Transcript p value - Toadovision

Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Welcome and Meeting Objectives
David Goldsmith
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics Advisory Board
Introduction
David Goldsmith
Aims and objectives 1
Strengths and gaps DCOR 2001-2004 CMS and RIND
results
•
Which data are compelling and why? What are the key conclusions
about Renagel? Which data are inconclusive?
•
What are the remaining gaps and how can we explain these gaps
using other existing evidence outside of DCOR (prospective trials,
retrospective studies, epidemiologic data, etc)?
•
What are the remaining gaps for future research?
Aims and objectives 2
Strengths and gaps of the DCOR/RIND cost
effectiveness analysis
•
•
•
•
Is the overall framework of the model robust and supportable from a
clinical perspective?
Is the overall framework of the model robust, transparent and does
it meet the standards set by NICE?
Is the data used appropriately from DCOR and other sources, or
are there better sources of data
Are the assumptions of the model reasonable and accurate, or how
can we improve them further from a clinical perspective?
Aims and objectives 3
How best to communicate the evidence to NICE,
reimbursement and guidelines authorities
•
•
•
•
What are the key benefits of Renagel that should be communicated
to reimbursement authorities and guidelines/listings?
How should the Renagel clinical-economic story be told?
Is the evidence convincing enough and how simple/complex should
the story be made? What is the sequence of the critical messages
and supporting data in that story?
What targets (first-line, sub-groups, etc) should we be aiming for
versus expect in the NICE CKD guidelines
What we know and what we don’t know
As we know, there are known
knowns. There are things we know
we know. We also know there are
known unknowns. That is to say we
know there are some things we do
not know. But there are also
unknown unknowns, the ones we
don't know we don't know.
The CKD Forecast for England
No. of patients
70000
60000
50000
Transplant
PD
HD
40000
30000
20000
10000
0
2000
2005
2010
2015
Year
2020
2025
2030
Roderick et al 2004
Dialysis – Tariffs
Financial envelope (a strait-jacket ?)
•
•
•
•
Dialysis
Hospitalizations
Transport / outpatients
Drug costs
–
–
–
–
EPO
phosphate binders
ACEI / ARB
statins
DRUGS
FIXED DIALYSIS COSTS
ERYTHROPOIETINS
£3000+
CINACALCET
£3-9000
SEVELAMER / FOSRENOL
£1500
DRUGS
FIXED DIALYSIS COSTS
ERYTHROPOIETINS
DRUGS
FIXED DIALYSIS COSTS
£3000+
ERYTHROPOIETINS
£6000+
CINACALCET
£9000
SEVELAMER / FOSRENOL
£3000
PARACALCITOL
£2000
DRUGS
FIXED DIALYSIS COSTS
DCOR 2-year CMS data summary - Mortality
• Primary endpoint (all-cause mortality): No significant
difference between sevelamer and calcium (22% vs 23%).
• All-cause mortality in patients aged ≥65 years: Significant
difference in favour of sevelamer compared to calcium (28%
vs 32%).
• Cardiovascular mortality: No significant difference between
treatment groups (11.8% vs 11.7%).
DCOR 2 year summary - Hospitalization
• Trend towards fewer all-cause, CV and other hospitalizations
with sevelamer vs calcium.
• Anderson-Gill regression models showed significantly lower
relative risk of all-cause hospitalization with sevelamer vs
calcium (RR 0.88, 95%, p=0.011) and for other hospitalization
(RR 0.82, 95%, p=0.0045).
• Sevelamer patients had significantly fewer days hospitalized
for other causes compared to calcium group and a trend
toward fewer days hospitalized for all causes (p=0.0644) and
CV causes (p=0.0582).
DCOR 2 year summary - Medicare expenditures
and clinical events
• Sevelamer patients had lower mean total, inpatient, skilled
nursing facility and other Medicare allowable expenditures vs
calcium patients primarily due to a difference in inpatient costs
($1546 vs $1779 PMPM).
• Sevelamer patients used significantly more vitamin D.
• The incidence of clinical events was lower for sevelamer in
22/30 types of inpatient event and 17/19 types of outpatient
event.
RIND Kidney International 2005
• Subjects with no evidence of coronary calcification at baseline
showed little evidence of disease development over 18
months (independent of phosphate binder therapy), while
subjects with at least mild calcification had significant
progression at 6, 12 and 18 months.
• Subjects treated with calcium showed more rapid and more
severe increases in coronary artery calcification score (CACS)
compared to those on sevelamer (p=0.056 at 12 months;
p=0.01 at 18 months).
RIND Mortality (Spiegel, Poster at NKF 2006)
• 114 patients from the primary RIND study were followed for up to 5
years to determine the relationship between phosphate binder and
mortality in patients new to haemodialysis.
• Baseline CACS was a strong predictor of mortality.
• Mortality was significantly lower in subjects randomized to
sevelamer vs calcium (p=0.0214).
• There was a trend toward improved survival with sevelamer for all
levels of baseline CACS, even when adjusted for baseline CV risk
using the Framingham Risk Index.
• Conclusion: The results provide strong evidence that phosphate
binder selection and calcification are integrally linked with mortality
in dialysis patients.
What we don’t know
•
•
•
•
Where does new data take us?
Is the current model fit for purpose?
Is a new model needed and what should it be like?
Is DCOR strong enough alone or do other data need to
be combined with it (eg RIND) ?
• Do DCOR and/or RIND change current clinical practice?
• How can the rationale convince in the clinical setting, ie
can it support arguments of cost at the clinical level?
Deliverables
• What is the minimum/probable/best destination for sevelamer?
• What are the specific targets en route and how can they be
reached?
• What specific actions are needed?
• What is a realistic timeframe?
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Business Overview
Doug Jermasek
Senior Vice President,
Global Marketing & New Product Development
The Genzyme Commitment
• Genzyme’s mission is to:
– Address serious diseases with unmet medical needs
– Provide breakthrough therapies and services that significantly improve patients’ lives
• Kidney disease is an illness in need of new solutions
– The complexity and severity of kidney disease has lead Genzyme to focus on many
aspects of renal care
• Genzyme has made a long-term commitment to renal
care
–
–
–
–
–
Clinical research
Product development
Education
Access
Support
Our Global Corporation
• Founded in 1981
• From small start up to more than 8,000 employees
• Headquarters in Cambridge, MA, and presence in over
30 countries
• Helping patients in more than 80 countries
• 14 manufacturing sites
• 9 genetic testing lab sites
Our Focus: Seeking Frontiers
Serious diseases with
unmet medical needs
+
Breakthrough therapies
and services
 Significant improvement
to patients’ lives
=




Patient-focused
Variety of technology
platforms
Life-changing standard of
care products
Creating sustainable value
Our Technology Platforms
•
•
•
•
•
•
•
•
Protein therapies
Polymer therapeutics
Small molecule therapeutics
Biomaterials
Cell therapy
Gene therapy
Diagnostics products and services
Pharmaceuticals
Our Products
LSDs
Renal
Orthopaedics
Transplant &
Immune Disease
Oncology/
Endocrinology
Campath®
Cerezyme®
Renagel®
Synvisc®
Thymoglobulin®
imiglucerase for injection
sevelamer hydrochloride
hylan G-F 20
Anti-thymocyte Globulin [Rabbit] alemtuzumab for injection
Fabrazyme®
Hectorol®
Carticel®
Clolar®
doxercalciferol
autologous cultured
chondrocytes
clofarabine
agalsidase beta
Thyrogen®
Aldurazyme®
thyrotropin alfa for injection
laronidase
Diagnostic Products and Services
Our Late-Stage Pipeline
Research
Pre-Clinical
Myozyme - Pompe disease
Synvisc - for hip U.S.
