A Pilot Study of Urinary Markers of Endothelial Function & Oxidant Stress for the Prediction of Cardiovascular Disease Risk with Antiretroviral therapy.

A Pilot Study of Urinary Markers of Endothelial Function & Oxidant

Stress for the Prediction of Cardiovascular Disease Risk with

Antiretroviral therapy

Michael S. Boger1, Ginger Milne2, Husamettin Erdem1, Valerie Mitchell1, David W. Haas1, Jason Morrow2, Todd Hulgan1

Divisions of 1Infectious Diseases and 2Clinical PharmacologyDepartment of Medicine

Vanderbilt University School of Medicine

Introduction

• Potent antiretroviral therapy (ART) dramatically improves morbidity & mortality from HIV infection

• ART may increase cardiovascular disease (CVD) risk Friis-Moller N et al. NEJM 2003; 349: 1993-2003DAD Study Group. NEJM 2007; 356: 1723-35

• Traditional Framingham prediction may underestimate CVD risk in HIV infected patients on ART

Law MG et al. HIV Med 2006; 7: 218-30De Socio GV et al. J Infect 2008; 57: 33-40

Introduction

• Inflammation, oxidant stress, & endothelial dysfunction are central to the pathogenesis of atherosclerosis/CVD Deakin S et al. Atheroscler Thromb Vasc Biol 2007; 27: 1390-5

• HIV-infection &/or ART may influence these factors

• hsCRP correlates with CVD outcomes in the general population; limited data in HIV infection

Pearson TA et al. Circulation 2003; 107: 499-511

• A urinary marker (isoprostane) of oxidant stress correlated with traditional CVD risk factors

Wang B et al. Atherosclerosis 2006; 184: 425-30

Basarici I et al. Coron Artery Dis 2007; 18: 615-20

• Vascular reactivity appears impaired in HIV infectionTorriani F et al. 4th IAS Conference 2007. Abstr WEAB302 Solages A et al. Clin Infect Dis 2006; 42: 1325-32

Introduction

• Eicosanoids are involved in inflammation, endothelial function, & oxidant stress

• Urinary assays for eicosanoids are noninvasive, precise, accurate, & reproducible

Milne GL et al. J Biol Chem 2005; 280: 25178-84

• We suspect that quantifying eicosanoids may help assess cardiovascular risk in this population

Eicosanoids: Inflammation, Endothelial Function, & Oxidant Stress

Membrane Phospholipids

Arachidonic Acid

COX

Prostaglandin-H2

Thromboxanes Prostacyclin Prostaglandins

Isoprostanes

Autoxidation

(Platelets) (Endothelium) (Smooth Muscle)

8-ISO-PGF2α(PGF2α)

11 DTxB2(TxB2)

2,3 DN-6KETO(PGI-M)

PGE-M

Khanapure SP et al. Curr Top Med Chem 2007; 7: 311-40

Vasoconstriction

Platelet aggregation

Oxidative tissue damage

Vasoconstriction

Platelet activation

Chemotaxis

Vasodilation

Inhibits platelet aggregation

Inhibits vascular smooth muscle proliferation

Vasoconstriction

Vascular smooth muscle proliferation

Eicosanoids: Inflammation, Endothelial Function, & Oxidant Stress

Membrane Phospholipids

Arachidonic Acid

COX

Prostaglandin-H2

Thromboxanes Prostacyclin

Isoprostanes

Autoxidation

(Platelets) (Endothelium) (Smooth Muscle)

8-ISO-PGF2α(PGF2α)

11 DTxB2(TxB2)

2,3 DN-6KETO(PGI-M)

PGE-M

Khanapure SP et al. Curr Top Med Chem 2007; 7: 311-40

Prostaglandins

Research Hypotheses & Aim

Hypotheses• ART metabolic effects include increased eicosanoid

production• These biomarkers are coupled with inflammation,

oxidant stress, & endothelial dysfunction

Aim• To correlate novel markers of inflammation, oxidant

stress, & endothelial function with serum lipids & hsCRP in HIV infected patients on ART

Study Population

• HIV infection > 12 mos • On ART including ≥ 2 NRTIs• HIV RNA < 400 copies/mL

• No CAD or DM• No lactic acidosis• No malignancy• No aspirin use• No tobacco use

Methods & Statistical Analyses

• Cross-sectional pilot study of a prospective cohort• Urine eicosanoids quantified by liquid

chromatography/mass spectroscopy• Spearman’s correlation• Wilcoxon rank-sum

Characteristics of Study Subjects(N=33)

Age in years, median (IQR) 45 (39-51)

Female sex, n (%) 8 (24%)

Non-white race, n (%) 18 (55%)

BMI, median (IQR) 26 (24-27)

hsCRP mg/dL, median (IQR) 2.2 (0.76-5.87)

CD4 cells/mm3, median (IQR) 515 (324-738)

HIV RNA copies/mL, median (range)

Non-HDL mg/dL, median (IQR)

Lipoatrophy, n (%)

50 (50-470)

124 (106-170)

9 (27%)

PI-based ART, n (%)

