www.lipid.orgwww.lipid.org
EPIDEMIOLOGY OF HDL AND ITS EVOLUTION INTO A THERAPEUTIC TARGET
Peter P. Toth, MD, PhD, FAAFP, FNLA, FAHA, FCCP, FACCDirector of Preventive Cardiology
Sterling Rock Falls ClinicSterling, Illinois
Professor of Clinical Family and Community MedicineUniversity of Illinois School of Medicine
Peoria, Illinois
www.lipid.orgwww.lipid.org
WHAT DIFFERENTIATES HDL PARTICLES FROM OTHER
LIPOPROTEINS?
IT APPEARS TO BE PRIMARILY ANTIATHEROGENIC.
www.lipid.orgwww.lipid.org
Potential Antiatherogenic Actions of HDL
Chapman MJ, et al. Curr Med Res Opin. 2004;20:1253-1268.Assmann G, et al. Annu Rev Med. 2003;53:321-341.
Antioxidative Activity
AntithromboticActivity
Anti-infectious Activity
EndothelialRepair
AntiapoptoticActivity
ReverseCholesterolTransportCellular
CholesterolEfflux
Anti-inflammatoryActivity
VasodilatoryActivity
HDL
Apo A-I
Apo A-II
www.lipid.orgwww.lipid.org
WHERE ARE HDL PARTICLES PRODUCED?
www.lipid.orgwww.lipid.org
HDL BIOGENESIS
www.lipid.orgwww.lipid.orgZhang, Y. et al. Circulation 2010;121:1347-1355.
Because adipose inflammation is a hallmark of central obesityand type 2 diabetes mellitus, loss of adipocyte lipidation of HDL may
contribute directly to lower HDL-C levels in these inflammatory,insulin-resistant states
www.lipid.orgwww.lipid.org
Early case-control studies (1950s) in the US and Israel showed that patients with CAD tend to have lower HDL levels than controls
The Tromso Heart Study showed that CAD patients have HDL levels 35% lower than controls and that patients with low HDL are 3 times more likely to develop CAD than those with elevated LDL
Low HDL is associated with significant disability and the loss of ability to execute daily living activities among patients over 65 years of age. Among the elderly, low HDL predicts risk for future MI and stroke better than LDL.
CAD, coronary artery disease; LDL, low-density lipoprotein; MI, myocardial infarction.
HDL as CAD Risk Factor
www.lipid.orgwww.lipid.org
Increased risk for MI and CAD mortality1
Stroke in the elderly2
Sudden death3
Restenosis after angioplasty4
Severe premature atherosclerotic disease in the proximal left main coronary artery5
Increased risk for retinopathy in diabetic patients6
For every 1 mg/dL drop in HDL-C, risk for NIDDM increases 4%7
Low HDL Portends
1. Goldbourt U et al. Arterioscler Thromb Vasc Biol 1997;17:107-13.2. Qeverling-Rijnsburger AWE et al. Arch Intern Med 2003;163:1549-54.3. Burke AP et al. JAMA 1999;281:921-6.4. Shah P et al. Circulation 1992;85:1279-85.5. Pearson T et al. Am J Epidemiol 1979;109:285-95.6. Sasongko et al. Diabetes care 2011; 34: 474-479.7. Stern MP et al. Ann Intern Med 2002;136:575-581.
www.lipid.orgwww.lipid.org
AdjustedNonadjustedControls (N = 601)Risk Factor
–12%*1%Diabetes mellitus
34%*26%26%LDL-C 160 mg/dL
–41%*21%Hypertension
57%*63%*19%HDL-C 35 mg/dL
–67%*29%Cigarette smoking
*Significantly different from controls (P < 0.001).CHD, coronary heart disease; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol.
Frequency of Low HDL-C Compared with Other Risk Factors in Men with Premature CHD
Genest JJ et al. Am J Cardiol. 1991;67:1185-1189.
www.lipid.orgwww.lipid.org
EPIDEMIOLOGY OF HDL‐C
www.lipid.orgwww.lipid.org
5-9 10-14 15-19 20-24 25-29 30-34 35-39
HDL-C (mg/dL)
0
2
4
6
8
10
12
14
16
% o
f the
Pop
ulat
ion Men
Women
Miller M & Zahn M. Curr Opin Cardiol 2004;19:380-384.
<20 mg/dL occurs in 1/200 men & 1/400 women
<10 mg/dL occurs in ~ 1/20,000
< 40/50 mg/dL men/women are the bottom tertile of population
National Health And Nutrition Examination Survey III: Distribution of Low HDL-C Bottom Tertile of
Population
www.lipid.orgwww.lipid.orgBruckert E Eur Heart J Suppl 2006;8:F17-F22.
Mean HDL-cholesterol in mmol/L (M) by country and prevalence (P) of low HDL-cholesterol adjusted for eight confounding factors by country;
low HDL-cholesterol was defined as <1.03 mmol/L (40 mg/dL) in men and <1.29 mmol/L (50 mg/dL) in women.
www.lipid.orgwww.lipid.org
Prevalence (%) of different severities of obesity in the Pan-European Survey of low HDL-cholesterol (excludes 3% of the overall population
for whom obesity status was not recorded)
Bruckert E Eur Heart J Suppl 2006;8:F17-F22.
www.lipid.orgwww.lipid.org
Prevalence of low HDL-cholesterol, hypertriglyceridaemia, or both according to receipt or not of lipid-modifying treatment in the Pan-
European Survey of low HDL-cholesterol.
