HIV and Aging: a Time for a New Paradigm Amy C. Justice, MD, MSCE, PhD Professor, Yale University Section Chief, General Internal Medicine VA Connecticut.

HIV and Aging: a

Time for a New Paradigm

Amy C. Justice, MD, MSCE, PhDProfessor, Yale University

Section Chief, General Internal MedicineVA Connecticut Healthcare System

Outline• Epidemiology, demography of aging with HIV• Describe Veterans Aging Cohort Study (VACS)

– HIV Associated Non AIDS (HANA) Conditions– VACS Risk Index

• A new approach to comparative effectiveness and personalized medicine

Antiretroviral Therapy in 2011• Once a day pill well tolerated and

achieves viral suppression in 84%*

• Median CD4 counts increasing

• Viral load declining

• AIDS defining events are rare

*Gallent JE. et al. Tenofavir DF, Emtricitabine, and Efavirenz vs. Zidovudine, Lamivudine, and Efavirenz for HIV. NEJM 2006 354:251-60.**McKinnell JA. et al ARV Prescribing Patterns in Treatment-Naïve Patients in the United States. AIDS Patient Care and STDs 2010 24:79-85

More People Living with HIV Infection Every Year (+38K/yr*)

CDC surveillance data Each year: 56K new infections-18K deaths=38K*

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

17%19%

21% 22%25%

27% 27% 29%33%

35%37%

39%41%

44%45%

47%50%

Projected Proportion of those Living With HIV in United States 50+ Years*

2001-2017

NY City

US VA in 2003

As of 2008:San Francisco

*Data from 2008, onward projected based on 2001-2007 trends (calculated by author), 2001-2007 data from CDC Surveillance Reports 2007. New York and San Francisco data from their Departments of Public Health

Projected

>50% of Deaths Attributed to Non-AIDS Events

Cumulative Mortality by COD Among Those on cART (1996-2006) ART-CC, CID 2010: 1387-1396

Prevalence and Incidence Are Linked

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 200840

45

50

55

60

44 4445

4647

48 48 48 4849

50 50

4546 46

4748

4950 50

5152

53 53

Median Age of Prevalent and Incident Veterans with HIV by Calendar Year

Incident Prevalent

AIDS Events Increasingly Rare

ART-CC, Archives Int Med 2005: 165 416-423

AIDS Events Variably Associated with CD4 and Survival

By Median (IQR) CD4 By Relative Hazard of Death

ART-CC, CID 2009;48:1138-51

Losina et al CID 2009

Life Expectancy is not “Normal”

Risk-adjusted HIV negative

Optimal care HIV postive

Mean age seroconversion of 33 years

HIV Epidemiology & Field Services Semiannual Report, NYCDOH. April 2010

Death Rate Disparities by HIV, Race/Ethnicity and Age

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 20070

50

100

150

200

250

300

350

400

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

40% 41% 39%41% 42%

44%42%

45% 46% 47% 48%

30%32% 34%

36% 36%38%

41% 39% 41% 39% 39%

269277 275

284293

313296

312323

333 336

203211

246261

234

272 274261

272 273 266

<50 years ≥50 years <50 years ≥50 years

Year

Med

ian

CD4

coun

t (ce

lls/m

m3)

at fi

rst p

rese

ntati

on fo

r HIV

clin

ical c

are

Prop

ortio

n of

pati

ents

with

a C

D4 co

unt ≥

350

cells

/mm

3 at

firs

t pre

sent

a-tio

n fo

r HIV

clin

ical c

are

Delayed Presentation By Age (NA ACCORD)

Altoff K. et al. In press JAIDS

Major Observations

• On ART, HIV is a complex chronic disease, not unlike insulin dependent diabetes or cancer in partial remission

• Annual new HIV infections exceed deaths; the population on ART is rapidly growing and aging

• We need an effective and efficient approach to caring for these individuals

VACS Long Term Objectives

1. Fully characterize treated HIV infection as a model of complex chronic disease with a dominant index condition

2. Use this model, risk stratification, and electronic medical records systems to revolutionize health care

VACS 8• SUBJECTS: 3,640 HIV infected; 3,640 uninfected

– Group matched: age, race/ethnicity, and site

• SITES: Manhattan, Bronx, Washington DC, Baltimore, Pittsburgh, Atlanta, Houston, Los Angeles