Tolevamer - C. difficile
Hylastan - OA of knee
Sevelamer Carbonate - kidney disease
DX-88 - hereditary angioedema
Campath - multiple sclerosis
Tasidotin HCL - cancer
Gene therapy - peripheral arterial disease
Cell therapy - ventricular restoration
Iron Chelator - iron overload diseases
Thyrogen - goiter
DENSPM - liver cancer
Phase 1
Phase 2
Phase 3
Phase 4
Renal Disease is a growing concern
1
CURRENTLY THERE ARE OVER 1 MILLION
DIALYSIS PATIENTS WORLDWIDE. THIS IS
EXPECTED TO DOUBLE IN THE NEXT DECADE.2
Majority of CKD patients die before reaching Stage 5
1.
2.
3.
4.
U.S. Renal Data System, USRDS 2004 Annual Data Report; Atlas of End-Stage Renal Disease in the United States. National
Institutes of Health, Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2004.
Lysaght MJ. Maintenance dialysis population dynamics: current trends and long-term implications. J Am Soc Nephrol.
2002;13:S37-40.
Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS. Prevalence of chronic kidney disease and decreased kidney function
in the adult US population: Third National Health and Nutrition Examination Survey. Am J Kidney Dis. 2003;41:1-12.
U.S. Renal Data System, USRDS 1998 Annual Data Report. National Institutes of Health, Institute of Diabetes and Digestive
and Kidney Diseases, Bethesda, MD, 1998. Cited by: National Kidney Foundation. K/DOQI Clinical Practice Guidelines for
Chronic Kidney Disease: Evaluation, Classification, and Stratification—Quick Reference Clinical Handbook. New York, NY:
National Kidney Foundation; 2002.
Our Growing Focus on Renal Disease
RBRT
PKD
FSGS
Bone & Mineral Disease
& Product Synergies
↓ Renal Function
↓ 1,25 D Production
Decreased VDR
expression
↑ PTH
Altered Parathyroid Gland Function
➙ Hyperplasia ➙ SHPT
↑ Phosphate Retention
↓ Ca+
↑ PO4
Hyperphosphatemia
CONSEQUENCES
Renal Osteodystrophy Fractures Calcification CV Disease
MORBIDITY & MORTALITY
Genzyme Renal Continuum of Care
Hectorol ®
®
Renagel
LR-103
Underlying Disease
HTN
BCI-202
SHPT
Transplant
Thymoglobulin
Progression
DN
Deferitrin
CKD
CKD
Fabry
ESRD
®
Renagel
Renagel ®
Infection
ESRD
ESRD
Tolevamer
Fabrazyme ®
APKD
Sevelamer carbonate
Alternate Phosphate Binder
Internal dvpt
FSGS
GC-1008
®
CV Death
Acute Renal
Failure
RenaMed
Collaboration
Renal Division Pipeline
Alternate
Phosphate
Binder
Oncea-day
RenaMed
RBRT
Powder
Sevelamer
Carbonate
CKD
BCI 202
SHPT
Renagel
1.0 mcg
Tolevamer
Cyclacel
PKD
GC 1008
FSGS
Vials
Phase 2
Phase 1
Pre-Clinical
Hectorol
Phase 3
Commercialized
Products
Renal Replacement Therapy for ESRD
• From a health economics standpoint, caring for
ESRD patients is cost ineffective relative to other
healthcare investments
– yet most societies routinely provide dialysis
• Why are we so inconsistent in the way that we
evaluate the economics of dialysis-related services?
•
•
•
•
Dialyzers
Vascular access
Anemia management
Binders
• In for a penny, in for a pound?
Lest we forget…
• Behind the products…
…there are economics
• Behind the economics…
…there is disease
• Behind the disease…
…there are patients
Thank You!
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Renagel Clinical Overview
Robert Guiberteau
Great Fosters, UK
28th -29th June 2006
Clinical Overview
Health Economics
Advisory Board
Great Fosters June 28-29th 2006
Robert Guiberteau MD
VP Global Medical Programs
consequences of hyperphosphatemia
from the 60’s to the 90’s
Pi
Ca **
PTH Resistance
Calcitriol
Calcitriol Resistance
PTH Secretion
Parathyroid Cell Growth
Increased Risk
of Bone Disease
Increased Risk
of Bone Disease
consequences of hyperphosphatemia
since 1998
P
Calcitriol
Ca++
Calcitriol
Resistance
PTH Resistance
PTH Secretion
Risk of Calcification
Parathyroid
Cell Growth
Mortality
the history of the paradigm shift
– 1996
Rapid progression of Coronary calcifications
Braun
– 1997
Negative Calcium Balance???
Hsu
– 1998
Hyperphosphatemia as independant RRF
Block
– 1999
Hyperphosphatemia, a silent killer?
Amann
– 2000
Oral Calcium load is an independant RRF
Goodman
– 2001
Calcifications predict outcome
Blacher
– 2002
Calcifications, a modifiable risk factor
Chertow
– 2004
Increased S.Calcium as independant RRF
DOPPS
– 2004
Low PTH > High PTH as an increased RRF
Stevens
High Serum Phosphorus Increases Mortality Risk
Serum phosphorus >6.5 mg/dL in 39% of patients
2
1.5
Relative
Mortality
Risk (RR)
1.39**
1.18*
1.02
1.00
1.00
1
*P = 0.03
** P < 0.0001
REFERENCE
0.5
1.1 - 4.5
4.5 - 5.5
5.6 - 6.5
6.6 - 7.8
7.9 - 16.9
Serum Phosphorus Quintile (mg/dL)
Block, et al. Am J Kidney Dis. 31:607-17. 1998
association between mortality risk
and serum phosphate in CKD patients
Relative risk of death
2.2
Reference group
2.0
n=40538
1.8
1.6
1.4
1.2
1.0
0.8
0
<3
3–4
4–5
5–6
6–7
7–8
8–9
>9
Serum phosphorus (mg/dL)
Multivariable-adjusted RR of death
Block et al. JASN 2004 15: 2208–18
relationship between serum phosphate
and morbi / mortality in patients with normal GFR
Tonelli M. Circulation 2005 112:2627-33
the history of the paradigm shift
– 1996
Rapid progression of Coronary calcifications
Braun
– 1997
Negative Calcium Balance???
Hsu
– 1998
Hyperphosphatemia as independant RRF
Block
– 1999
Hyperphosphatemia, a silent killer?
Amann
– 2000
Oral Calcium load is an independant RRF
Goodman
– 2001
Calcifications predict outcome
Blacher
– 2002
Calcifications, a modifiable risk factor
Chertow
– 2004
Increased S.Calcium as independant RRF
DOPPS
– 2004
Low PTH > High PTH as an increased RRF
Stevens
coronary calcifications in young adults
Calcification Score
10000
1000
100
10
1
0.1
0
5
10
15
20
25
30
35
Age (years)
Goodman WG et al. N Engl J Med. 2000;342:1478-1483.
arterial calcification score
and daily dose of Ca containing PO4 binders
CaCO3 (g Ca element/day)
5.0
P<0.0001
4.2
3.3
2.5
1.7
0.8
0.0
0
1
2
3
4
Arterial calcification score
Guerin et al NDT 2002
the history of the paradigm shift
– 1996
Rapid progression of Coronary calcifications
Braun
– 1997
Negative Calcium Balance???
Hsu
– 1998
Hyperphosphatemia as independant RRF
Block
– 1999
Hyperphosphatemia, a silent killer?