Bruckert E Eur Heart J Suppl 2006;8:F17-F22.
www.lipid.org
0.00
5.00
10.00
15.00
20.00
25.00
30.00
HDL 1 HDL 2 HDL 3 HDL 4 HDL 5
≤ 2%3‐4%5‐9%≥ 10%
CVD
Morta
lity R
ate (p
er 10
00 pe
rson y
ears)
0.73 0.50 0.360.39 0.30
6.52
3.891.60 3.08
2.26
9.87
5.31 5.053.69
3.22
24.69
28.84
23.32
17.0915.74
SCOR
E Ca
tegor
y
www.lipid.org
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
< 1.291.3 ‐ 1.6
> 1.6
Age < 55Age 55‐65Age > 65
HDL Tertiles
CVD
Morta
lity R
ate (p
er 10
00 pe
rson y
ears)
0.370.22
0.20
3.3
2.01.7
13
11
8
www.lipid.org
0.002.004.006.008.00
10.0012.00
14.00
16.00
18.00
< 1.091.09 ‐ 1.34
> 1.34
Age < 45Age 45‐55Age 55‐65Age > 65
HDL Tertiles
CVD
Morta
lity R
ate (p
er 10
0 per
son y
ears)
0.800.55
0.36
4.0
2.82.7
9.6
6.4
4.9
17
15
13
www.lipid.org
0.000.501.001.502.002.503.003.504.004.505.00
HDL 1 HDL 2 HDL 3 HDL 4 HDL 5
TC 1TC 2TC 3TC 4TC 5
HDL Quintile
CVD
Morta
lity R
ate (p
er 10
00 pe
rson y
ears)
0.9
0.2 0.2 0.50.2
1.21.0
0.80.3 0.7
2.2
1.41.1
0.7 0.4
2.7
1.81.4
1.2 1.2
4.94.7
2.1 2.11.7
www.lipid.org
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
HDL 1 HDL 2 HDL 3 HDL 4 HDL 5
TC 1TC 2TC 3TC 4TC 5
HDL Quintile
CVD
Morta
lity R
ate (p
er 10
00 pe
rson y
ears)
2.8
1.9
1.21.8
1.2
3.7
2.62.4
2.11.9
5.7
2.93.5
2.5 2.7
5.7
4.63.9
3.52.9
7.3
5.4 5.55.0
4.4
www.lipid.orgwww.lipid.org
Relative risk of CHD for sex-specific lipid factor quintiles with adjustment for age and race in ARIC women and men N=12,339, 10-
year follow-up, 725 events
Sharrett AR, et al. Circulation. 2001;104:1108-1113.
Women Men
mg/dL 39 48 56 65 81 31 38 43 49 62 HDL-C Cholesterol (mmol/L) HDL-C Cholesterol (mmol/L)
Rel
ativ
e R
isk
HDL-C Provides Substantial CHD Prediction: ARIC Study
www.lipid.orgwww.lipid.org
Significant association found between incidence of CHD and HDL‐C levels after adjustment for other risk factors (P<0.001)12
Incidence of CHD Per 1000 SubjectsOver a 6-year Period According to HDL-C Level12
*PROCAM: Prospective Cardiovascular Münster study.The independent effect of raising HDL-C or lowering TG on the risk of coronary and cardiovascular morbidity and mortality has not been determined.
PROCAM* Results at 6 Years of Follow-up12Levels of LDL-C and Percent Change in HDL-C
www.lipid.orgwww.lipid.orgCastelli WP. Can J Cardiol. Jul 1988;4 Suppl A:5A-10A.
Framingham Heart StudyRisk of Coronary Artery Disease in Men Aged 50-70 by
LDL and HDL Cholesterol Levels
www.lipid.orgwww.lipid.orgAbbott RD, et al. Arteriosclerosis. May-Jun 1988;8(3):207-211.
23-46 47-55 56-66 67-1390
10
Rat
e pe
r 100
/12
year
s
267-412
Cholesterol (mg/dL)
238-266
212-237
124-211
HDL-C(mg/dL)
12 Year Incidence of Myocardial InfarctionFramingham Cohort-Women
www.lipid.orgwww.lipid.org
Objectives To evaluate if apoB and apoA-I add predictive
information of cardiac risk - fatal myocardial infarction - up and above TC, TG, and LDL-C
Subjects 98,722 males and 76,831 females, <20 – >80 years consecutively referred 1985–1996, no diagnosis or
treatment known, 67 months mean follow up Fatal myocardial infarction males 864, females 359
24
AMORIS
www.lipid.orgwww.lipid.org
0.92 1.28
HDL-cholesterol (mmol/L)
Ris
k ra
tio
0.30
0.6
0.8
0.4
1.0
1.91 2.251.20 1.57 1.81
Men Women
1.52
§
0.5
0.7
0.9
0.30.4
1.0
0.5
0.7
0.90.8
0.6
*§
§
§‡
AMORIS, Fatal Myocardial Infarction
www.lipid.orgwww.lipid.org
1.12 1.30
Apolipoprotein A-I (g/L)
Ris
k ra
tio
0.30
0.6
0.8
0.4
1.0
1.64 1.831.24 1.43 1.57
Men Women
1.43
§ §§
0.5
0.7
0.9
0.30.4
1.0
0.5
0.7
0.90.8
0.6
**
§
AMORIS, Fatal Myocardial Infarction
www.lipid.org
4.03.6
2.3
1.7
4.1
3.1
1.8 1.72.31.6
1.1 1.21.4
0.8 1.1 1.0
0.0
1.0
2.0
3.0
4.0
5.0
1.12 1.30 1.42 1.65
0.901.10
1.391.80
Men <70 years
Ris
k ra
tio
ApoA-I (g/L)
ApoB (g/L)
AMORIS, Fatal Myocardial Infarction
www.lipid.org
9.1
5.76.7
3.9
8.7
5.1
2.5
5.17.0 7.0
3.85.2
5.8
4.3 4.5
1.0
0.0
2.0
4.0
6.0
8.0
10.0
1.24 1.43 1.57 1.83
0.811.06
1.281.71
Women 70 years
Ris
k ra
tio
ApoA-I (g/L)
ApoB (g/L)
AMORIS, Fatal Myocardial Infarction
www.lipid.orgwww.lipid.orgMcQueen M et al. Lancet 2008;372:224-233.
