Cardiovascular risk A key long-term comorbidity for HIV+

Dr Graeme Moyle

Chelsea and Westminster Hospital

London, United Kingdom

1999-2000 2001-2 2003-4 2005-6 2007-8 2009-11

Smith CJ et al. Lancet 2014

ICD-10 (until 2004)

CoDe (from 2004 on)

Age-standardised incidence rates (D:A:D Study)

All-cause (both AIDS and non-AIDS) mortality has

decreased with cART

SCOPE cohort: subclinical atherosclerosis in HIV

Hsue PY et al, AIDS 2009

mean

intima-media

thickness

mm

HIV is associated with a relative increase

in cardiovascular events

Age Group

Inci

de

nce

Rat

e o

f A

MI p

er

1,0

00

p

ers

on

ye

ars

PLHIV HIV-negative

Drozd et.al. 2017 JAIDS

“U-shaped” relationship between BMI and poor outcomes in PLHIV

Achhra et.al. 2018 JAIDS

SMART Study Group. N Eng J Med. 2006;355: 2283-2296.

No. at Risk

Treatment

Interruption 2720 2070 1663 1292 1041 867 693 543 443 375 273 157

Continuous Treatment

2752 2077 1692 1307 1070 899 713 563 462 380 282 165

44 40 36 32 28 24 20 16 12 8 4 0 0.00

0.05

0.10

0.15

0.20

Major Cardiovascular, Renal, or Hepatic Disease

Hazard ratio, 1.78; 95% CI, 1.1-2.5; P=0.009

Cu

mu

lati

ve

Pro

bab

ilit

y o

f E

ve

nt

Months

Treatment

Interruption

Continuous Treatment

N=5472 HIV-infected patients with a CD4+ cell count >350mm3

Endpoint Hazard Ratio

(95%CI)* P Value

Death, any cause 1.8 (1.2-2.9) 0.007

Major cardiovascular, renal or hepatic disease

1.7 (1.1-2.5) 0.009

Fatal or non-fatal CVD

1.6 (1.0-2.5) 0.05

*Treatment Interruption vs. Continuous Treatment

SMART Study: HIV Viremia Can Contribute to CV Risk

Lundgren J, et al; 8th IAS, Vancouver, Canada, July 19-22, 2015; Abst. MOSY0301.

Higher impact of smoking than that of HIV infection

in mortality of HIV-infected Persons

Helleberg M et al. Clin Infect Dis 2014

Pro

port

ion s

urv

ivin

g

Cardiovascular risk: calculation

Framingham risk score

AHA/ACC pooled cohort equations CV risk

calculator (PCE)

Friis-Moller et al, Eur J Prev Cardiol 2016

DAD model

includes age, gender, BP, smoking, FH, DM, TC, HDL,

CD4, cumulative exposure to PIs and NRTIs, current

use of ABC

predicted risk more accurately than Framingham

Non-modifiable risk factors common to CKD and CVD

1. Booth GL, et al. Lancet 2006;368:29–36; 2. Mocroft A, et al. AIDS 2010;24:1667–78; 3. Hunt SC, et al. Am J Prev Med 2003;24:136–42;

4. NICE CKD Guidelines 2008. http://www.nice.org.uk/nicemedia/pdf/CG073NICEGuideline.pdf; 5. NICE CVD Guidelines 2010. http://www.nice.org.uk/nicemedia/live/13024/49273/49273.pdf

Non-modifiable risk factors CKD CVD

Advancing age1,2 + +

Gender1,2 + (female) + (male)

Family history3,4 + +

Ethnicity4,5 + (Black/Asian) + (Asian)

Major modifiable risk factors CKD CVD

Diabetes1,2 + +

Hypertension1,2 + +

Dyslipidaemia/abnormal lipids1,3 + +

Smoking1,4 + +

HIV disease5–7 + +

Metabolic syndrome8,9 + +

Hepatitis C virus infection2,10 + +

Obesity1,11 + +

Many modifiable risk factors are common to CKD and CVD

1. NICE CVD Guidelines 2010. Available at http://www.nice.org.uk/nicemedia/live/13024/49273/49273.pdf; 2. Mocroft A, et al. AIDS 2010;24:1667–78; 3. CKD in Adults: UK Guidelines http://www.renal.org/CKDguide/full/UKCKDfull.pdf;

4. NICE CKD Guidelines http://www.nice.org.uk/nicemedia/pdf/CG073NICEGuideline.pdf; 5. SMART Study Group. N Eng J Med 2006;355:2283–96; 6. Klein D, et al. CROI. 2011;Abstract 810;