• BASELINE: 2002 (8 years)

VACS Virtual Cohort

• Subjects– 40,594 HIV infected Veterans– 81,188 Age, Race, Region Matched 2:1

• Scope– 1998 to present– Baseline

• HIV infected patients at initiation of HIV care• Controls selected and followed in same year

Arbitrated Clinical Events in VACS 8

• ART Initiation (Complete, paper in process)• Symptomatic Cirrhosis (Decompensated Liver

Disease—paper in press)• Major Cancers (Nearly Complete)• Myocardial Infarction (Underway)• Stroke (Planned)• COPD and Pneumonia (Planned)

HIV Associated Non AIDS Conditions (HANA)

*More AIDS and “Non-AIDS” Events Among Rx. Sparing Arm (HR 1.7 in SMART) NEJM 2006;355:2283-96

Rx. Sparing

Rx.Intensive

Total

All Cause Death 55 30 85

Serious OI 13 2 15

Nonserious OI 63 18 81

Major CAD, Renal, or Liver Disease

65 39 104

Non AIDS Events Are Associated with HIV Disease Progression*

Definition: HIV Associated Non AIDS Conditions (HANA)

• After adjustment for established risk factors, association with HIV remains– Compare to demographically and behaviorally similar

uninfected controls– Weaker (<2 fold) associations may be due to

inadequate adjustment for risk factors

• May be due to HIV, ART or both

HIV status and the risk of AMI

Risk Factors HR for AMI with 95% CI

HIV infection 1.94 (1.58-2.37) Age (10 yrs) 1.39 (1.26-1.54) Race/ethnicity African American Hispanic Other

0.79 (0.64-0.98) 1.39 (1.03-1.88) 0.42 (0.22-0.80)

Hypertension 1.36 (1.10-1.68) Diabetes 2.01 (1.68-2.53) Hypercholesterolemia 1.30 (1.06-1.59) Ever Smoking 1.87 (1.38-2.52) HCV infection 1.10 (0.88-1.38) EGFR<30 ml/min/1.73m2 4.93 (3.12-7.77) BMI ≥ 30kg/m2 0.90 (0.72-1.12) History of cocaine abuse or dependence

1.42 (0.97-2.09)

History of alcohol abuse or dependence

0.80 (0.56-1.11)

Freiberg M.S. et al. HIV is Associated with Clinically Confirmed MI. CROI 2011 Abstract# W-176

Fragility Fractures HIV+/- (n= 125,259)

HIV Model Full Model HIV+ Men

HIV 1.32 (1.20, 1.47) 1.10 (0.97, 1.25) --

Age (10 yr increments) -- 1.32 (1.25, 1.40) 1.52 (1.39, 1.66)

White race -- 1.80 (1.60, 2.03) 1.85 (1.52, 2.25)

Alcohol abuse -- 1.80 (1.50, 2.17) 1.50 (1.12, 2.02)

Liver disease -- 1.38 (1.10, 1.73) 1.39 (1.03, 1.87)

Smoker -- 1.21 (1.04, 1.42) 1.30 (1.00, 1.67)

Any PPI use -- 1.70 (1.51, 1.92) 1.55 (1.28, 1.89)

BMI -- 0.82 (0.79, 0.85) 0.87 (0.77 0.99)

BMI2 -- 1.002 (1.002, 1.003) 1.00 (1.00, 1.00)

Current corticosteroid use -- 1.45 (1.21, 1.74) 1.41 (1.06, 1.88)

CD4/100 cells/mm3 -- -- 1.01 (0.98, 1.05)

Current TDF use -- -- 1.29 (0.99, 1.70)

Current PI use -- -- 1.41 (1.16, 1.70)

Womack J. et al. PLoS ONE February 2011 | Volume 6 | Issue 2 | e17217

Possible HANA• Targeted Disease

– Vascular: Myocardial Infarction, Thrombosis, and Stroke

– Bone: Osteoporosis and Avascular Necrosis

– Cancer: “infectious” e.g. Anal and “non infectious” e.g. Lung

– Lung: pneumonia and COPD

– Neurological: Peripheral neuropathy, ?dementia

• General Organ Injury– Liver Fibrosis: risk of, progression to, cirrhosis and hepatoma

