hdl plays no role in the pathogenesis of atherosclerosis...hdl plays no role in the pathogenesis of...
TRANSCRIPT
HDL plays no role in the pathogenesis of
atherosclerosis
Børge G Nordestgaard
Professor, Chief Physician, MD, DMSc
University of Copenhagen & Copenhagen University Hospital
Conflict of Interest Disclosure
Consultancies or talks sponsored by AstraZeneca, Merck, Omthera, Sanofi-Aventis,
Regeneron, ISIS Pharmaceuticals, Aegerion, Dezima, Fresenius, B Braun, Kaneka, Pfizer,
Amgen, Lilly, Kowa, Denka Seiken
Rebellious
Hans Christian Andersen’s fairy tale: The Emperor's New Clothes
HDL
Borge
HDL supportersHDL supporters
”HDL mafia”
”HDL mafia”
Two weavers promise
an Emperor a new suit of
clothes that is invisible to
those who are unfit for their
positions, stupid, or
incompetent.
The good cholesterol
Low HDL
protects against
heart disease
Medical textbooksLay press
Brain washed
Main focus on LDL cholesterol
• LDL receptor – Goldstein & Brown
• Oxidized LDL – Steinberg
• Statins – Endo, 4S trial and others
• European guidelines – LDL only
• American guidelines – LDL mainly
Next focus on HDL cholesterol• Strong epidemiology – all studies
• Animal studies
• The ”HDL mafia”
• Big Pharma
• But then……
HDL-C
CVD
2007
Drug targets based on genetic evidence
LDL-C Lp(a)
HDL-C ? Remnant-C / TG
Nordestgaard Eur Society Cardiology Congress 2011
Clinical focus on lipoproteins for CVD prevention
1980 1990 20102000
100%
0%
50%
Nordestgaard 2014
2020
LDL
HDLRemnants
TGs
4S trial
Epidemiology
Clinicians
Trials
”Postprandial”
Epidemiology
Genetics & ”failed” HDL trials
Nobel prize 85 & statins
Evidence for risk factors causing
atherosclerotic disease?
LDL Many other
risk factors
Epidemiology √ Many studies √
Genetics √ FH + SNPs No
Animal models √ Many models Sometimes
Mechanism √ Understood Speculations
Intervention √ Statin trials No
Nordestgaard 2015
Randomized trial vs. Mendelian randomization
Randomization methods
Placebo Drug: (lipo)protein
levels or
Cardiovascular disease or
Confounders
evenly distributed
Random distribution of alleles
Confounders
evenly distributed
Cardiovascular disease or
Normal
allele
Allele: (lipo)protein
levels or
Reverse causation
HDL
Remnants
LDL
HDL cholesterol
Remnant or
VLDL cholesterol
LDL cholesterol
Triglycerides
total cholesterol minus LDL-C minus HDL-C
no direct assay available yet
< 1 1 – 2 2 – 3 3 – 4 4 – 5 5
2309 6040 3023 1294 527 477
Women
Triglycerides, mmol/L < 1 1 – 1.99 2 – 2.99 3 – 3.99 4 – 4.99 5
Triglycerides, mg/dL < 89 89 – 176 177 – 265 266 – 353 354 – 442 443
Observations, No. 4564 7355 2298 652 212 141
Cholesterol, mmol/L (mg/dL)
Remnant 0.35 (14) 0.64 (25) 1.08 (42) 1.53 (59) 1.62 (63) 2.33 (90)
LDL 2.95 (114) 3.41 (132) 3.64 (141) 3.70 (143) 3.92 (151) 3.54 (137)
HDL 2.11 (81) 1.86 (72) 1.57 (61) 1.41 (55) 1.32 (51) 1.13 (44)
Lipoprotein cholesterol, mmol/ L
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
Lip
opro
tein
chole
stero
l, m
mol/
L
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
Lip
opro
tein
chole
stero
l, mg/d
L
0
50
100
150
200
250
*** ***
*** *** *** ***
*** ***
*** ***
*** ***
***
***
***
*** *** ***
***
***
*** ***
*** ***
***
*** *** *** ***
***
Remnant
LDL
HDL
Remnantp < 0.001
LDLp < 0.001
HDLp < 0.001
Trend
N =
Triglycerides(mmol/L)
Men
Copenhagen General Population Study
Freiberg, Nordestgaard 2011
***
***
Remnants
Evidence for lipoproteins causing
atherosclerotic disease?