Odd
s ra
tio (9
5% C
I)
Decile Median (SD distance from overall median)
-1 0 1 21.00
2.00
4.00
8.00 ApoB/ApoA-ITC/HDL-C
1 2 3 4 5 6 7 8 9 10ApoB/ApoA-I 0.43 0.53 0.60 0.66 0.72 0.78 0.85 0.93 1.04 1.28TC/HDL-C 2.74 3.37 3.82 4.23 4.64 5.08 5.58 6.21 7.15 9.20
Risk of MI for increasing decile medians (adjusted for age, sex
and region) of ratios of apoB/ApoA-I and TC/HDL-C
TC/HDL-C
ApoB/ApoA-I
Lipids, Lipoproteins & Apolipoproteins as Markers of risk of MI in52 Countries in The INTERHEART Study
www.lipid.org
Risk factor Odds ratio adjusted for age, sex, and smoking (99% CI)
Odds ratio adjusted for all (99% CI)
ApoB/ApoA-1 (fifth quintile compared with first)
3.87 (3.39-4.42) 3.25 (2.81-3.76)
Current smoking 2.95 (2.72-3.20) 2.87 (2.58-3.19)Diabetes 3.08 (2.77-3.42) 2.37 (2.07-2.71)Hypertension 2.48 (2.30-2.68) 1.91 (1.74-2.10)Abdominal obesity 2.22 (2.03-2.42) 1.62 (1.45-1.80)Psychosocial 2.51 (2.15-2.93) 2.67 (2.21-3.22)Vegetable and fruits daily 0.70 (0.64-0.77) 0.70 (0.62-0.79)Exercise 0.72 (0.65-0.79) 0.86 (0.76-0.97)Alcohol intake 0.79 (0.73-0.86) 0.91 (0.82-1.02)All combined 129.2 (90.2-185.0) 129.2 (90.2-185.0)
Yusuf S. European Society of Cardiology Congress 2004; August 28-September 1, 2004; Munich, Germany.
INTERHEART: Risk of Acute MI Associated with Risk Factorsin the Overall Population
www.lipid.orgwww.lipid.org
Characteristic Men with event (n=97) Men w/o event (n= 837) P-value
BMI (kg/m2) 27.6 27.6 n.s.Total cholesterol (mg/dl) 259.6 243.3 0.001HDL cholesterol (mg/dl) 46.8 51.8 0.0037
TC/HDL-Ratio§ 5.66 4.83 < 0.0001Systolic BP (mmHg) 139.0 136.6 n.s.Diastolic BP (mmHg) 84.9 84.5 n.s.Regular smoker % 53.9 28.0 <0.0001Physical activity % 25.7 35.6 0.05Diabetes % 8.4 3.1 0.0096Education (<12 years) % 74.2 73.7 n.s.Lp-PLA 2 (ng/ml) 292.3 263.4 0.0013C-reactive protein§ (mg/L) 2.49 1.54 < 0.0001
§ geometric means calculated from the log-transformed distributionKoenig et al. AHA 2003.
Age-Adjusted Baseline Characteristics of 934 men, Aged 45-64 YearsParticipating in the MONICA Augsburg Survey 1984/85
With Follow-up 1998
www.lipid.orgwww.lipid.org
1.63 0.91 0.78
18.32
11.75
3.29
<35 35-44 45+HDL Cholesterol (mg/dL)
0
5
10
15
20
Dea
ths/
1000
per
son-
yr CVD at BaselineAbsent (n=1940)Present (n=471) Trend p<0.001
Pekkanen J, Linn S, Heiss G, et al. N Engl J Med. Jun 14 1990;322(24):1700-1707.
11.2
CHD Mortality and HDL CholesterolLRC Men 10 Year Follow up
www.lipid.org
HDL-C = high-density lipoprotein cholesterol; CVD = cardiovascular disease; LDL-C = low-density lipoprotein cholesterol; TC = total cholesterol; RR = relative risk; CI = confidence interval.aRelative risks are adjusted for age at baseline as a continuous variable using Cox proportional hazards models.bHigher Χ2 values indicated better prediction of CVD mortality by that lipid level when added to age-only model.*Measured by an increase of 0.78 mmol/L (30 mg/dL).†Measured by an increase of 0.26 mmol/L (10 mg/dL).
Risk for CHD based on Lipid Fractions: Lipid Research Clinics Program Study
Cui Y et al. Arch Intern Med. 2001;161:1413-1419..
Comparison of lipid levels in predicting CVD mortality in men and womenRR (95% CI)a Χ2 for addition to modelb
MenNon-HDL-C* 1.19 (1.13 to 1.26) 24.3LDL-C* 1.11 (1.02 to 1.22) 5.0TC 1.16 (1.08 to 1.23) 14.4HDL-C† 0.77 (0.69 to 0.86) 23.2
WomenNon-HDL-C 1.15 (1.06 to 1.25) 8.3LDL-C 1.08 (0.96 to 1.22) 1.8TC 1.10 (0.99 to 1.22) 2.8HDL-C 0.77 (0.69 to 0.88) 18.5
www.lipid.orgwww.lipid.org
Hazard Ratio for Ischemic Heart Disease (IHD) as a Function of High-Density Lipoprotein (HDL) Cholesterol in
Quintiles in the Copenhagen City Heart Study
Frikke-Schmidt, R. et al. JAMA 2008;299:2524-2532.
www.lipid.orgwww.lipid.org
Hazard Ratios for Coronary Heart Disease Across Fifthsof Usual Lipids or Apolipoproteins
Di Angelantonio E, Sarwar N, Perry P, et al. JAMA. Nov 11 2009;302(18):1993-2000.
Quintile Mean (SD Distance From Overall Mean)
Haz
ard
Rat
io
www.lipid.orgwww.lipid.org
Hazard Ratios for Coronary Heart Disease or Ischemic Stroke Across Quintiles of Usual Triglyceride, HDL-C, and Non–HDL-C Levels
Coronary Heart DiseaseHDL-CTriglyceride Non-HDL-C
HDL-CTriglyceride Non-HDL-CIschemic Stroke
Haz
ard
Rat
io
Usual Mean Level, mg/dLH
azar
d R
atio
Usual Mean Level, mg/dL
Haz
ard
Rat
io
Usual Mean Level, mg/dL
Haz
ard
Rat
io
Usual Mean Level, mg/dL
Haz
ard
Rat
io
Usual Mean Level, mg/dL
Haz
ard
Rat
io
Usual Mean Level, mg/dL
Di Angelantonio E, Sarwar N, Perry P, et al. JAMA. Nov 11 2009;302(18):1993-2000.
www.lipid.orgwww.lipid.org
Hazard Ratios for Coronary Heart Disease Across Fifths of Non-HDL-Cby Levels of HDL-C and Fifths of HDL-C by Levels of Non-HDL-C
Haz
ard
Rat
io
Haz
ard
Rat
io
Usual Mean Non-HDL-C Level, mg/dL Usual Mean HDL-C Level, mg/dL
Non-HDL-C by levels of HDL-C HDL-C by levels of Non-HDL-C
Di Angelantonio E, Sarwar N, Perry P, et al. JAMA. Nov 11 2009;302(18):1993-2000.
www.lipid.orgwww.lipid.orgShahar, E. et al. Stroke 2003;34:623-631.
