: Diode Array
263.801_271.9
5.22e4
N
O
MeSO2
L-873,724
~50% metabolism
OH
M7
M12
16.12
CF3
17.63
N
H
%
R
M13
H
N
CF3
N
N
H
O
R
M7: leucine hydroxylation
17.85
?
M8
14.45 14.72
OH
15.03 15.77
CF3
10
N
H
Time
12.00
13.00
14.00
15.00
16.00
O
M13: acid
?
19.58
OH
17.00
18.00
19.00
R
N
H
O
M12: lactone formation
30
CF3
O
R
OH
O
M8: hydroxyacid
Modifications to Reduce Metabolism
Modify P2 to block metabolism
CF3
N
H
H
N
N
O
Add substituent to block amide
cleavage and lactonization
MeO2S
31
L-873,724 has Similar Potency in Whole Cells
and Purified Cathepsins
 Modifications result in odanacatib which maintains excellent selectivity
and activity in both enzymatic and cell-based assays
L-873724
Odanacatib
Inhibition of Cathepsins, IC50 (nM)
Cathepsin K
Enzyme Cell
L-873724
Odanacatib
0.2
0.2
3
5
Cathepsin B
Cathepsin L
Cathepsin S
Enzyme
Cell
Enzyme
Cell
Enzyme
Cell
5240
1034
4800
1050
264
2995
1220
4843
178
60
94
45
J. Gauthier et. al. BMCL 2008, 18, 923
32
Odanacatib has Improved Pharmacokinetics in
Preclinical Species
 Modification of L-873724 to block sites of oxidative metabolism provides
a superior pharmacokinetic profile.
 High recovery (>99%) after incubation with dog and human hepatocytes
Odanacatib
L-873724
Pharmacokinetics
Dog
Rat
Cl(ml/min/kg)
L-873724
Odanacatib
6.3
2
J. Gauthier et. al. BMCL 2008, 18, 923
t1/2 (h)
2.7
6
Cl(ml/min/kg)
1.12
0.13
33
Rhesus Monkey
t1/2 (h)
6
57
Cl(ml/min/kg) t1/2 (h)
7.5
6.1
1.7
18
Mean Concentration Profile of Odanacatib in Women
Administered Once-Weekly Doses for 3 Weeks
0
0
6
W
O
g
m
5
2
Cat B,L,S,F IC90
W
O
g
m
0
5
0
0
5
W
O
g
m
0
0
1
0
0
4
0
0
3
0
0
2
M
n
,
n
o
i
t
a
r
t
n
e
c
n
o
C
0
0
1
Cat K IC90
(bone res)
0
4
1
3
1
2
1
1
1
03
1k
9e
e
8W
7
s
6y
a
5D
4
3
2
1
0
7
6
1
5k
e
4e
W
3
2s
y
1a
D
0
︵
︶
︵
︶
– Human half-life similar to dog; human t1/2 = 60-70 h
– At therapeutic exposures, odanacatib maintains selectivity over other cathepsins
Stoch, A. et. al. Clin. Pharm. & Ther. 2009, 86, 175
34
Effect of Odanacatib on Bone Resorption After 3
Weeks of Once
-Weekly Dosing
Once-Weekly
S-CTx Mean Percent Change (95%CI) from Baseline
S-CTX-I % Change from Baseline
75
Placebo
50 mg
5 mg
100 mg
25 mg
50
25
0
-25
-50
-75
-100
0 1
3
5
7
9
11
14
16
18
21
24
Day
Dose
Dose
Stoch, A. et. al. Clin. Pharm. & Ther. 2009, 86, 175
Dose
N = 9-12/group
35
28
Selectivity of Other Cathepsin K Inhibitors
 Relacatib has low selectivity towards related cathepsins (L, V, B, S) in
enzymatic assays
− Relacatib clinical development was halted after Phase I studies
 Balicatib is very selective towards related cathepsins in enzymatic assays
− Development of balicatib was discontinued during Phase IIb due to
morphea-like skin changes
− Cathepsin B, K, L, S and V are all expressed in human skin and have
collagenase activity
− Postulate that morphea is due to lysosomotropic-based loss of selectivity
Basic
O
O
N
H
H
N
O
N
H
O
O
N
N
O
S
N
N
O
Relacatib (GSK)
Brömme, D. Expert Opin. Investig. Drugs. 2009, 18, 585
Desmarais, S. et al Molecular Pharmacology 2008, 73, 147
Balicatib (Novartis)
36
H