LDL TG &
Remnants
Epidemiology √ √ Many studies
Genetics √ √ Remnant
hyperlipidemia
Animal models √ √ Many models
Mechanism √ √ Similar to LDL
Intervention √ (√) More needed
Nordestgaard 2015
Copenhagen General Population Study (CGPS)
Copenhagen City Heart Study (CCHS)
N=15,000
N=110,000+
37 yrs follow-up
10 yrs follow-up
No losses to
follow-up
1977-2014
Copenhagen
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Nonfasting triglycerides, mmol/L27% 46%
27% 0.1%
Frac
tio
n o
f p
op
ula
tio
n
Reference In extreme groupsNordestgaard & Varbo, Lancet 2014; 384: 626-635
Nonfasting triglycerides, mmol/L
Copenhagen City Heart Study and Copenhagen General Population Study
Haz
ard
rat
io (
95
%C
I)
Nordestgaard & Varbo, Lancet 2014; 384: 626-635
Myocardial infarctionN = 96,394 (Events = 3,287)
0
1
2
3
4
5
6
1 2 3 4 5 6 7
Nonfasting triglycerides, mmol/L
Copenhagen City Heart Study and Copenhagen General Population Study
Haz
ard
rat
io (
95
%C
I)
Ischemic Heart DiseaseN = 93,410 (Events = 7,183)
Haz
ard
rat
io (
95
%C
I)
Mainly fasting triglycerides, mmol/L
Emerging Risk Factors Collaboration
JAMA 2009
Coronary Heart DiseaseN = 302,430 (Events = 12,785)
0
1
2
3
4
5
0
1
2
3
4
5
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Nordestgaard & Varbo, Lancet 2014; 384: 626-635
In decilesIn extreme groups
Nonfasting triglycerides, mmol/L
Copenhagen City Heart Study and Copenhagen General Population Study
Haz
ard
rat
io (
95
%C
I)
Ischemic StrokeN = 97,442 (Events = 2,994)
Haz
ard
rat
io (
95
%C
I)
Mainly fasting triglycerides, mmol/L
Ischemic StrokeN = 173,312 (Events = 2,534)
0
1
2
3
4
5
0
1
2
3
4
5
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Nordestgaard & Varbo, Lancet 2014; 384: 626-635
Emerging Risk Factors Collaboration
JAMA 2009
In quintilesIn extreme groups
Nonfasting triglycerides, mmol/L
Copenhagen City Heart Study and Copenhagen General Population Study
Haz
ard
rat
io (
95
%C
I)
All-cause mortalityN = 98,515 (Events = 14,547)
0
1
2
3
4
5
1 2 3 4 5 6 7
Nordestgaard & Varbo, Lancet 2014; 384: 626-635
In extreme groups
Lipoprotein Cardiovascular
Disease Risk
Genotype
Mendelian randomization hypotheses
established but causal?
effect size?
pleiotropic effects?statistical power?