Multivariable-adjusted risk functions of incident ischemic stroke among men in relation to LDL-C, HDL-C, apoB, and triglycerides (TG)
www.lipid.orgwww.lipid.orgWilley, J. Z. et al. Arch Neurol 2009;66:1400-1406.
HRs and 95% CIs for Lipid Parameters as Continuous Variables and Risk of Ischemic Stroke
www.lipid.orgwww.lipid.org
Cooperative Lipoprotein Phenotyping Study Western Collaborative Group Study Israeli Ischemic Heart Disease Study Belgian Interuniversity Research on Nutrition
and Health Honolulu Heart Study
The following other observational studiesalso confirm low HDL as an importantindependent CAD risk factor:
HDL as CAD Risk Factor
www.lipid.orgwww.lipid.org
*95% confidence intervals for adjusted proportional hazards regression coefficients.
CHD = coronary heart disease; HDL-C = high-density lipoprotein cholesterol
% C
hang
e in
Ris
k pe
r In
crem
ents
(1 m
g/dL
)in
HD
L-C
*
FHS LRCF CPPT MRFIT FHS LRCF
CHD IncidenceMen Women
Coronary Heart Disease Incidence Is Related toHDL-C Levels in Various Clinical Trials
-10
-8
-6
-4
-2
0
Gordon DJ, et al. Circulation. 1989;79:8-15.
www.lipid.orgwww.lipid.orgFoody, J. M. et al. Circulation 2000;102:III-90-III-94.
Survival and HCL-C. Adjusted for age, HTN, DM, body mass index,TC, TG, LVEF, disease extent, and ITA use
www.lipid.orgwww.lipid.org
Event-free survival and HDL-C. Adjusted for age, HTN, DM,body mass index, TC, TG, LVEF, disease extent, and ITA use
Foody, J. M. et al. Circulation 2000;102:III-90-III-94.
www.lipid.orgwww.lipid.org
0.0
3.83.0
11.3
0
3
5
8
10
13
15 High HDL-Cc
Low HDL-Cd
N = 1032a98% of patients were discharged on 40 mg atorvastatin with mean LDL-C of 105 mg/dLbTarget lesion revascularization or major cardiac eventcMean HDL-C of 55 mg/dL (>40 mg/dL in men and >45 mg/dL in women)dMean HDL-C of 32 mg/dL (<40 mg/dL in men and <45 mg/dL in women)
Wolfram RM. Am J Cardiol. 2006;98:711-717.
Patie
nts,
%
0 to 30 Days >30 Days to 1 Year
Post Drug-Eluting Stent Implantationa
50
60
70
80
90
100
Surv
ival
Pro
babi
lity,
%
60 120 180 240 300 360
Days
P<.001 P<.001
P=.033
HDL-C 1 mg/dL Eventsb by 4%
High HDL-Cc
Low HDL-Cd
Eventb-Free Survival Death
0
Impact of Low HDL-C on Clinical Outcomes
www.lipid.org
Rel
ativ
e R
isk
of C
HD
HDL-C (mg/dL)
Low
High
Low High25 45 65
1
2
3
4
Higher the Level of HDL-C, Lower the Risk of CHD
www.lipid.orgwww.lipid.org
ARE THERE COMPLICATIONS/EXCEPTIONS TO THE RULE THAT WHEN IT COMES TO HDL, HIGHER IS BETTER?
www.lipid.orgwww.lipid.org
Unadjusted Rates of the Combined Primary End Point
deGoma, E. M. et al. J Am Coll Cardiol 2008;51:49-55.
All-cause mortalityMyocardial injury orIHD hospitalization
Even
t rat
e (%
)
www.lipid.orgwww.lipid.org
Adjusted Odds Ratios for the Combined Primary End Point
All-cause mortalityMyocardial injury orIHD hospitalization
Adj
. Odd
s R
atio
Mor
e E
vent
sFe
wer
Eve
nts
deGoma, E. M. et al. J Am Coll Cardiol 2008;51:49-55.
www.lipid.orgwww.lipid.orgWim A. van der Steeg, et al. J Am Coll Cardiol 2008;51:634–42.
European Prospective Investigation into Cancer and
Nutrition - Norfolk Study (EPIC-Norfolk)
Incremental Decrease in End Points Through Aggressive Lipid Lowering
(IDEAL)
1 2 3 40 5 6 8 9 1 2 3 40 5 6 8 9
<8.60
8.60-9.05
9.05-9.53
9.53-9.85
9.85-10.07
>10.07
Adjusted for apoA-I & apoB
OR for MCE
HD
L S
ize
(nm
)
RR for MCE
HD
L-C
(m
g/dL
)
<40
40-49
50-59
60-69
70-79
>80
Adjusted for apoA-I & apoB
NMR measured particle size
The data suggest that very high values of plasma HDL-C and HDL particle size can confer increased risk of atherosclerotic disease. Although this epidemiological observation parallels the findings of some previous animal studies , there is no clear biological explanation of how HDL can become pro-atherogenic.
www.lipid.orgwww.lipid.orgRobins, S. J. et al. Arterioscler Thromb Vasc Biol 2011;31:1208-1214.
The Prevalence of IR by the 4th Quartile of HOMA-IR is Shown across Quartiles of Fasting Plasma HDL-C and across Quartiles of Fasting
Plasma Triglycerides for Men and Women
www.lipid.orgwww.lipid.org
Unadjusted Kaplan-Meier Curves Showing the Cumulative Incidence of CHD Events in the Entire Study Sample (N=2910) with and without IR
and with Lower or Higher Values of Fasting Plasma HDL-C (A) and with Lower or Higher Values of Plasma Triglycerides (TG) (B)
Robins, S. J. et al. Arterioscler Thromb Vasc Biol 2011;31:1208-1214.
www.lipid.orgwww.lipid.org
Plasma TG (mmol/L)0.0 1.0 2.0 3.0 4.0 5.0
0.0
0.5
1.0
1.5
2.0
2.5H
DL-
C (m
mol
/L)
R = -0.48; P < 0.001.