Amann
– 2000
Oral Calcium load is an independant RRF
Goodman
– 2001
Calcifications predict outcome
Blacher
– 2002
Calcifications, a modifiable risk factor
Chertow
– 2004
Increased S.Calcium as independant RRF
DOPPS
– 2004
Low PTH > High PTH as an increased RRF
Stevens
calcification score
•
Probality of all-cause survival according to calcification score. Comparison
(log-rank test) between curves was highly significant ( Chi D =42.66 ; P<0.0001).
Probality of survival
1
Calcification score : 0
0.75
Calcification score : 1
Calcification score : 2
0.50
Calcification score : 3
0.25
Calcification score : 4
0
0
20
40
60
Duration of follow-up (months)
80
Blacher et al Hypertension 2001
smooth muscle cell matrix mineralization
Adapted from Yang Kidney Int 2004 66:2293-9
arterial calcification score and CRP
in ESRD patients
50.0
CRP (mg/l)
41.7
33.3
P<0.0001
25.0
16.7
8.3
0.0
0
1
2
3
4
Arterial calcification score
London et al NDT 2003
absorption of phosphate binder
• Calcium
: up to 20 %
• Aluminium
: 0.06 – 0.10 %1
• Lanthanum
: 0.00003 %2
1 De
2
Broe ME et al. NDT 19:114-118, 2004
Joy MS Am J Kidn Dis 42:96-107, 2003
renagel a new phosphate binder
N H2
H3P O4
N H3C l
N H2
Combination of anionic and hydrogen bonding.
NH
H
O
N H3+ -O P O H
O
H
N H2
+ HCl
efficacy
Ca (TTG)*
Renagel (TTG)*
Lanthanum**
Ca (CARE)***
8.5
8
Serum P (mg/dl)
7.5
7
6.5
6
K/DOQI Limit
5.5
5
4.5
0
4
8
12
16
20
24
28
32
36
40
44
48
52
Weeks
* Chertow G.
** Hutchison A.
*** Qunibi W. et al.
KI 2002
WCN 2003Berlin
KI 2004 65: 1914-1926
the history of the paradigm shift
– 1996
Rapid progression of Coronary calcifications
Braun
– 1997
Negative Calcium Balance???
Hsu
– 1998
Hyperphosphatemia as independant RRF
Block
– 1999
Hyperphosphatemia, a silent killer?
Amann
– 2000
Oral Calcium load is an independant RRF
Goodman
– 2001
Calcifications predict outcome
Blacher
– 2002
Calcifications, a modifiable risk factor
Chertow
– 2004
Increased S.Calcium as independant RRF
DOPPS
– 2004
Low PTH > High PTH as an increased RRF
Stevens
Median Percentage Change
TTG: percentage change in coronary
calcification scores at 52 weeks
35
30
25%*
25
20
15
10
6%
5
0
Calcium
Sevelamer
*Within treatment P<0.0001; between treatment groups P=0.02.
Patients with a baseline score >30.
Chertow KI 2002
TTG: percentage change in coronary
calcification scores at 2 years
100
80
Sevelamer (n=17)
Calcium (n=23)
p**=0.040
p*=0.000
(%)
60
p**=0.239
40
p**=0.129
p*=0.004
20
0
-20
p*=0.108
p*=0.020
p*=0.99
p*=0.40
week 26
p*: Wilcoxon signed rank test
week 52
final (1.8-2 yrs)
p**: Wilcoxon rank sum test
Asmus Nephrol Dial Transplant 2005 20:1653-61
RIND: Effects of Sevelamer and Calcium
on Coronary Artery Calcification
In Patients New to Hemodialysis
P=0.01
Median Increase CACS
140
120
100
80
60
40
20
0
Sevelamer
Ca Salts
The Median Absolute Increase in CACS at 18 Months Was 11-Fold Greater in
the Calcium Treated Group Compared to the Sevelamer-Treated Group
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
TTG: sevelamer and bone density
3
Percent change (median)
2
*
1
0
-1
Sevelamer
-2
Calcium
-3
-4
-5
-6
-7
*
Trabecular**
*Wilcoxon signed rank test, within group P<0.05
**Wilcoxon rank sum test, between group P<0.05
Cortical
Raggi et al. J Bone Miner Res. 2005;20:764–772
TTG:
change in bone attenuation
and coronary calcium score
Sevelamer
Calcium salts
80
15
60
10
40
5
20
5
20
10
Agatson score
Hounsfield units
20
100
40
Change in trabecular
bone attenuation
Change in coronary
calcium score
Raggi et al. J Bone Miner Res. 2005;20:764–772
Osteoblast surface (%) (%)
bone histomorphometry and arterial calcifications
in
ESRD
30.0
22.5
15.0
7.5
0.0
0
1
2
3
4
Arterial calcification score
London et al JASN 2004
P<0.0001
sevelamer hydrochloride restores bone
in osteopenic OVX rats
Bone
formation
Restoration
of lost bone
Bone
resorption
BV/TV, Tr.n
Osteocalcin
MAR, BFR
sC-telopeptide
BMD
BV/TV
Vukicevic
Abstract ASN 2005
Reversal of the Adynamic Bone Disorder and
Established Vascular Calcification in Chronic Kidney
Disease by Sevelamer Carbonate Therapy
• “The LDLR-/- high fat fed mouse exhibits an adynamic bone
disorder in the presence of CKD characterized by decreased
osteoblast surface, decreased bone formation rates and normal
osteoclast surfaces similar to the excess representation of the ABD
as the form of renal osteodystrophy observed in human diabetic
nephropathy.
• Treatment with sevelamer significantly increased osteoblast
surfaces and bone formation rates ameliorating the adynamic
bone disorder of the LDLR-/- high fat fed animals with CKD.”
Hruska K. Submitted for publication
Sevelamer and bone: clinical data
Impact of phosphate binders on bone
histomorphometry: results of a one-year
bone biopsy study
A. Ferreira J. Frazao H.Malluche
A. Ferreira J. Frazao H.Malluche
Submitted for publication
Trends in Activation Frequency
Renagel
Calcium
N=33
N=35
Away from normal
Towards normal
0%
9%
3%
12 %
No change
67 %
54 %
Towards normal
Away from normal
21 %
17 %
3%
14 %
renagel a efficient phosphate binder
with additional benefits……..
• Bioindicators
–
–
–
–
–
–
–
Lowering LDL C
Lowering biomarkers of inflammation
Lowering biomarkers of oxydative stress (animals)
Lowering uric acid
Lowering serum glucose
Binding to p cresol, indoxyl sulfate and indole (in vitro)
Increasing serum fetuin A
• Intermediate indicators
– Preserving RRF (animals)
– Halting progression of CV calcifications
– Maintening / Improving bone health
• Improving patient outcomes
new terminologies and recommendations
from osteodystrophy to CKD-MBD
Moe, Drueke Am J Kidney Dis 2004 43:553
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Outcomes Evidence:
DCOR Trial and CMS 3-Year results
RIND Trial Results
Jose Menoyo
Great Fosters, UK
28th -29th June 2006
Prevalence of Vascular Calcification in
CKD-IV
Patients New to Dialysis and Established Patients
100%
83%***
80%
57% **
60%
40% *
40%
20%
0%
Russo et al
*Russo et al AJKD 2004 (CrCl =33 ml/min)
**Spiegel D et al. Hemod Internat 2004: 8:265
***Chertow et al KI 2002
RIND
TTG
Effects of Sevelamer and
Calcium on Coronary Artery
Calcification in Patients New
to Hemodialysis
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
RIND Trial
Objective
• The primary aims of this study were to assess
the degree of coronary artery calcification in a
cohort of patients new to hemodialysis, and to
compare the impact of sevelamer vs. calciumbased phosphate binders on the development
and progression of coronary artery
calcification
Study Design
USUAL clinical practice
Maintain dialysate Ca=2.5 mEq/L
RANDOMIZE
w/in 90 days
of HD
Sevelamer
Extended treatment
Calcium
binder
Extended treatment
0 EBCT scan
6 mo 12 mo 18 mo EBCT scans
Titrate dose
P < 6.5 mg/dL
Ca2+ < 10.2 mg/dL
•Renagel patients can receive Ca supplementation at night.