– Hematologic Disease: anemia, thrombocytopenia

– Decreased Renal Function: most is not HIVAN

General Observations on HANA• Multiple interacting HIV and non HIV causes

– HIV typically not the most influential risk factor

• Incidence of event different from relative risk

• Adjusted relative risk HIV+/- highly variable– Association with CD4 variable– Degree to which these occur “prematurely”

difficult to quantify– Competing risk of death is changing and

unmasking risk associated with HIV

Warning!

• All these conditions have multiple, interacting, causes among HIV+/-

• The mix of causes driving these events among HIV+ may differ from HIV-

• Until we understand this mix, we must focus on what drives health outcomes in our patients

Veterans Aging Cohort Study Risk Index (VACS Index)

Justice, AC. et. al, HIV Med. 2010 Feb;11(2):143-51. Epub 2009 Sep 14.

An index composed of routinely collected laboratory values that accurately predicts all cause mortality among those with HIV infection

Rationale for Multivariable Risk Index

• A single, summary measure of disease

• Identifies important thresholds for lab tests

• Resolves conflicting results

• Informs prioritization

• Has major statistical advantages– Decreased measurement error– Each person has a measurable outcome at any time point

Justice AC. HIV and aging: time for a new paradigm. Curr HIV/AIDS Rep. 2010 May;7(2):69-76.

Veterans Aging Cohort Study Risk Index (VACS Index)

• Composed of age and laboratory tests currently recommended for clinical management

– HIV Biomarkers: HIV-1 RNA and CD4 Count

– “non HIV Biomarkers”: Hemoglobin, hepatitis C, composite markers for liver and renal injury

• Developed in US veterans, validated in Europe and North America

32

Composite Biomarkers

32

AGE * AST PLT * sqrt(ALT )

FIB 4 =

eGFR = 186.3 * CREAT -1.154 * AGE -0.203 * FEM_VAL * BLACK_VAL

FEM_VAL = 0.742 if female, 1 if male

BLACK_VAL = 1.21 if black, 1 otherwise

Index ScoreRestricted VACS

Age (years) <50 0 050 to 64 23 12> 65 44 27

CD4 > 500 0 0cells/mm3 350 to 499 10 6

200 to 349 10 6100 to 199 19 1050 to 99 40 28< 50 46 29

HIV-1 RNA < 500 0 0copies/ml 500 to 1x105 11 7

> 1x105 25 14

Hemoglobin > 14 0g/dL 12 to 13.9 10

10 to 11.9 22< 10 38

FIB-4 < 1.45 01.45 to 3.25 6> 3.25 25

eGFR mL/min > 60 045 to 59.9 630 to 44.9 8< 30 26

Hepatitis C Infection 5

Age

HIV SpecificBiomarkers

Biomarkers of General Organ System Injury

Tate J. et al. IDSA 2010 Vancouver, BC October 21-24th. Poster 1136

VACS Index Thresholds and Weights

Age

PLT ALT AST 20 30 40 50 60 70

100 35 35 1.18 1.77 2.37 2.96 3.55 4.14

  44 1.49 2.23 2.97 3.72 4.46 5.21

  53 1.79 2.69 3.58 4.48 5.38 6.27

  44 35 1.06 1.58 2.11 2.64 3.17 3.69

  44 1.33 1.99 2.65 3.32 3.98 4.64

  53 1.60 2.40 3.20 4.00 4.79 5.59

  53 35 0.96 1.44 1.92 2.40 2.88 3.37

  44 1.21 1.81 2.42 3.02 3.63 4.23

    53 1.46 2.18 2.91 3.64 4.37 5.10

FIB 4 Values by Age, ALT, and AST(Platelets 100k)

FIB 4 >3.25 is worth 25 points, 1.45-3.25 is worth 6 points

35

0%

20%

40%

60%

80%

100%

0 20 40 60 80 100Risk Score

Mo

rta

lity

Justice AC. HIV and Aging: Time for a New Paradigm. Curr HIV/AIDS Rep. 2010 May;7(2):69-76

y = 0.0091x - 0.0318

R2 = 0.9916

0%

20%

40%

60%

80%

100%

0 20 40 60 80 100Risk Score

Mo

rta

lity

Justice, AC. et. al, HIV Med. 2010 Feb;11(2):143-51. Epub 2009 Sep 14.