Quebec Cardiovascular Study
Sniderman AD et al. Am J Cardiol. 2001;87:792-793.
www.lipid.orgwww.lipid.org
Cholesterol/HDL-C
0.0
3.0
4.0
5.0
T1< 4.97
T24.97–6.37
T3 637
< 50th< 1.56
> 50th 1.56
1.0
2.0
TG
IHD, ischemic heart disease.
Quebec Cardiovascular Study
Sniderman AD et al. Am J Cardiol. 2001;87:792-793.
www.lipid.orgwww.lipid.org
Atherogenicity of Triglycerides and Etiology of Abnormal Lipid Metabolism
Adapted from Bays H, et al. Expert Opin Pharmacother. 2003;4:1901-1938.
Liver
VLDLTG
HDL
LDL
TG CECETP
CETPTG CE
Renal Clearance
TG
Lipases
Small, denseHDL-P
TG
Small, denseLDL-P
Lipases
HDL-C
VisceralAdiposopathy
FFA
Elevated TG
TG
VLDL-C
? LDL-C Non–HDL-C
Apo B
LDL P #
Apo A-I Apo A-I
Apo A-I
Apo B Apo B
Apo B
www.lipid.orgwww.lipid.org
Framingham Offspring Study
1000
1200
1400
1600
1800
20 40 60 80 100
100
120
140
160
180
0 100 200 300 400
100
120
140
160
180
1000
1200
1400
1600
1800
LDL
Part
icle
s (n
mol
/L)
LDL
Cho
lest
erol
(m
g/dL
)
HDL Cholesterol (mg/dL) Triglycerides (mg/dL)
LDL Particles
LDL Cholesterol
LDL Particles
LDL Cholesterol
Otvos et al., Amer J Cardiol 2002;90:22i-29i.
Relations of LDL Particles and LDL Cholesterol to Levels of HDL Cholesterol and Triglycerides
www.lipid.orgwww.lipid.org
01020304050607080
Downs JR et al. JAMA 1998;279:1615-1622.
HDL-C Levels
Placebo
Lovastatin
<34 mg/dl 35–39 mg/dl >40 mg/dl
71
40
68
41 4435
44% RR 40%
RR
20% RR
AFCAPS/TexCAPS: Risk Reduction by HDL-C Tertile at Baseline
www.lipid.orgwww.lipid.org
0.30
0.25
0.20
0.15
0.10
0.05
0HDL-C 35 mg/dL HDL-C 35 mg/dL
Dec
reas
e in
min
imum
lum
en
diam
eter
(mm
)
P = 0.09
P = 0.0004
Fluvastatin
Placebo
P for interaction = 0.01
Ballantyne CM et al. Circulation 1999;99:736–743.
-0.065
-0.274
-0.036
-0.083
Lipoprotein and Coronary Atherosclerosis Study (LCAS) Benefit of Statin Therapy in Patients With Low HDL-CLRC Men 10 Year Follow up
www.lipid.orgwww.lipid.org
29.9
20.922.6
17
0
5
10
15
20
25
30
35
Placebo
Simvastatin
Mean LDL = 131 mg/dL.
HDL-C LevelBelow Tertile*(< 35 mg/dL)
HDL-C LevelAbove Tertile*( 42 mg/dL)
Abs
olut
e Ev
ent R
ate
(%)
*P < 0.05
25% 18.7%
CHD Event Rate and HDL-C Levels in the Heart Protection Study (HPS)
Heart Protection Study Collaborative Group. Lancet. 2002;360:7-22.
www.lipid.orgwww.lipid.org
*CHD death, nonfatal MI, and stroke.
0.4 0.6 0.8 1 1.2 1.4 1.6
LDL-C
HDL-C
mg/dL
43
53
Interaction Pvalue
43–53 0.007
Pravastatin Better Placebo Better
1320.88
132–159 0.70.88
1590.77
0.64
0.93
1.09
Primary End Point* by Tertiles of LDL-C and HDL-C in PROSPER
Shepherd J et al. Lancet. 2002;360:1623-1630.
www.lipid.orgwww.lipid.org
• Patients with stable coronary disease (N=143) received gemfibrozil + niacin + cholestyramine or placebo
– Baseline LDL-C 128 mg/dL• Group with greatest HDL-C increase had greatest chance of being CV event
free– HDL-C increase was a significant predictor of event-free survival even after
adjustment for LDL-C changes (p<0.01)
Devendra GP, et al. J Cardiovasc Pharmacol Ther. Dec 2010;15(4):380-383.
Post-hoc Analysis of AFREGS
www.lipid.orgwww.lipid.org
Angiographic Effects of Lipid Drug Classes Meta‐Analysis, 12 Trials
%s = 3.0 – 0.076 (%HDL-C) + 0.06 (%LDL-C) R2 = 0.96; P<.004
S+N=simvastatin + niacinHATS (HDL - Atherosclerosis Treatment Study) data not shown in original study.
ASTEROID
Brown BG, et al. Curr Opin Lipidol. 2006;17:631–636.
www.lipid.orgwww.lipid.org
LDL-C <87.5 mg/dL HDL-C >7.5%
Nicholls SJ, et al. JAMA. 2007;297(5):499-508.
Regression of Coronary Atherosclerosis Associated with On-Treatment Levels of LDL-C and
Percent Change in HDL-C
www.lipid.orgwww.lipid.org
Percentage increase in HDL-C and percentage relative risk reduction in CHD: data from 17 prospective, randomized trials combining a total of 44,170 patients and 3,869 CHD events
Alsheikh-Ali AA, et al. AHA Scientific Sessions, 2004.
HDL-C and CHD Risk: Meta Analysis
www.lipid.orgwww.lipid.orgBarter P, et al. N Engl J Med. 2007;357:1301-1310.
Low HDL-C was Associated with Higher CV Events in Patients with LDL-C < 70 mg/dL
Post-hoc Analysis of TNT
www.lipid.orgwww.lipid.orgJafri H, et al. Ann Intern Med. 2010;153:800-808.
Black circles indicate patients who are receiving statin interventions, and green circles indicate patients who are receiving a nonstatin control.
Associations Between HDL-C and Cardiovascular Outcomes: Statin Treatment Does Not Alter Inverse Relationship
www.lipid.orgwww.lipid.org
Incidence rates per 100 person years of exposure for the JUPITER primary endpoint according to on-treatment concentrations of HDL-C and on treatment concentrations of ApoA1
ApoA1HDL-C
Ridker PM, Genest J, Boekholdt SM, et al. Lancet. Jul 31 2010;376(9738):333-339.