•Dialysate Ca concentration was maintained at 2.5 mEq/l (1.25 mmol/l)
throughout the study period.
Titrate dose
P < 6.5 mg/dL
Ca2+ < 10.2 mg/dL
PTH 150-300 pg/mL
Effects of Sevelamer and Calcium on
Coronary Artery Calcification
In Patients New to Hemodialysis
Screened (N=385)
Declined/Screen Failures (N=237)
Randomized (N=148)
Calcium (N=75)
Sevelamer (N=73)
Baseline EBCT (N=67)
Baseline EBCT (N=62)
Adverse Event (N=1)
Transplanted (N=3)
Death (N=1)
Other (N=4)
Lost to Follow-Up (N=1)
Transferred to PD (N=2)
Adverse Event (N=1)
Transplanted (N=2)
Death (N=1)
Other (N=2)
Transferred to PD (N=2)
Available for Analysis (N=55)
(at Least 1 Post Baseline EBCT)
Available for Analysis (N=54)
(at Least 1 Post Baseline EBCT)
RIND Baseline Demographics
• Age:
• 42% Caucasian
•
• 32% African-American
•
•
•
•
•
•
•
59
BP:
149/80
Hg:
11.8
Ca+:
9.3
Phos:
5.3
Albumin:
3.6
PTH:
363
Creatinine:
7.4
Days from
1st HD to EBT: 84.6
• 62% Male
• 57% Diabetic
• 96% HTN
• 23% History of CAD
• 30% History of ASVD
• 16% History of CHF
Effects of Sevelamer and Calcium on
Coronary Artery Calcification
In Patients New to Hemodialysis
Sevelamer
Calcium
Phosphorus
5.2 (0.9)
5.1 (0.8)
Corrected Calcium
9.1 (0.5)
9.6 (.5)
47 (7)
49 (8)
PTH
298 (152)
243 (136)
P<0.05
LDL
60 (34)
81 (26)
P<0.05
Albumin
3.8 (0.3)
3.8 (0.4)
CRP
9.1 (9.7)
10.5 (10.3)
Hypercalcemia
(Ca>10.2mg/dl)
22%
54%
P<0.0001
Severe
Hypercalcemia
(Ca>11mg/dl)
5%
24%
P<0.02
Ca x P
Mean (SD)
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
P<0.05
Average Phosphorus Control by Binder
Mean Serum Phosphorus by Study Month
6.0
Serum Phosphorus
5.8
5.6
5.4
5.2
5.0
4.8
4.6
4.4
0
1
2
3
4
5
6
7
8
9
10
11
12
Study Month
Renagel
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
Calcium
13
14
15
16
17
18
Mean Serum Calcium Level by Binder
Mean Serum Calcium by Study Month
Serum Calcium (Corrected)
9.7
9.5
9.3
9.1
8.9
8.7
8.5
0
1
2
3
4
5
6
7
8
9
11
10
12
Study Month
Renagel
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
Calcium
13
14
15
16
17
18
Effects of Sevelamer and Calcium on
Coronary Artery Calcification
In Patients New to Hemodialysis
350
Median CACS
300
250
200
150
100
50
0
N=54 N=55
Baseline
N=51 N=53
6 months
N=45 N=47
12 months
Sevelamer
Calcium
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
N=40 N=45
18 months
Effects of Sevelamer and Calcium on
Coronary Artery Calcification
In Patients New to Hemodialysis
Median Increase CACS
140
P=0.01
120
100
80
60
40
20
0
Sevelamer
Ca Salts
The Median Absolute Increase in CACS at 18 Months Was 11-Fold
Greater in the Calcium Treated Group Compared to the
Sevelamer-Treated Group
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
Subjects With or Without Progression of
Coronary Artery Calcium Scores
Percentage Among Subjects with Baseline Coronary Artery Scores >30
Calcium
Sevelamer
6 Months
38
63
6 Months
62
12 Months
44
32
86
12 Months
14
56
93
60
18 Months
18 Months
40
Progression
7
Regression/Stable
Progression of CACS = Greater Than 15% Increase from Baseline
Fisher Exact Test P Value <0.05 for Between Group Differences at 12 and 18 Months
Block, GA, Kidney Int; Vol68(4): 1815-1824, 2005
Effects of Sevelamer and Calcium on
Coronary Artery Calcification
• Substantially greater proportion of patients
new to dialysis had no visible coronary
calcification compared with previous reports of
patients on chronic hemodialysis
• Subjects with no evidence of coronary artery
calcification upon initiation of dialysis showed
little evidence of coronary artery calcification
development during follow-up
Block, GA et al. Kidney Int 2005; 68:1815-1824
Calcium Containing Phosphate
Binders are Associated with
Increased Mortality Risk in
Hemodialysis Patients
Compared to
Sevelamer
DM Spiegel, P Raggi, A Bellasi, L Kooienga, GA
Block
NKF Spring Clinical Meeting 2006
Survival by Baseline
Coronary Calcium Score
Mortality Analysis Comparing Baseline Calcium Score
Survival Distribution Function
1.00
0.75
0.50
0.25
P = 0.0035
CCS=0
CCS< 400
CCS >= 400
0.00
0
6
12
18
24
30
36
42
48
54
60
Months
No. at Risk
CCS = 0
CCS < 400
CCS >= 400
43
35
36
41
34
34
41
33
32
38
29
28
32
25
23
18
14
15
4
1
4
66
Survival by Binder Randomization
Mortality Analysis Comparing Calcium Versus Sevelamer
1.00
Survival Distribution Function
10 Deaths
0.75
P = 0.0214
0.50
24 Deaths
0.25
Calcium
Renagel
0.00
0
6
12
18
24
30
36
42
48
54
60
Months
No. at Risk
Calcium
Renagel
60
54
57
52
54
52
49
46
38
42
22
25
5
4
66
Mortality by Baseline Calcification Score
Summary
• Sevelamer confers a survival benefit over
calcium-based phosphate binders
• Within each strata of coronary artery calcification
sevelamer demonstrated a improved survival
• Baseline CACS as measured by EBCT scan is a
strong predictor of survival in patients new to
dialysis
DCOR:
Dialysis Clinical Outcomes
Revisited
Outcomes Studies in Hemodialysis
HEMO Study (n=1,846)
–
–
Standard or high dose of dialysis and a low or high flux dialyzer
Risk of death from any cause was same for both dialyzer doses
and both dialysis groups
U.S. Normal Hematocrit Trial (n=1,223)
–
Patients with clinical evidence of congestive heart failure or
ischemic heart disease
– Study terminated after 29 months because the group targeted to
normal values had a higher mortality rate
4D (n=1,255)
–
Diabetic patients randomized to atorvastatin/day or matching
placebo
– Atorvastatin had no statistically significant effect on
cardiovascular death, nonfatal myocardial infarction, and/or
stroke in patients
The DCOR Trial
•
Largest prospective randomized clinical outcomes trial
conducted in the hemodialysis population
•
First large scale head-to-head trial comparing clinical
outcomes in patients randomized to alternative phosphate
binder therapies
•
Real-world study design
– Broadly applicable across clinical practice settings
– Health economic outcomes data allow for
analysis of cost-effectiveness
DCOR Study Objectives
• Primary Objective
– To compare the association of sevelamer use versus
calcium-based phosphate binder use on all-cause
mortality in hemodialysis patients
• Secondary Objectives
– To compare the association of sevelamer use and
calcium-based phosphate binder use with:
• Cause-specific mortality (cardiovascular, infection, other)
• All-cause and cause-specific hospitalization and morbidity
• Medicare expenditures
Study Design
•
Multi-center, open-label, parallel design trial
•