VACS Index Highly Predictive of Long Term (5 Year) All Cause Mortality

Individual Scores

Aggregated Scores

Discrimination of VACS vs. Restricted Index

Justice AC. et al. A Prognostic Index for those Aging with HIV. CROI 2011 Poster # 793

Subgroup VACS IndexC-stat

Restricted IndexC-stat

p-value**

Overall 0.80 0.75 <0.0001MaleFemale

0.810.81

0.750.77

<0.001<0.001

WhiteBlackHispanic

0.790.810.90

0.740.760.78

<0.001<0.001<0.001

Age<50>= 50

0.810.74

0.750.69

<0.001<0.0001

HIV-1 RNA<500>=500

0.770.78

0.680.74

<0.0001<0.0001

Calibration of VACS vs. Restricted Index (5 Year Mortality)

Justice AC. et al. A Prognostic Index for those Aging with HIV. CROI 2011 Poster # 793

The Holy Grail: Surrogate Endpoint

• Must be an accurate predictor of target outcome

• Respond to changes in risk of the outcome due to treatment

• Detect differences in outcome due to treatment among different treatment arms

VACS Index Response to 1st Year of cART (+/- 80% adherence)

39Solid lines indicate >80% adherence

VACS Index Correlated with Biomarkers of Inflammation

Justice AC et al,“Biomarkers of Inflammation, Coagulation, and Monocyte Activation are Strongly Associated with the VACS Index among Veterans on cART” CROI 2011 Poster # 796

eGFR

Age

HIV-1 RNA

Hemoglobin

FIB-4

CD4 count

Rest. index

VACS index

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

IL-6sCD14d-Dimer

VACS Index Summary

• Is associated with markers of inflammation

• Accurately predicts mortality among HIV patients in the US and Europe

• Responds to changes in risk associated with ART initiation

• Will likely prove a more reliable surrogate endpoint than any single biomarker

Why Is This Important?

• Uses lab tests currently part of routine care

• Identifies modifiable risk at earlier thresholds

• Incorporates age, and effects of HANA and toxicity

• Computation easy, can be included in lab reports and available through websites/apps

• Offers approach to personalizing and prioritizing care that goes beyond CD4 count and HIV-1 RNA

Example: Framingham Index

• Assigns points based on 6 factors (5 modifiable)

• Estimates risk of MI or death over the next 5-10 years ranging from 1% to >56%

• Assumes that change in risk due to smoking cessation is same as never having smoked, etc.

43D’Agostino RB. Et al. Validation of the Framingham Coronary Heart Disease Prediction Scores: Results of a Multiple Ethnic Groups Investigation. JAMA 2001;286:180-187

Framingham Risk Assessment

http://hp2010.nhlbihin.net/atpiii/calculator.asp?usertype=prof

Risk score results:

Age: 60

Gender: male

Total Cholesterol: 280 mg/dL

HDL Cholesterol: 100 mg/dL

Smoker: Yes

Systolic Blood Pressure: 120 mm/Hg

On medication for HBP: No Risk Score* 10%

* The risk score shown was derived on the basis of an equation. Other NCEP materials, such as ATP III print products, use a point-based system to calculate a risk score that approximates the equation-based one.

To interpret the risk score and for specific information about CHD risk assessment as part of detection, evaluation, and treatment of high blood cholesterol, see ATP III Executive Summary and ATP III At-a-Glance.