HDL-C and Risk Reduction in JUPITER
www.lipid.orgwww.lipid.org
• Electronic medical record study of 1,512 high-risk patients
• Low HDL-C present in 66% of overall patients
– Most prevalent (79%) in patients with LDL-C 70 mg/dL
Alsheikh-Ali AA, et al. Am J Cardiol. 2007;100:1499-1501.
Low HDL-C defined as 40 mg/dL in men and 50 mg/dL in women
Prevalence of Low HDL-C in Patients with Documented CHD
Or CHD Risk Equivalents and Controlled LDL-C
www.lipid.orgwww.lipid.org
WHEN IT COMES TO THE GENETICS OF HDL, PUZZLE PIECES DO NOT ALWAYS FIT TOGETHER NEATLY
www.lipid.orgwww.lipid.org
Tangier disease (deletion of nucleotides 3283 & 3284) produces a premature stop codon, resulting in the expression of a nonfunctional molecule1
Up to one half of the cases of familial hypoalphalipoproteinemia are due to a mutation in the gene for ABCA11
The R219K variant of ABCA1 is associated with increased ABCA1 expression, higher mean serum HDL levels, and lower frequency of CAD in Dutch men2
ABCA1 Mutations
1. Remaley et al. Proc Natl Acad Sci USA 1999;96:12685-90.2. Clee SM et al. Circulation 2001;103:1198-205.
www.lipid.orgwww.lipid.org
Table 2. Summary of Data Available on CETP Genotypes, CETP Phenotypes, Lipid Levels, and Coronary Outcomesa.
Frikke-Schmidt, R. et al. JAMA 2008;299:2524-2532.
Characteristics of Participants of Danish Descent in the3 Studies Conducted in Copenhagen, Denmark
www.lipid.orgwww.lipid.org
Plasma High-Density Lipoprotein (HDL) Cholesterol and Apolipoprotein A-I Levels for Heterozygous Carriers of ABCA1 Mutations in the
Copenhagen General Population Study
Frikke-Schmidt, R. et al. JAMA 2008;299:2524-2532.
www.lipid.orgwww.lipid.org
Risk of Ischemic Heart Disease as a Function ofABCA1 Mutations in the 3 Different Studies
Frikke-Schmidt, R. et al. JAMA 2008;299:2524-2532.
www.lipid.orgwww.lipid.org
APOPROTEIN A1
www.lipid.orgwww.lipid.org
ApoA-INichinan significantly decreases the ability of apoA-I to bind fibroblasts and promote lipid efflux1
ApoA-IZavalla results in about a 4-fold reduction of serum HDL and increased risk for premature CAD2
Patients with apoA-I(Lys 0) have low HDL due to enhanced catabolism3
ApoA-I (L178P) leads to endothelial dysfunction, increased arterial wall thickness, and premature CAD4
Apoprotein A-I Mutations
1. Huang W et al. Arterioscler Thromb Vasc Biol 2000;20:210-6.2. Miller M, et al. Arterioscler Thromb Vasc Biol 1998;18:1242-7.3. Tilly-Keisi M et al. Arterioscler Thromb Vasc Biol 1997;17:873-80.4. Hovingh GK et al. J Am Coll Cardiol 2004;44:1429-35.
www.lipid.orgwww.lipid.org
Effect of APOA1 genotype and genotype combinations on HDL cholesterol levels and theoretically predicted and observed risks of
IHD and MI
Haase, C. L. et al. J Clin Endocrinol Metab 2010;95:E500-E510.
www.lipid.orgwww.lipid.org
Effect of APOA1 genotype and genotype combinations onapoA-I levels and theoretically predicted and observed risks of
IHD and MI
Haase, C. L. et al. J Clin Endocrinol Metab 2010;95:E500-E510.
www.lipid.orgwww.lipid.org
Intravenous injections of HDL and apoA-I prevent atherogenesis in rabbits fed high-cholesterol diets1,2
Intravenous injection of HDL in rabbits with established atherosclerotic disease results in significant slowing in rates of disease progression and can stimulate regression of established plaques3
Transgenic apoE-deficient mice overexpressing human apoA-I have elevated serum HDL levels and resistance to atherosclerosis4,5,6
Mice with an LDL receptor deficiency fed an atherogenic diet experience a dramatic 70% decrease in aortic atheromatous plaque burden within 4 weeks after being injected with an adenovirus that contains the human apoA-I gene7
The residual plaque in treated animals also has fewer foam cells, has a lower density of macrophages, and is more fibrotic, possibly rendering it more stable.
Apoprotein A-I and Atherosclerosis
1. Badimon J et al. J Clin Invest 1990;85:1234-41.2. Miyazak et al. Arterioscler Thromb Vasc Biol 1995;15:1882-8.3. Badimon J et al. Circulation 1992;86(suppl III):86-94.4. Plumb A et al. Proc Natl Acad Sci USA 1994;91:9607-11.5. Paszty C et al. J Clin Invest 1994;94:899-903.6. Benoit P et al. Circulation 1999;99:105-10.7. Tangirala R et al. Circulation 1999;100:1816-22.
www.lipid.orgwww.lipid.org
Apo B
Total Cholesterol (mg/dl)
LDL Cholesterol
HDL CholesterolTriglyceridesApo A-I
Age (y)
nA-IM Carriers
103 ±35
189 ±50
132 ±35
20 ±10
186 ±107
78 ±28
42 ±17
21Controls
90 ±28
215 ±56
143 ±46
49 ±20
120 ±61
132 ±27
42 ±17
42
97 ±20
191 ±33
133 ±29
27 ±6
142 ±49
97 ±13
Low-HDL
44 ±17
21
A-IMilano Cardiovascular Study
Sirtori CR, et al. Circulation. Apr 17 2001;103(15):1949-1954.
www.lipid.orgwww.lipid.org
0.5
0.6
0.7
0.8
0.9
1.0
0.5
1.0
1.5
2.0
Aver
age
IMT
(mm
)
Max
imum
IMT
(mm
)
Mean ±SEM
A-IMCarriers
ControlsLow-HDL A-IMCarriers
ControlsLow-HDL
A-IMilano Cardiovascular StudyCarotid Intima-Media Thickness
Sirtori CR, et al. Circulation. Apr 17 2001;103(15):1949-1954.
www.lipid.orgwww.lipid.org
N C
Apo-AI
Cys
N C
Apo-AIMilano monomer
Cys-
N
Cys
N
Apo-AIMilano dimer
www.lipid.orgwww.lipid.orgRoma P, Gregg RE, Meng MS, et al. J Clin Invest. Apr 1993;91(4):1445-1452.