Patients randomized to receive sevelamer or
a calcium-based phosphate binder in a 1:1
fashion
•
Randomization stratified by age, race, gender
and diabetic status
•
1,000 patients per arm
•
Patients enrolled from March 2001 through
Patient Enrollment
•
Inclusion criteria
– > 18 years
– On dialysis for > 3 months
– Required phosphate binder therapy
– Had Medicare as their primary insurance
•
Exclusion criteria
– Current dysphagia, swallowing disorders, severe
gastrointestinal motility disorders, or bowel
obstruction
– Hypersensitivity to sevelamer
Statistical Methods
• Sample Size Calculation:
– Assuming a mortality rate of 200 per 1000 patient
years in the control group, a two sided test, and a
preserved overall significant level of 0.05
– A total of 1000 patients per treatment arm would
provide 80% power to detect a 22% decrease in allcause mortality
Statistical Methods
• Survival probabilities (Kaplan-Meier curves) were
calculated for time to death by treatment group
overall
• Cox regression models were run to determine relative
risks and 95% confidence intervals adjusting for the
following pre-specified prognostic factors (age, race,
gender, diabetes, cause of ESRD, and vintage)
• Treatment interactions with each of the prognostic
factors were assessed
• A statistically significant interaction was required as a
gating step prior to subset analysis
Results
Patient Disposition
Randomized
N = 2,103
Renagel
N = 1,053
Never Dispensed
N = 20
Calcium
N = 1,050
Dispensed
N = 1,033
Never Dispensed
N = 43
Dispensed
N = 1,007
Terminated Early (N=484)
• Consent Withdrawn (N=62)
• Investigator Decision (N=90)
• Lost to Follow-up (N=107)
• Adverse Event (N=81)
• Renal Transplant (N=59)
• Changed Dialysis Modality (N=22)
• Other (N=63)
Terminated Early (N=491)
• Consent Withdrawn (N=61)
• Investigator Decision (N=141)
• Lost to Follow-up (N=98)
• Adverse Event (N=50)
• Renal Transplant (N=46)
• Changed Dialysis Modality (N=25)
• Other (N=70)
Completed Study*
N = 549
Completed Study*
N = 516
* Completers either died or followed through the end of the study.
All patients dispensed study medication were included in the data analysis.
Demographics
Variable
Race [N (%)]
Caucasian
Black
Asian
Other
Age
[Mean ± SD, (years)]
[Median (years)]
Gender [N (%)]
Male
Female
Diabetes Status [N (%)]
No
Yes
Sevelamer
(N=1033)
Calcium
(N=1007)
p-value*
0.36
508 (49)
482 (47)
6 (1)
37 (4)
479 (48)
470 (47)
11 (1)
47 (5)
0.69
60 ± 14
62
60 ± 15
62
0.96
562 (54)
471 (46)
546 (54)
461 (46)
0.93
509 (49)
524 (51)
499 (50)
508 (50)
* Fisher’s exact test for categorical variables and Wilcoxon rank sum test for continuous variables
Renal History
Variable
Sevelamer
(N=1033)
Calcium
(N=1007)
0.54
Primary Cause of CKD [N (%)]
Diabetes
Hypertension/Large Vessel Disease
Glomerulonephritis
Secondary GN/Vasculitis
Interstitial Nephritis/Pyelonephritis
Neoplasms/Tumors
Miscellaneous Conditions
Cystic/Hereditary/Congenital Disease
440 (43)
342 (33)
108 (10)
16 (2)
24 (2)
10 (1)
66 (6)
27 (3)
432 (43)
338 (34)
97 (10)
22 (2)
29 (3)
8 (1)
47 (5)
34 (3)
0.58
Dialysis Vintage
[Mean ± SD, (months)]
[Median (months)]
pvalue*
38 ± 39
24
38 ± 40
24
* Fisher’s exact test for categorical variables and Wilcoxon rank sum test for continuous variables
Laboratories (Mean + SD)
Parameter
Sevelamer
(N=843)
Calcium
(N=843)
p-value*
Phosphorus
5.8 ± 1.3
5.7 ± 1.3
< 0.01
Calcium
9.2 ± 0.7
9.5 ± 0.7
< 0.0001
53.7 ± 12.0
53.6 ± 12.9
0.60
278
226
< 0.0001
Total Cholesterol
145.6 ± 33.8
160.8 ± 34.7
< 0.0001
LDL Cholesterol
69.0 ± 25.9
84.9 ± 31.0
< 0.0001
1.6 ± 0.3
1.6 ± 0.3
0.11
Calcium x Phosphorus Product
iPTH ^
Kt/V
* Wilcoxon Rank Sum Test
^ Presented as median
Results
All-Cause Mortality
Cumulative Incidence of All-Cause Mortality
0.6
0.5
0.4
Primary endpoint results
inconclusive across entire
study population
0.3
p = 0.30
0.2
0.1
Calcium
Sevelamer
0.0
0
No. at Risk
Calcium
1007
Sevelamer 1033
1
2
3
Time on Study (Years)
640
656
430
449
161
195
4
Pre-Specified Subset Analysis
Gender
[male/female]
Race
[black/non-black]
Age
[< 65 or ≥ 65 years]
Diabetes
[yes/no]
Dialysis vintage
≤ or > 2 years
Cause ESRD
diabetes, hypertension, other
p-value
Cox Regression Model
Variable
0.67
0.84
0.03
0.35
0.83
0.82
Of the pre-specified
variables, only age
demonstrated a
statistically significant
treatment interaction
Subset analysis of
patients < or ≥ 65 years
was undertaken in
accordance with prespecified statistical plan
Cumulative Incidence of All-Cause Mortality
All-Cause Mortality
in Patients ≥ 65 years
0.8
Sevelamer therapy resulted in
a statistically significant
reduction in the relative risk
for all-cause mortality in
pre-specified subset
[RR 0.78 (0.62-0.97)]
0.7
0.6
0.5
0.4
↓ 22%
p = 0.03
0.3
0.2
Calcium
Sevelamer
0.1
0.0
0
No. at Risk
Calcium
556
Sevelamer 585
1
2
3
Time on Study (Years)
366
381
245
253
98
99
4
Cumulative Incidence of All-Cause Mortality
All-Cause Mortality Patients < 65 years
0.8
Calcium
Sevelamer
0.7
0.6
0.5
0.4
Results inconclusive for
patients <65 years
0.3
0.2
p = 0.31
0.1
0.0
0
1
2
3
Time on Study (Years)
No. at Risk
Calcium
451
Sevelamer 448
274
275
185
196
62
97
4
All-Cause Mortality for Patients Other Than
the Pre-specified Age Cut-point (< or ≥ 65
years)
1.4
Relative Risk
1.2
1.0
0.8
0.6
Statistically significant
reduction of relative risk for allcause mortality was observed
for patients >45 years of age
treated with Sevelamer
0.4
0.2
All
25
30
35
40
45
50
55
Age Cut point (years)
60
65
70
Cause specific Mortality
• Results for CV and infection mortality were
similar to all-cause mortality
• With only 53% of deaths due to CV causes, the
study is not powered to detect differences in
CV death
All-Cause Hospitalizations
Variable
Number of Hospitalizations
[per patient year]
Mean ± SD
Median
Number of Days Hospitalized
[per patient year]
Mean ± SD
Median
Sevelamer
(N=1033)
Calcium
(N=1007)
p-value*
2.1 ± 4.4
1.0
2.3 ± 4.9
1.3
0.06
15 ± 28
5.0
17 ± 32
5.8
0.09
* Wilcoxon rank sum test
Conclusions
• Results of the DCOR trial failed to reach
statistical significance for the primary endpoint of all-cause mortality (RR 0.91; p = 0.3)
• Statistically significant reduction in all-cause
mortality was observed for patients receiving
sevelamer in a pre-specified subset analysis of
patients ≥ 65 years of age (RR 0.78, p = 0.03)
• A strong trend toward reduced hospitalizations
in sevelamer-treated patients was observed (p
= 0.06)
QUESTIONS?