N
O
CN
Odanacatib has Similar Potency in Whole Cells
and Purified Cathepsins
Inhibition of Cathepsins, IC50 (nM)
Cathepsin K
Enzyme Cell
Odanacatib
Balicatib
0.2
0.6
5
22
Cathepsin B
Cathepsin L
Enzyme
Cell
Enzyme
Cell
1034
4800
1050
61
2995
503
4843
48
Cathepsin S
Enzyme
45
60
65000 2900
Basic
Odanacatib (MK-0822)
Balicatib
The low level of off-target activity of odanacatib in enzyme
assays is maintained in whole cell assays
37
Cell
Effect of Cat K Inhibitors on Collagen Accumulation
in Dermal Fibroblasts
(Fluorescence intensity, fold Increased over vehicle)
Intracellular Collagen Accumulation
 Develop model of intracellular collagen accumulation in primary human
dermal fibroblasts
7
6
5
Balicatib
Relacatib
Odanacatib
4
3
2
1
0
0.01
0.1
1
10
Inhibitor Concentration (µM)
 Odanacatib has a minimal effect on intracellular collagen accumulation
compared to balicatib and the pan-cathepsin inhibitor relacatib
J. Gauthier et. al. BMCL 2008, 18, 923
38
Odanacatib Inhibits Bone Resorption with Less
Reduction of Bone Formation
Bone formation factors
IGF-1, TGF-BMP. Wnt 10b
 Bisphosphonates and
denosumab reduce the
number of osteoclasts
−Fewer resorption pits,
reduced new bone
formation
Osteoblast (Ob)
Ob progenitor
Osteoclast
 Odanacatib reduces the
activity of Cat K in the
osteoclast
Vehicle
−Same number of, but
shallower resorption pits
−Allows new bone
formation
Odanacatib
O
O
C
C
P
P
O
O
C
C
OC: Osteoclast
P: Resorption Pit
Leung et al. Bone 2011, 49, 623
39
P
P
Differential Effects on Bone Turnover Markers

Odanacatib inhibits bone resorption (uNTx) while exhibiting less reduction of
bone formation (BSAP) than observed with alendronate.
Historical Comparison to Alendronate
Urinary NTx
Serum BSAP
1 Binkley
et al., ASBMR 2011
2 Merck data on file
Geometric Mean Percent Change from Baseline (SE)
Odanacatib Phase IIB1
20
0
20
Year
1
2
3
Alendronate Phase III2
4
5
0
-20
-20
-40
-40
-60
-60
-80
-80
-100
2
3
1
2
1
2
-100
Year
1
Year
4
5
20
20
0
0
-20
-20
-40
-40
-60
-60
-80
-80
-100
-100
40
3
4
5
4
5
Year
3
Odanacatib: Phase IIb Study 5-Year Data
Bone Mineral Density (DXA)
 Odanacatib progressively increases BMD over 5 years at lumbar spine, femoral
neck and total hip
Historical Comparison to Alendronate
Lumbar
Spine
Femoral
Neck
Total
Hip
Mean Percent Change from Baseline (±SE)
Odanacatib Phase IIB1
Alendronate Phase III2
15
15
10
10
5
5
0
1
2
3
Year
4
5
0
10
10
5
5
0
1
2
10
3
Year
4
5
2
1
2
1
2
10
3
Year
4
5
3
4
5
3
4
5
Year
5
5
0
0
1
1
2
3
Year
4
0
5
41
Year
Summary
 Early inhibitors were found to accumulate in lysosomes thereby lowering
their selectivity towards related cathepsins in functional assays
 Optimized leads produced selective non-basic inhibitors
 Optimized non-basic leads reduced metabolism and improved
pharmacokinetic properties
 In post-menopausal women, odanacatib treatment:
− Demonstrates less suppression of markers of bone formation than has been
observed with alendronate*
− Progressively increases BMD over 5 years at lumbar spine, femoral neck,
and total hip
* Historical comparison
42