A-IM Dimer Has A Long Half-Life
www.lipid.orgwww.lipid.org
AIM INFILTRATION INTO THE CORONARY WALLREDUCES STENT RESTENOSIS
Control
AIMPROXIMALSEGMENT
DISTAL SEGMENT
Kaul S, et al. Circulation. May 27 2003;107(20):2551-2554.
www.lipid.orgwww.lipid.orgNissen SE, et al. JAMA. Nov 5 2003;290(17):2292-2300.
Overall r Apo A-I Milano ( ETC-216 ) produced 4.5 % net regression of coronary atheroma volume in
6 Weeks
N=47 patients with ACS
Baseline Post-Rx
-1.06
0.14
-1.5
-1
-0.5
0
0.5
-14.1
-2.9
-20
-15
-10
-5
0
(N=36) (N=11) (N=36) (N=11)
Mean Change in Atheroma Volume (%)
Mean Change in Total Atheroma Volume
(mm3)
P<0.02
P<0.97
(P<0.001)
P<0.97
A-Im A-ImPlacebo Placebo
10 mm coronary segments with the largest plaqueshowed 11 % net regression: Nichols JACC 2006
Rapid Regression of Human Coronary Plaque after 5 Weekly Intravenous Injections of Recombinant rApo A-Imilano (ETC-216)
www.lipid.orgwww.lipid.org
CHOLESTEROL ESTER TRANSFER PROTEIN AND CETP INHIBITION
www.lipid.orgwww.lipid.org
Reverse Cholesterol Transport
Toth PP. Am J Cardiol. Nov 7 2005;96(9A):50K-58K; discussion 34K-35K.
www.lipid.orgwww.lipid.org
No evidence of premature atherosclerosis in any of the 10 homozygotes.Trend toward longevity in 2 families with a 100 year old heterozygote.
Lipid Profile of 5 Families With HALP
Group
Homozyg.(10)
Heterozyg.(20)
Unaffected(10)
Age (yr)
58(51-68)
49(19-100)
48(20-71)
Total
271*
195
189
HDL
164*(209%)
66(25%)
53
LDL
77*(-44%)
111(-5%)
117
A-I B
0.54*(-31%)
0.66(-15%)
0.78
2.13*(72%)
1.49*(20%)
1.24
CETP(g/mL)
0*
1.4*(0.3)
2.3
Cholesterol (mg/dL) Apoproteins(mg/mL)
Are CETP Gene Mutations Atheroprotective?
Inazu et al. N Engl J Med. 1990;323:1234-1238.
*Significant difference from unaffected P<0.05; numbers in parentheses are % difference from unaffected controls.
www.lipid.orgwww.lipid.org
CETP Mutation:
Curb JD, et al. J Lipid Res. May 2004;45(5):948-953.
HDL-C <60 mg/dl HDL-C 60 mg/dl
CH
D In
cide
nce
(rat
e/10
00 p
erso
n-ye
ars)
171/1713*
31/509**5/76
2/42
*Number of CHD events/men at risk
**Significantly lower risk compared to men with HDL-C <60 mg/dl and without a CETP mutation (p<0.05)
absentpresent
18
15
12
9
6
3
0
Age-Adjusted 6-Year CHD/CVD Rates for Elderly Japanese American Men With and Without CETP Mutations
www.lipid.orgwww.lipid.org
Table 2. Summary of Data Available on CETP Genotypes, CETP Phenotypes, Lipid Levels, and Coronary Outcomesa.
Thompson, A. et al. JAMA 2008;299:2777-2788.
Table 2. Summary of Data Available on CETP Genotypes, CETP Phenotypes, Lipid Levels, and Coronary Outcomesa.
www.lipid.orgwww.lipid.org
Figure 2. Associations of CETP Genotypes With CETP Phenotypesand Lipid Levels
Thompson, A. et al. JAMA 2008;299:2777-2788.
www.lipid.orgwww.lipid.org
Figure 4. CETP Genotypes and Coronary Risk, Grouped by Recorded Study Characteristics
Thompson, A. et al. JAMA 2008;299:2777-2788.
www.lipid.orgwww.lipid.org
Figure 5. Observed Per-Allele Odds Ratios for Coronary Disease With CETP Variants vs Odds Ratios Derived From Available Prospective
Studies of HDL-C Levels
Thompson, A. et al. JAMA 2008;299:2777-2788.
www.lipid.orgwww.lipid.orgRidker P M et al. Circ Cardiovasc Genet 2009;2:26-33.
Figure 2. Distribution of HDL-C (left), ApoA1 (middle), and Kaplan Meier estimates of cumulative incidence of myocardial infarction (right)
according to the genotype of SNPs at the CETP locus
www.lipid.orgwww.lipid.orgCopyright ©2009 American Heart Association
Vasan, R. S. et al. Circulation 2009;120:2414-2420.
Kaplan-Meier curves showing survival free from CVD over the follow-up period in individuals above vs below the median plasma CETP activity
www.lipid.orgwww.lipid.org
Multivariable-adjusted relations of plasma CETP activity (continuous variable) to the incidence of CVD on follow-up
Copyright ©2009 American Heart Association
Vasan, R. S. et al. Circulation 2009;120:2414-2420.
www.lipid.orgwww.lipid.org
Primary Endpoint: Time to First MCVE*: Kaplan-Meier Plot
* Major cardiovascular event: CHD death, non-fatal MI, stroke or hospitalization for unstable angina
www.lipid.orgwww.lipid.org
LIPOPROTEIN LIPASE
www.lipid.orgwww.lipid.org
Ser447(stop) in LPL was found to increase HDL levels in Dutch men, possibly from increased lipolytic activity1
Gly188→Glu in LPL develop hypertriglyceridemia and low HDL/apoA-I and have increased risk for developing ischemic heart disease2
The Asn291→Ser in LPL is a key contributor to the development of low HDL and premature CAD. The Copenhagen City Heart Study found that this mutation increases the risk for CAD and ischemic cerebrovascular disease 2-fold in women3,4
Lipoprotein Lipase Mutations
1. Kuivenhoven J et al. Arterioscler Thromb Vasc Biol 1997;17:595-9.2. Nordestrgaard B et al. Circulation 1997;96:1737-44.3. Reymer P et al. Nat Genet 1995;10:28-34. 4. Wittrup HH et al. Circulation 2000;101:2393-7.
www.lipid.orgwww.lipid.org
Associations between plasma high density lipoprotein (HDL) cholesterol and triglyceride levels and 6 polymorphisms of the
lipoprotein lipase gene, grouped by various study characteristics using a dominant genetic model (carriers vs. noncarriers).