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
DCOR study: CMS-based data analysis
Jeremy Heaton
Outline
• CMS data used in analysis
• Study rules
• Randomized versus dosed population
• 90-day versus study completion follow-up
•
•
•
•
•
Baseline data
Mortality
Hospitalization
Morbidity
Costs
CMS data background
• Medicare tracks vital status (mortality)
• Medicare pays for:
–
–
–
–
Dialysis procedures
Hospitalizations
Physician visits, outpatient procedures
Some IV medications (EPO, iron, vitamin D, antibiotics
administered in dialysis unit)
– Other treatments, excluding most oral medications
• All payments occur following adjudication by Medicare
• Payments finalized over time
• Data presented reflect up to December 31, 2004
What are CMS data?
Structure of the CMS database
CMS = Centers for Medicare and Medicaid Services
USRDS 2004 Annual Data Report
Methods – patient matching
• 2101 patients in CMS database were matched to patients in the
Case Report Form (CRF) database ITT population
• Patients were matched based on:
–
–
–
–
–
–
–
SSN
first/last name initial
date of birth
dialysis initiation date
gender
race
primary cause ESRD
Data used in CDRG data analysis
CRF
CMS
•
•
•
Baseline age, race, gender,
diabetes, vintage, ESRD cause
Randomization date, drug
dispensed date, stop date,
study completion date
•
•
•
•
•
•
Pre-study comorbidities (CHF,
ASHD, other cardiac, PVD,
CVA/TIA; diabetes-by CRF)
CVD history (0, 1, ≥1)
Baseline Hb, URR, Vit D use
Death date and cause
Hospitalization date and cause
Morbidity (inpatient and
outpatient) and cause
Costs (total, inpatient,
outpatient, skilled nursing
facility, other)
Randomized versus Dosed populations
• Analyses were conducted for both randomized and dosed
populations (majority):
– Randomized population – followed from date of randomization
– Dosed population – followed from date of first dose of study medication
Four rules used
Event
Death (mortality analysis only)
Hospitalization
Early
Morbidity
Cost
Randomization
or first dose date
termination
+90 days
Study completion
12/31/03
12/31/04
Early termination: withdrawal pt consent, renal transplant, modality
change; investigator decision, lost-to-follow-up, adverse event, other
Importance of data captured by CMS
• Support results from prospective DCOR trial
• Provide additional details otherwise impossible without great
expense
– Hospitalization dates, causes, costs
– Morbidity (inpatient, outpatient), dates, causes, costs
– Cost breakdown
• Follow more early termination patients
– More death, hospitalization events
Risk factors adjusted for in analyses
• Clinical trial
–
–
–
–
–
–
Age
Race
Gender
Diabetes status (2 categories)
Vintage
ESRD Cause
• CMS analysis
–
–
–
–
–
Age
Race
Gender
Vintage
Pre-study comorbidities
• Diabetes (per CRF-3 categories)
• CHF
• ASHD
• CVA/TIA
• PVD
• Other cardiac
CMS 3-year results
• New data
–
–
–
–
Haemoglobin
URR
Vitamin D use
Baseline co-morbidities
• CHF
• PVD
• ASHD
• CVA/TIA
• Other cardiac
• 90-day (mainly) and SC follow-up data
Baseline characteristics
Baseline characteristics: Dosed population
Hb, URR, vitamin D use
Sevelamer
(n=1031)
Calcium
(n=1007)
p value*
Variable
Total
(n=2038)
Haemoglobin, g/dL
Mean ± SD
Median
11.9 ± 1.3
11.9
11.9 ± 1.4
11.9
11.8 ± 1.3
11.8
0.9289
URR, n (%)
< 60 %
60- < 64 %
65- < 70 %
70- < 75 %
≥ 75%
Unknown %
99 (5)
123 (6)
371 (18)
612 (30)
706 (35)
127 (6)
54 (5)
61 (6)
198 (19)
308 (30)
343 (33)
67 (7)
45 (5)
62 (6)
173 (17)
304 (30)
363 (36)
60 (6)
0.6696
Vitamin D use, n (%)
No
Yes
687 (34)
1351 (66)
339 (33)
692 (67)
348 (35)
659 (65)
0.4261
* Fisher’s exact test for categorical variables and Wilcoxon rank sum test for continuous variables
Baseline characteristics: Dosed population
Cardiovascular comorbidities
Total
(n=2038)
Variable
Sevelamer
(n=1031)
Calcium
(n=1007)
Cardiovascular History #
[N (%)]
p value*
0.9269
0 co-morbidities
686 (33.7)
351 (34.0)
335 (33.3)
1 co-morbidity
491 (24.1)
248 (24.1)
243 (24.1)
> 1 co-morbidity
861 (42.2)
432 (41.9)
429 (42.6)
* Fisher’s exact test for categorical variables and Wilcoxon rank sum test for continuous variables
Mortality
All-cause mortality
Kaplan-Meier survival curves of time to all-cause mortality, by treatment
(90-day follow-up rule, ITT population)
p = 0.5332
Hospitalization
Multiple hospitalizations (per 100 patient-years)
Sevelamer
Calcium
p value
All-cause
170
191
0.02
CV
55
60
0.05
Infection
25
30
0.11
Vascular Access
33
34
0.94
Fracture
3
3
0.21
Other
55
64
0.10
Hospital days (mean per patient-year)
Dosed population, SC rule
Sevelamer
Calcium
Mean hospital days per patient year
p=0.052
13.9
14
12.1
12
10
8
6
p=0.155
p=0.06
4.6
3.7
4
4
p=0.1
p=0.85
2.7
2
3.9
2
2.2
2
p=0.21
0.38 0.48
0
All-cause
CV
Infection
VA
Fracture
Other
p value: Wilcoxon rank sum test was used
Follow-up time calculated as total number of patient-years at risk of hospital admission including time in hospital
Unadjusted analyses
Multiple Hospitalizations
(Anderson-Gill Regression Model)
Relative Risk for Multiple Hospitalization
Sevelamer (Referent: Calcium)
All-cause
Cardiovascular
Infection
Vascular access
Fracture
Other
0.4
0.6
0.8
1.0
Relative Risk
1.2
1.4
1.6
p < 0.05
Adjusted for demographics and pre-study comorbidities, dosed population
Genzyme Corporation Protocol GTC-68-401
Centers for Medicare and Medicaid Services Data Analysis
10.2.1.1
Hospitalization days per patient
(Poisson Main-Effects Regression Model)
Relative Risk for Days Hospitalized Per 100 Patient Years
Sevelamer (Referent: Calcium)
All-cause
Cardiovascular
Infection
Vascular access
Fracture
Other
0.4
0.8
0.6
1.0
Relative Risk
1.2
1.4
1.6
p < 0.05
Adjusted for demographics and pre-study comorbidities, dosed population
Genzyme Corporation Protocol GTC-68-401
Centers for Medicare and Medicaid Services Data Analysis
10.2.1.1
Morbidity
Morbidity analyses
• Categories
–
–
–
–
Cardiovascular (CV)
Vascular access (VA)
Infection
Fracture
• Included both inpatient and outpatient events
• Rules were created to avoid counting same event twice
Morbidity results (adjusted analyses)
• First morbidity (Cox regression analysis)