7. Campbell LJ, et al. HIV Med 2009;10:329–36; 8. WHO CVD Guidelines. http://www.who.int/cardiovascular_diseases/guidelines/Full%20text.pdf; 9. Royal College of Physicians CKD Guidelines. http://www.nice.org.uk/nicemedia/live/12069/42116/42116.pdf;

10. Butt AA, et al. Clin Infect Dis 2009;49:225–32; 11. Ix JH & Sharma KJ. Am Soc Nephrol 2010;21:406–12

NA-ACCORD:

Contributions to MI Risk in HIV+ Subjects

Althoff K, et al. 24th CROI; Seattle, WA; February 13-16, 2017. Abst. 130.

Population attributable fractions have been adjusted for all the risk factors in the figure, as well as age, sex, race, HIV transmission risk, diabetes, and stage 4 chronic kidney disease.

Population attributable fractions and 95% confidence intervals for traditional and HIV-related factors, and hepatitis C virus infection, NA-ACCORD (1 Jan 2000 – 31 Dec 2013)

38% 41% 43%

2% 3%

10% 6%

2% 8%

0%

80%

Po

pu

lati

on

att

rib

uta

ble

tr

ac

tio

n (

PA

F)

Source: Haffner SM et al. NEJM 1998;339:229–234

Patients with DM but no CHD experience a similar rate of MI as patients

without DM but with CHD

Eve

nts

*/

10

0 p

ers

on

-ye

ars

Prior CHD

45 DM

No DM

No prior CHD

50

40

30

20

10

0

19 20

3.5

*Fatal or non-fatal MI

CHD=Coronary heart disease, DM=Diabetes mellitus, MI=Myocardial infarction

East-West Study

Diabetes Mellitus: Risk of Myocardial Infarction

CV Risk Modification in HIV

Petoumenos K for D:A:D . 20th CROI 2013.

Reduce TC 1 mmol/L

Stop smoking

CVD hazard in D:A:D*

Reduce sysBP 10 mmHg

*relative to 40 y.o. HIV+ male

NNT to Harm with MI: ABC in D:A:D

Risk Underlying 5

year risk (%) 5 year NNH

Underlying 10

year risk (%) 10 year NNH

Low (40 yo

non-smoker

with good

lipids and BP

0.1 1111 0.3 370

Smoker 0.4 277 1.5 92

Smoker and

DM 1.1 101 3.1 35

Smoker and

raised lipids 3.1 35 7.5 14

Previous CVD 5 22 10 14

Kowalska JD, et al . HIV Med. 2010 Mar;11(3):200-8. NNT, number needed to harm; MI, myocardial infarction; ABC, abacavir;

DM, diabetes mellitus; CVD, cardiovascular disease

Impact on CV risk of selected interventions

Intervention Type

HIV+ Patient Base Case Profile: 50 years

old, Male, On Abacavir, Smoker, w/

Hypertension, w/ Hyperlipidemia

1

Abacavir substitution with an alternative

antiviral without association to higher MI

rate [1][2][3]

2 Prescribing anti-hyperlipidemia

medication [4][8]

3 Prescribing anti-hypertensive medication

[4][7][8]

4

Counseling including standard treatment

for smoking cessation such as nicotine

patch and varenicline [4][5][6]

Hsue P, et al 692

D:A:D:

Cumulative Exposure to DRV/r Increase MI Risk

Ryom L, et al. 24th CROI; Seattle, WA; February 13-16, 2017. Abst. 128LB.

*Adjusted for gender, race, HIV exposure group, enrolment cohort, baseline date, prior CVD, nadir CD4 count, current CD4 count**, dyslipdaemia**, BMI**, diabetes**, eGFR**, age (all as fixed variates at baseline), HBV, HCV, smoking, family history of CVD, VL., hypertension, AIDS, cumulative exposure to darunavir/r, atazanavir/r, lopinavir/r & indinavir & recent exposure to abacavir (≤6 months) (all as time-updated variables).

**Factors considered to potentially lie on the causal pathway between PI/r exposure and CVD and values hence fixed at baseline.