Sagoo G S et al. Am J Epidemiol. 2008;168:1233-1246.
www.lipid.orgwww.lipid.org
Meta-analyses of studies of coronary heart disease and 7 polymorphisms of the lipoprotein lipase gene, grouped by various
study characteristics using a dominant genetic model(carriers vs. noncarriers).
Sagoo G S et al. Am J Epidemiol. 2008;168:1233-1246.
www.lipid.orgwww.lipid.org
HEPATIC LIPASE
www.lipid.orgwww.lipid.org
Allelic variation in the gene for HL gives rise to 25% of the interindividual differences in serum HDL levels1,2
G250A in the promoter decreases HL activity, raises HDL2, and increases the concentration of large, buoyant LDL3
In the Familial Atherosclerosis Treatment Study, combination therapy with either niacin and colestipol or lovastatin and colestipolwas associated with decreased hepatic lipase activity and coronary artery plaque regression as a result of elevations in LDL particle buoyancy and serum HDL2 concentrations4
C480T in the promoter significantly decreases HL activity, and is associated with elevations in apoA-I and HDL5
C514T increases the size of HDL2 fraction but has no effect on LDL particle size6
Hepatic Lipase Mutations
1. Cohen et al. J Clin Invest 1994;94:2377-84. 2. Guerra R et al. Proc Natl Acad Sci USA 1997;94:4532-7.3. Zambon A et al. Aterioscler Thromb Vasc Biol 1998;18:1723-29.4. Zambon A et al. Circulation 1999;99:1959-64.5. Murtomaki S et al. Aterioscler Thromb Vasc Biol 1997;17:1879-84.6. Couture P et al. Arterioscler Thromb Vasc Biol 2000;20:815-22.
www.lipid.orgwww.lipid.org
HDL FUNCTIONALITY
www.lipid.orgwww.lipid.org
SAA1
PON1ApoA-I
SAA2
ApoHApoA-IV
C3
C4A
C4B
VTN
ApoA-II
ApoL-1
ApoDSAA4
ApoC-IV
ApoC-III
ApoC-II
ApoC-ILCATCETPPLTP
ApoM
ApoF
ApoE
HRP SERA1
KNG1
ORM2
AGT
AHSG AMP
SERF2 SERF1
Clusterin
PON3
C9
Lipid Metabolism
ITIH4TTR RBP4 TFFGA
HPX
Proteinase Inhibitor
ComplementRegulation
Acute Phase Response
Gene Ontology Analysis of HDL Proteins
Slide from J Heinecke.
www.lipid.orgwww.lipid.orgAssmann G et al. Ann Rev Med 2003;54:321-341.
The antioxidative effects of HDL are related to presence
of α-tocopherol and other antioxidants including apoA-I and apoJ. These scavenge
ROS and prevent lipid peroxides (LOOX) and oxidized CE (CE-OOX).
LDL
Phospholipids Fatty Acids
Oxidized Phospholipids
(POVPC, PGPC)
LDL-oxO2- , OH-
HDL12-lipooxygenase
HODE HETE
Oxidized phospholipids on LDL are deleterious. These are
generated by seeding molecules like HPODE and
HPETE as catalysts generated by 12-liopoxygenase and lipid
peroxides (LOOX) and oxidized CE (CE-OOX).
HPODE = Hydroperoxyoctadecadienoic acid HPETE = Hydroperoxycicosatetraenoic acid
HDL overtakes seeding molecules and degrades them using paraoxonase
(PON), PAF-AH or glutathione peroxidase
(GPX)
A-I, J
α-TOC PON, PAF-AH
GPX
HDL: Anti-inflammatory Activity
www.lipid.orgwww.lipid.orgFogelman AM Nature Medicine 2004;10:902-903.
A-I A-I
Anti-inflammatory HDL
Pro-inflammatory HDL
A-I
Myeloperoxidase damages apoA-I at the tyrosine molecules
Nitrotyrosine
Chlortyrosine
↓apoA-I ↓ Paraoxonase, other factors ↑ Proinflammatory factors,
other factors
In inflammatory states HDL can change its
properties and become pro-inflammatory
HDL Inflammatory Activity
www.lipid.orgwww.lipid.org
A-I
A-I
A-I
A-IIA-I
A-I
J
J
A-I ApoAI
A-II ApoAII
PON1 Paraoxonase
LCAT Lecithin:cholesterol acyltransferase
CETP Cholesteryl ester transfer protein
PLTP Phospholipid transfer factor
J ApoJ
sPLA2 Secretory phospholipase A
SAA Serum Amyloid A
Normal HDL
APR HDL
Rohrer M et al. Curr Opin Lipidol 2004;15:269-278.
Acute Phase Reaction (APR) and HDL
www.lipid.orgwww.lipid.org
Coronary Artery Disease Status According to Quartile of Efflux Capacity
www.lipid.orgwww.lipid.org
Odds Ratios for Coronary Artery Disease According to Efflux Capacity
and Selected Risk Factors
www.lipid.orgwww.lipid.org
CONCLUSIONS• HDL‐C is consistently shown to be an independent risk factor for CHD based on observational studies around the world in both men and women•Meta‐analyses and multivariate regression support the hypothesis that raising HDL‐C reduces risk for CVD•HDL particles have a complex proteosome that confers diverse protective and potentially toxic functionality •Polymorphisms in components of the HDL proteosome do not always impact risk for CHD as predicted•A variety of approaches to HDL therapy are being tested prospectively in randomized trials