– Final Model: All RR (when no interactions) favour sevelamer, but all
are NS
• CV: interaction with CVA/TIA (SC only)
• VA: interaction with diabetes (90 d, SC)
• Infect: interaction with dialysis duration (90 d only)
• Multiple morbidity (Anderson-Gill model)
– Final model: All RR favour sevelamer, but all are NS
• VA: interaction with diabetes (90 d, SC),
• Fracture: interaction with ASHD (SC only); too few events to
evaluate
Cost analysis
Mean Annual Medicare costs
Dosed population, 90-day follow-up
$70,000
$66,348
$68,688
$60,000
$44,760
$50,000
$44,532
$40,000
$30,000
Sevelamer
Calcium
$18,528 $20,748
$20,000
$10,000
$960 $1,320
$2,088 $2,100
$0
Total
Inpatient
Outpatient
Difference:
Skilled nursing
facility
Other
($2,340)
($2,220)
$228
($360)
($12)
(3.5%)
(12.0%)
0.5%
(37.5%)
(0.6%)
Inpatient actual annual cause-specific costs
Dosed population, 90-day follow-up
$25,000
$20,000
$6,432
$15,000
$5,316
$2,592
$3,372
$10,000
$4,176
$4,128
$5,916
$6,240
Sevelamer
Calcium
$5,000
$0
Other
Infection
Fracture
VA
CVD
Annual Medicare allowable expenditures: selected sub-groups
Dosed population, 90-day follow-up
Mean patient cost per year
$80,000
p = 0.028
$75,000
p = 0.030
$70,000
p = 0.034
p = 0.063
Calcium
Sevelamer
$65,000
$60,000
$55,000
$50,000
N
934 961
428 415
468 484
454 473
Total
>=65
Diabetic
Dialysis
<2 yrs
Total cost per patient shown
Summary – mortality
• All-cause and cause-specific mortality NS at 3 yrs
• Trend for sevelamer benefit is seen for all-cause and other
cause mortality
• CRF and CMS results are consistent using 90-day rule, even
though more deaths captured by CMS data
• Significant age interaction seen with all-cause mortality
Summary – hospitalization
Multiple hospitalization (Anderson-Gill)
•
Dosed population, 90-day rule
–
•
Dosed population, SC rule
–
•
Positive sevelamer effect on all-cause (RR=0.9, p=0.05) and ‘other’ (RR=0.85, p=0.02),
but not CV hospitalization
Positive sevelamer effect on all-cause (RR=0.89, p=0.01), infection (RR=0.03) and
“other” (RR=0.86, p=0.02), but not CV hospitalization
Diabetes: significant interaction in multiple analyses
Hospital days (Poisson regression)
•
•
•
All-cause: Positive sevelamer effect seen with both 90-d (RR=0.84, p=0.01) and
SC (RR=0.86, p=0.02)
Infection: Positive trend with 90-d (RR=0.78, p=0.06), positive effect with SC
(RR=0.76, p=0.02)
‘Other’: Positive effect with 90-d (RR=0.81, p=0.04) only, trend with SC
(RR=0.86, p=0.11)
Summary – morbidity
• First morbidity (adjusted, Cox regression):
– All RR (when no interactions) <1 favouring sevelamer, but all are NS
– Scattered interactions; not consistent
• Multiple morbidity (adjusted, Anderson-Gill model):
– All HR < 1 favoring sevelamer, but all are NS
– Scattered interactions; not consistent
Conclusions
•
There was no statistically significant difference for all-cause mortality
overall.
– All-cause mortality was significantly reduced by 19% in patients over 65 years.
•
All-cause hospitalisation was significantly reduced by 10 %.
•
Overall, sevelamer resulted in less Medicare expenditure, reducing inpatient
expenditure by 12.0% annually.
•
Expenditures were consistently less for sevelamer in sub-groups, including:
– older patients
– diabetic patients
– patients on dialysis for at least 2 years.
DCOR study: CMS-based data analysis
Principal Investigator
Wendy L. St. Peter, Pharm.D., FCCP, BCPS
Principal Biostatistician
Jiannong Liu, Ph.D.
Biostatisticians
Qiao Fan, M.S. and Eric Weinhandl, M.S.
Company
Chronic Disease Research Group
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Discussion and Conclusions
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Debrief Day One
Workshop One: Weighing the Evidence
Group A: Orangery
Group B: Mowbray
Great Fosters, UK
28th -29th June 2006
Share Discussion from Workshop One
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
What Price Health?
The case for sevelamer
Heba Elgazzar MSc
Global Health Outcomes Genzyme
Overview

Context

Health Economic Analysis

Conclusions and Questions
Context
Health Care in CKD
Health Care in CKD
WHO Goal by 2015:
Reduce mortality in chronic diseases
by 2% annually
Number of dialysis patients:
1.45 million
Source: WHO ‘Preventing chronic diseases: a vital investment’, 2006; Fresenius Annual Report, 2005.
Health Care in CKD
Dialysis is the only treatment in the
world for which all governments pay.
Source: Jonssen, Pharmacoeconomics 2005.
Annual Resources Per Capita
$70,000
$60,000
$50,000
$40,000
General
Hemodialysis
$30,000
$20,000
$10,000
$0
Brazil
USA
Germany
UK
WHO, 2003; Romão Jr, 2005; USRDS, 2003; IGES 2006; Grun et al, 2004.
Resources per patient
Hospitalization
15%
Dialysis
70%
Complex
Medications
11%
Basic
Medications
5%
Ploth et al, 2003; Romão Jr 2005; Grun et al, 2003.
Key Questions asked by Health
Authorities
1.
Perceived Medical and Clinical Need:

“Is it needed?”
2.
Effectiveness:
 “Does it work?”
3.
Budgetary impact and cost effectiveness:
 “Can we afford it and is it worth it?”
Health Economic Analysis - Aims
1.
Demonstrate value-for-money
2.
Support reimbursement and patient access to
health care coverage
Comparative evidence in ESRD
Risk Factor
Relative
Risk on
Mortality*
Intervention
Kt/V
Vascular access
1.16
1.23
Dialysis mode
Hg levels
1.14
Erythropoeitin
Cholesterol
--
Statins
Parathyroid
hormone
1.01
Vitamin D
Cinacalcet
Phosphate
1.11 / 1/04
Calcium
Lanthanum
Sevelamer
*Port et al, 2004 DOPPS
Comparative evidence in ESRD
Risk Factor
Relative
Risk on
Mortality*
Intervention
Mortality
Morbidity
Quality
of Life
Kt/V
Vascular access
1.16
1.23
Dialysis mode
++
++
+
Hg levels
1.14
Erythropoeitin
--
--
+
Cholesterol
--
Statins
--
--
--
Parathyroid
hormone
1.01
Vitamin D
Cinacalcet
---
+
--
---
Phosphate
1.11 / 1/04
Calcium
--
--
--
Lanthanum
--
--
--
Sevelamer
+
+
--
*Port et al, 2004 DOPPS
Health economics of Sevelamer —
Treat-to-Goal Predictive model
CVD events predicted per 100 patients in long-term:
Sevelamer
Events per 100 patients
80
70
Calcium acetate
12%
RR
60
50
40
12%
RR
30
20
10
0
All CVD events
Huybrechts et al. Value in Health 2005.