Results View:

Uses of Framingham Index

• Assesses relative importance of CHD risk for individual patients

• Quantifies absolute level of CHD risk for individual patients

• Allows clinicians and patients to match the level of treatment to the level of risk

• CHD guidelines are based on these estimates

D’Agostino RB. Et al. Validation of the Framingham Coronary Heart Disease Prediction Scores: Results of a Multiple Ethnic Groups Investigation. JAMA 2001;286:180-187

Case Example50 year old, HIV infected male on ART with an HIV-1 RNA<500, CD4 count 500, normal hemoglobin, creatinine, AST, ALT, and platelets. HCV negative. score 8; expected mortality* 4%

–CD4 count 400 cells/mm3, score 18; expected mortality* 9% –Hemoglobin 12-13.9 g/dL, score 28; expected mortality* 15% –Hemoglobin 10-11.9 g/dL, score 40; expected mortality* 24%

*In all cases referring to estimated 5 year mortality risk.

Risk and Revolution of Care

Comprehensive Observational Data

Finely Grained Risk Assessment for Major

Outcomes

Identification of Modifiable Risk Factors

Link to Evidence Based Treatments through

Integrated Decision Support with Point and Click Action

RCTs of the Strategy of Care Tailored to Risk and Using

Change in Risk as Outcome

48

In Development: Interpretation

Your score is XX. Among 100 veterans in VA care with HIV infection with this score, we would expect that YY would be alive at five years and ZZ would have died. The figures in grey represent those expected to live 5 years and the figures in black represent those expected to have died.

49

Counseling (Hypothetical)

• Based on your drinking pattern and use of tobacco, you could reduce your 5 year risk of mortality to XX if you stopped both

• If you stop smoking, your risk will be YY and if you stop drinking your risk will be XX

• Websites where you can learn more about – How to stop drinking include XX– How to stop smoking include XX

• If you would like to help us improve this site click here50

Examples of Advice: Liver Disease

Because you appear to have liver injury and have HCV infection, there are a number of things you can do to reduce you VACS Index Score…

• Review all your medications with your provider to identify any potentially liver toxic medications

• Cut down or abstain from alcohol• Make sure not to skip doses of your ARVs• Talk to your provider about taking medications to

treat you HCV infection

Future Work

• Informatics: Develop information tool that calculates index, counsels on risk, identifies modifiable risk, and suggests patient action

• Observational Analyses: estimate likely effect size for potential interventions: eg, alcohol cessation, HCV treatment, adherence, etc.

• RCT: strategy trial among those with abnormal FIB 4 who drink alcohol

• PI and Co-PI: AC Justice, DA Fiellin

• Scientific Officer (NIAAA): K Bryant

• Participating VA Medical Centers: Atlanta (D. Rimland), Baltimore (KA Oursler, R Titanji), Bronx (S Brown, S Garrison), Houston (M Rodriguez-Barradas, N Masozera), Los Angeles (M Goetz, D Leaf), Manhattan-Brooklyn (M Simberkoff, D Blumenthal, H Leaf, J Leung), Pittsburgh (A Butt, E Hoffman), and Washington DC (C Gibert, R Peck)

• Core Faculty: K Akgun, S Braithwaite, C Brandt, K Bryant, R Cook, K Crothers, J Chang, S Crystal, N Day, R Dubrow, M Duggal, J Erdos, M Freiberg, M Gaziano, M Gerschenson, A Gordon, J Goulet, N Kim, M Kozal, K Kraemer, V LoRe, S Maisto, K Mattocks, P Miller, P O’Connor, C Parikh, C Rinaldo, J Samet

• Staff: H Bathulapalli, T Bohan, D Cohen, A Consorte, P Cunningham, A Dinh, C Frank, K Gordon, J Huston, F Kidwai, F Levin, K McGinnis, L Park, C Rogina, J Rogers, L Sacchetti, M Skanderson, J Tate, E Williams

• Major Collaborators: VA Public Health Strategic Healthcare Group, VA Pharmacy Benefits Management, Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC), Yale Center for Interdisciplinary Research on AIDS (CIRA), Center for Health Equity Research and Promotion (CHERP), ART-CC, NA-ACCORD, HIV-Causal

• Major Funding by: National Institutes of Health: NIAAA (U10-AA13566), NIA (R01-AG029154), NHLBI (R01-HL095136; R01-HL090342; RCI-HL100347) , NIAID (U01-A1069918), NIMH (P30-MH062294), and the Veterans Health Administration Office of Research and Development (VA REA 08-266) and Office of Academic Affiliations (Medical Informatics Fellowship).

Veterans Aging Cohort Study

National VACS Project Team 2010