Intermittent NSAID use

15 (45%)

20 (55%)

Results: Urinary Eicosanoids

Eicosanoid*

(ng/mg cr)

Reference1-4 Study Population

PGF2α 1.60 ± 0.60 2.01 ± 1.40

TxB2 0.37 ± 0.14 0.31 ± 0.17

PGI-M 0.14 ± 0.05 0.13 ± 0.06

PGE-M M: 10.40 ± 1.50

F: 6.00 ± 0.70

M: 13.57 ± 15.37

F: 6.57 ± 5.39Morrow JD et al. J Chromatogr 1993; 612: 179-85

Murphey LJ et al. Anal Biochem 2004; 334: 266-75

Daniel VC et al. J Chromatogr 1994; 653: 117-22

Morales CR et al. Clin Chim Acta 2001; 314: 93-9

* Data are mean ± 1 SD

Results: Gender Differences in Urinary Eicosanoids

PGF2α*(ng/mg cr)

Total (N= 33)

Men (N=25)

Women (N =8)

Overall 1.9 (1.3-2.4) 1.8 (1.3-2.4) 2.0 (1.3-2.3)

PI use

Yes

No

1.4 (0.76-2.3)

2.2 (1.4-2.5)

1.4 (0.73-2.4)

1.9 (1.4-2.4)

1.3 (0.89-1.9)

2.3 (2.1-5.5)

Lipoatrophy

Yes

No

2.3 (1.4-2.5)

1.8 (1.1-2.3)

2.3 (1.4-2.9)

1.6 (0.92-2.3)

1.6 (0.76-2.5)

2.0 (1.6-2.2)

*Data are median (IQR)

No association with age, race, CD4, HIV RNA, NSAID use, BMI

p=0.04p=0.07

Results: Gender Differences in Urinary Eicosanoids

PGI-M*(ng/mg cr)

Total (N= 33)

Men (N=25)

Women (N =8)

Overall 0.12 (0.08-0.16) 0.12 (0.08-0.16) 0.10 (0.07-0.17)

PI use

Yes

No

0.11 (0.07-0.15)

0.12 (0.09-0.17)

0.12 (0.07-0.15)

0.12 (0.09-0.17)

0.07 (0.04-0.24)

0.13 (0.07-0.17)

Lipoatrophy

Yes

No

0.15 (0.12-0.17)

0.11 (0.08-0.14)

0.15 (0.12-0.18)

0.11 (0.08-0.14)

0.11 (0.04-0.17)

0.10 (0.07-0.19)

*Data are median (IQR)

No association with age, race, CD4, HIV RNA, NSAID use, BMI

p=0.07

Results: Correlation of Urinary Eicosanoids & hsCRP

Eicosanoid* Total

(N= 33)

Men

(N=25)

Women

(N =8)

PGF2α 0.36, 0.04 0.33, 0.11 0.29, 0.49

TxB2 0.33, 0.06 0.12, 0.55 0.84, 0.01

PGI-M 0.39, 0.04 0.34, 0.12 0.89, 0.02

PGE-M 0.26, 0.19 0.25, 0.30 0.29, 0.54

*Data are correlation coefficient, p-value

Results: Correlation of Urinary Eicosanoids & Non-HDL Cholesterol

Eicosanoid* Total

(N= 33)

Men

(N=25)

Women

(N =8)

PGF2α 0.02, 0.91 0.03, 0.89 -0.19, 0.65

TxB2 0.38, 0.04 0.50, 0.02 -0.11, 0.80

PGI-M 0.19, 0.33 0.42, 0.06 -0.67, 0.15

PGE-M 0.23, 0.26 0.35, 0.14 -0.14, 0.76

*Data are correlation coefficient, p-value

Summary

• Minimally invasive, reliable methods to predict metabolic problems of ART are needed

• Urinary eicosanoids may be promising markers of CVD health in HIV patients on ART

• These markers of inflammation, oxidant stress & endothelial function correlated with non-HDL cholesterol & hsCRP

• Sex-specific associations of urinary eicosanoids with traditional CVD risk factors were seen

• Larger studies including CVD outcomes are needed to confirm & expand our findings

Acknowledgements

Jason Morrow, MD

Acknowledgements

• Persons with HIV infection who volunteered for this study

• NIH/NCCAM Career Development Award K23 AT002508

• Vanderbilt CTSA grant 1UL1RR024975 (NCRR/NIH)

• Vanderbilt-Meharry CFAR P30 AI54999

• NIH Molecular Basis of Infectious Diseases Training Program T32 AI07474-13

• Vanderbilt University GCRC

• Vanderbilt Human Analytical Isoprostane Core Facility

A Pilot Study of Urinary Markers of Endothelial Function & Oxidant

Stress for the Prediction of Cardiovascular Disease Risk with

Antiretroviral therapy

Michael S. Boger1, Ginger Milne2, Husamettin Erdem1, Valerie Mitchell1, David W. Haas1, Jason Morrow2, Todd Hulgan1

Divisions of 1Infectious Diseases and 2Clinical PharmacologyDepartment of Medicine

Vanderbilt University School of Medicine