Fatal CVD events
Treat-To-Goal Predictive Model
Less CVD
morbidity
Sevelamer predicted
cost-effective
Huybrechts et al, 2005, Kidney International; Value-in-Health
Predicted cost-effectiveness
Sevelamer
Source: National Institute for Health and Clinical Excellence, 2002.
DCOR: Economic Implications


Cause-specific hospitalizations
Expenditure
DCOR All-cause Hospitalizations
Variable
Number of Hospitalizations
[per patient year]
Mean ± SD
Median
Number of Days Hospitalized
[per patient year]
Mean ± SD
Median
Suki et al, ASN 2005
Sevelamer
(N=1033)
Calcium
(N=1007)
p-value*
2.1 ± 4.4
1.0
2.3 ± 4.9
1.3
0.06
15 ± 28
5.0
17 ± 32
5.8
0.09
* Wilcoxon rank sum test
Outcomes Benefit and Value
Conclusions
50% less mortality over 5 years in incident patients
34% less mortality in patients treated for at least 2 years
22% less mortality in patients over 65 years
10% less hospitalisation in prevalent patients over 3 years
Net cost: - 4.0% to + 1.0% relative to calcium
Economic Evaluation Questions






Which treatment effects?
Follow-up time?
Binder costs – compliance?
Sub-groups?
Quality of life gains (due to less hospitalisation)?
Sensitivity analysis parameters?
Cost-Effectiveness Analysis (CEA)
Jörgen Möller and Paul Trueman
Great Fosters, UK
28th -29th June 2006
Sevelamer (Renagel) in the
Management of CKD: A NICE
Perspective
Paul Trueman
29th June 2006
Items for Discussion
 Overview
 Sevelamer and NICE
 DCOR & RIND for the UK setting
 Key issues
 Summary
Overview
 Aims
 To demonstrate the cost-effectiveness of
sevelamer (Renagel) in the management of
chronic kidney disease in a UK setting
 To incorporate data emerging from the DCOR
and/or RIND trials
 To ensure that the revised economic model is
tailored for use in submission to NICE
Overview
 Key Issues for today
 Comparison of Sevelamer with similar technology
appraisals from NICE;
 Application of emerging evidence (DCOR/RIND);
 Utility measurement;
 Model parameters (population and horizon);
 Costs.
Case Study: Cinacalcet
 NICE Review
 “Cinacalcet hydrochloride is not recommended for the
routine treatment of secondary hyperparathyroidism in
patients with ESRD on maintenance dialysis therapy”
 Implications for sevelamer
 Similar patient group, so can we expect a similar decision?
Case Study: Cinacalcet
 Incremental cost-effectiveness ratio
 Manufacturer’s submission = £35,600
 Assessment Group’s model:
 Base case = £61,900
 Inclusion of dialysis costs = £72,000
 Alternative assumption = £43,000
 Best case = £38,900
 Treatment discontinuation = £57,400
 “The committee concluded that cinacalcet
hydrochloride was unlikely to be a costeffective use of NHS resources”
Case Study: Cinacalcet
 Key differences between sevalamer and cinacalcet
 Incremental cost-effectiveness ratio
 Comparator (‘cinacalcet + standard care’
versus ‘standard care alone’)
 “…it did not replace the need for dietary restrictions
and the use of other medications such as phosphate
binders and vitamin D sterols”
 Sevelamer positioned as an alternative calcium
containing phosphate binders
 Cinacalcet positioned as an adjuvant to other
therapies – less potential for cost offsets
Case Study: Cinacalcet
 Implications for sevelamer
 “…there was a lack of data relating to long-term treatment”
 “…there was considerable uncertainty in the extent to which
intervening to correct derangements in the levels of PTH,
calcium and phosphate (in particular by lowering PTH levels)
was effective in reducing the risk of adverse events”
 “…people on dialysis have many other factors that contribute
to their risk of serious adverse events, and these add to the
uncertainty in predicting clinical benefits from changes in
surrogate markers”
Case Study: Cinacalcet
 Implications for sevelamer
 Reviewing HTA report to determine how
utilities, costs and mortality were
addressed.
Patient Demographics
 Applying the DCOR and RIND results to the UK
population
Age
(years)
Male
UK (Renal
Registry)
DCOR
RIND
64.7
60.5
58.0
62.0%
54.4%
63.0%
Patient Demographics
UK (Renal
Registry)
DCOR
RIND
Caucasian
84%
48%
42%
Black
4%
47%
31%
Asian
10%
1%
-
Other
3%
4%
27%
Ethnicity
Patient Demographics
Primary cause of
ESRD
UK (Renal
Registry)
DCOR
RIND
Glomerulonephritis
10.4%
10.1%
7.0%
Pyelonephritis
7.0%
2.6%
-
Diabetes
18.0%
42.7%
53.0%
Reno-vascular
disease
7.5%
-
-
Hypertension
5.5%
33.3%
20.0%
Polycystic kidney
disease
5.4%
-
-
Patient Demographics
Median vintage
(prevalent
patients)
UK (Renal
Registry)
DCOR
RIND
Vintage (years)
2.70
3.16
-
Discussion Point – Outcomes
 Key outcomes from DCOR & RIND
Outcome
Suitable for model?
All-cause mortality
Yes, but modelled to specify cause
CV mortality
Yes, see above
Hospitalisations
Only for costing purposes
Time to 1st hospitalisation
No – only use overall rate
Medicare costs
No, not representative of ‘real world’
Non-fatal CV events
Yes, for quality of life and costing
Non-fatal other events
Yes, for quality of life and costing
Change in calcification
Not as main outcome (use as marker)
Discussion Point – Outcomes
 Currently no utility data for patients with ESRD and
adverse events (i.e. myocardial infarction)
 Options:
 Elicit utility scores
 Ideal, but insufficient time
 Obtain utility scores from literature
 Search already undertaken
 Additional search from York, but no positive
findings
 Based on assumptions
 i.e. multiplicative (or additive) effect of AEs +
ESRD, with extensive discussion
 No use of utility (i.e. use life years only)
 Discussion of likely impact (i.e. overestimation
of benefits)
Discussion Point - Scenarios
 Subgroup analysis




Incident / prevalent patients
Vintage
Age
Pre-end stage renal disease
 Time horizon
 Lifetime (captures long-term benefits)
 Five-year analysis (to address issues relating to
lack of long-term data)
Discussion Point - Costing
 Should we include the cost of dialysis?
 This will apply to both treatment arms (i.e. calcium binders
and sevelamer)
 However, increased life expectancy will be associated with
increased costs
 Are ‘background costs’ important when considering the costeffectiveness of sevelamer?
 NICE tend to account for all costs
 We should accommodate an alternative analysis
which includes dialysis costs in order to anticipate the
Assessment Group’s approach.
Summary
 Sevelamer differs from cinacalcet submission
 Improved ICER
 ‘Treatment versus treatment’ analysis, rather than ‘treatment
versus do nothing’





Are DCOR and/or RIND applicable to the UK setting?
Approach towards cost-utility analysis
Time horizon
Subgroup analysis
Inclusion of dialysis costs
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Workshop Two: CEA Critique
Group A: Orangery
Group B: Mowbray
Great Fosters, UK
28th -29th June 2006
Share Discussion from Workshop Two
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
NICE Discussion
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Assessment and Recommendations:
Presenting the Evidence to Health
Authorities Country by Country
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006
Meeting Wrap-up
David Goldsmith
Great Fosters, UK
28th -29th June 2006
Renagel Health Economics
Advisory Board
Great Fosters, UK
28th -29th June 2006