1
2 3
Causality: Instrumental Variable Analysis
Remnant cholesterol
mmol/L
HDL cholesterol
mmol/L
Hazard ratio for IHD
<0.4
0.4-0.6
0.6-0.7
0.7-1.1
>1.1
>2.0
1.7-2.0
1.4-1.7
1.2-1.4
<1.2
Quintiles
1.0 1.5 2.0 2.5N=57.000
CCHS+CGPS
Varbo et al JACC
2013; 61: 427-36
N=66.000 CCHS+CGPS+CIHDS
Remnant cholesterol
Plasma: observational
Genetic: causal
Remnant / HDL-C
Plasma
Genetic
HDL cholesterol
Plasma
Genetic
Hazard ratio for IHD per 1mM or
1.0 2.0 4.0
15 selected genetic variants
LDL cholesterol
Plasma
Genetic
12.000 IHD
Varbo et al JACC
2013; 61: 427-36
N=87.000 CARDIoGRAM
Triglycerides
Genetic unadjusted
Genetic LDL+HDL adjust
HDL cholesterol
Genetic unadjusted
Genetic LDL+TG adjust
Effect size (β) for CAD per 1SD or
0 0.3 0.6
Genome wide 185 SNPs
LDL cholesterol
Genetic unadjusted
Genetic TG+HDL adjust22.000 CAD
Do et al Nat Genet
2013; 45: 1345-52
APOC3Jørgensen et alNEJM 2014
01
01
APOC3TG and HDL Working GroupNEJM 2014
PCSK9CGPS & CCHS
0
1
Ischemic vascular disease
N alleles N total N events Risk estimate
75,465 10,770
260 27 0.59
110,472 33,889
498 113 0.60
64,492 10,665
1,697 250 0.90
- 44%
- 39%
- 12%
LDL cholesterol, mmol/L
Triglycerides, mmol/L Hazard ratio (95% CI)
Hazard ratio (95% CI)
0.0 0.5 1.0
0.0 0.5 1.0
0 1 2
0 1 2 3 4
Nordestgaard & Varbo, Lancet 2014; 384: 626-635
LDL
Remnants
LDL
Remnants
Plasma Intima
Triglycerides
Cholesterol Chylomicron
Inflammation
Macrophage
Foam cells
LPL LPL
FFA +Monoacylglycerol
Nordestgaard & Varbo, Lancet 2014; 384: 626-635
HDL cholesterol
Evidence for lipoproteins causing
atherosclerotic disease?
LDL TG &
Rem-
nants
Low HDL
Epidemiology √ √ √
Genetics √ √ None
Animal models √ √ (√)
Mechanism √ √ ?
Intervention √ (√) None
Nordestgaard 2015
Evidence for lipoproteins causing
atherosclerotic disease?
LDL TG &
Rem-
nants
Low HDL
Epidemiology √ √ √
Genetics √ √ None
Animal models √ √ (√)
Mechanism √ √ ?
Intervention √ (√) None
Nordestgaard 2013
Reverse cholesterol transport?
HDL+cholesterol
Human
evidence?
Evolution?
Causal factor
with variation
Glucose
Longterm
monitoring
HgbA1c
TG
RemnantsHDL
Nordestgaard et al. Current Drug Targets, 2009, 10, 328-335
Evidence for lipoproteins causing atherosclerotic disease?
LDL TG &
Rem-
nants
Low
HDL
HDL
function
ality
Epidemiology √ √ √ None
Genetics √ √ None None
Animal models √ √ (√) None
Mechanism √ √ (√) ?
Intervention √ (√) None None
Present Future The past?
HDL hypothesis
HDL functionality
Remnant cholesterol
HDL-C for risk
prediction
SCORE-HDL
Proportion classified as high CVD mortality risk
Martin Mortensen, Afzal, Nordestgaard, Falk. Eur Heart J 2015; in press
2003-2008: 46092 individuals without CVD, diabetes or statin use
6.8 years of follow-up
Copenhagen General Population Study
Martin Mortensen, Afzal, Nordestgaard, Falk. Eur Heart J 2015; in press
-20
-10
0
10
20
%R
ecla
ssif
icati
on
Inappro
pri
ate
A
ppro
pri
ate
-20
-10
0
10
20
%R
ecla
ssif
icati
on
Inappro
pri
ate
A
ppro
pri
ate
Fatal CVD Myocardial infarctionA C
Cases Non-cases Cases Non-cases
NRI, %
p value
NRI, %
p value
-16 +4 -10 +4
0.002 < 0.0001 < 0.0001 < 0.0001
Combined Combined
-12
0.02 0.006
-5
-20
-10
0
10
20
%R
ecla
ssif
icati
on
Inappro
pri
ate
A
ppro
pri
ate
B
Cases Non-cases
NRI, %
p value
-9 +4
<0.0001 < 0.0001
Combined
-5
<0.0001
Fatal CVD + nonfatal MI or stroke
Reclassification across 5% 10-year risk of fatal CVD
Using SCORE-HDL instead of SCORE
Copenhagen General Population Study
HDL
LDL
Rem-
nant
”GOOD”
BAD
UGLY
=innocent
Hans Christian Andersen’s fairytale: The Emperor's New Clothes
HDL
Borge
HDL supportersHDL supporters
”HDL mafia”
”HDL mafia”
Rebellious