Unadjusted CVD rate ratios per 5 years additional exposure: ATV/r 1.25 [1.10-1.43] and DRV/r 1.93 [1.63-2.28]

Adjusted* CVD rate ratios per 5 years additional exposure: ATV/r 1.03 [0.90-1.18] and DRV/r 1.59 [1.33-1.91]

Cumulative Years of Drug Exposure

Events

0 0-1 1-2 2-3 3-4 4-5 5-6 >6 0 0-1 1-2 2-3 3-4 4-5 5-6 >6

824 75 49 41 26 46 34 62 909 52 51 44 39 17 18 27

ATV/r DRV/r

20.0-

15.0-

10.0-

5.0-

4.0-

3.0-

2.0-

1.0-

ART classes and dyslipidaemia

NRTIs

TDF preferred (switching to TAF reduces lipid effect)

ABC associated with MI risk

NNRTIs

RPV, DOR preferred over EFV

PIs

ATV preferred even when boosted over LPV/rr

DRV/r associated with MI risk

InSTIs

Lipid neutral but no CV risk. Weight gain risk?

NEAT 022: Switch from PI/r- to DTG-based Regimen in Subjects with

Elevated CV Risk

Gatell J, et al. IAS 2017. Abstract 2461

• NEAT 022: randomised, open-label non-inferiority trial over 96 weeks in treatment-experienced

patients, designed to investigate lipid parameters in subjects with elevated CV risk*

– Changes in plasma lipids from BL to Week 48 were primary (TC) and secondary (other lipids) endpoints

– Co-primary endpoint: % of patients free of therapeutic failure at Week 48, in the ITT population

-8,7 -11,3

-18,4

-7,7

1,1

-7,0

0,7 0,5 4,2 2,0 2,5

0,4

-25,

-20,

-15,

-10,

-5,

0,

5,

TC Non-HDL-C

TG LDL-C HDL-C TC/HDLratio

p<0.001 p<0.001

p<0.001

p<0.001

p<0.001

P=0.286

• TC and other lipid

fractions (except HDL-

C) improved

significantly (p<0.001)

from BL with DTG.

• Virologic suppression

was maintained after

switch to DTG

Switching from a PI/r-based regimen to a DTG-based regimen improved

lipid profile versus continuing a P-based regimen in virologically

suppressed patients with elevated CV risk*

–7.0 –7.7 –8.7

–

11.3

–18.4

–5

–10

–5

–5

–15

Me

an

ch

an

ge

fro

m B

L t

o W

ee

k 4

8

(mm

ol/

L)

–20

–25

DTG† (n=205)

PI/r† (n=210)

Atorvastatin vs Pbo for Non-Ca++ Coronary Plaques

40 HIV-infected pts with subclinical coronary atherosclerosis and low density lipoprotein (LDL) cholesterol <130mg/dL

Coronary atherosclerotic plaque as assessed by coronary computed tomography angiography

Statin therapy was well-tolerated, with low incidence of clinical adverse events or laboratory abnormalities

-67,5

-45,

-22,5

0,

22,5

45,

NCPlaque % LDL-c mg/dl Lpa ng/dl

Changes in Non-Calcified Plaque and Key Lipids

Atorvastatin Placebo

Lo J, et al. 22nd CROI; Seattle, WA; February 23-26, 2015. Abst. 136.

p =.009 p =.0001 p =.0005

Beware of Drug Interactions

Boosters and statins esp simva-, fluva-statin

DRV induction and Atorva- and Rosuva-

Boosters and Ca Channel blockers e.g. amlodepine, diltiazem

Inducers (EFV>NVP) and statins, Ca Channels

DTG and metformin (+75%)

REPRIEVE: trial design (n=6500) http://reprievetrial.org/overview/

Screening and consent (asymptomatic HIV-infected persons with no history of CVD)

Randomisation

Mechanistic study primary (coronary plaque, vascular inflammation,

or immune activation)

Secondary end points (individual components of primary end point; all-cause death; incidence or progression

of non-calcified plaque; high-risk plaque; inflammatory, immunologic or metabolic biomarkers; predictors of statin effect; and non-AIDS-related comorbidities)

Composite primary end point (CVD-related death, myocardial infarction, unstable angina,

stroke, and arterial revascularization)

Placebo Pitavastatin 4mg daily

Management principles in HIV-infected persons

Same risk reduction strategies as general population

aspirin (?), statin, BP control, management of DM

Screen for DM and dyslipidaemia at baseline and every 3

to 6 and 6 to 12 months respectively

Use InSTI and avoid ABC in treatment naïve with

increased cardiovascular risk

Switch PI to InSTI has modest benefit on dyslipidaemia in

treatment experienced

atorvastatin, pitavastatin and rosuvastatin are statins of

choice for dyslipidaemia on ART