EXPLORE Study

1
HIVNET 015 The Explore Trial Susan
Buchbinder, MD Director, HIV Research Section San
Francisco Dept. of Public Health
2
(No Transcript)
3
(No Transcript)
4
Prevalence of HIV in US MSMNHBS MMWR
200554597-601
5
HIV Prevalence in South American CitiesS
Montano, JAIDS 2005 4057-64
6
Estimated number of adults and childrennewly
infected with HIV during 2004
Eastern Europe Central Asia 210 000 110 000
480 000
Western Central Europe 21 000 14 000 38 000
North America 44 000 16 000 120 000
East Asia 290 000 84 000 830 000
North Africa Middle East 92 000 34 000 350
000
Caribbean 53 000 27 000 140 000
South South-East Asia 890 000 480 000 2.0
million
Sub-Saharan Africa 3.1 million 2.7 3.8 million
Latin America 240 000 170 000 430 000
Oceania 5 000 2 100 13 000
Total 4.9 (4.3 6.4) million
7
Estimated US Cases HIV/AIDS by year of
diagnosis33 areas with name-based HIV infection
reporting
MMWR 2004531106-10
8
Background

• HIV epidemic throughout Americas (except parts of
  Caribbean) most concentrated in MSM
• HIV prevalence 10-30 HIV seroincidence 2-4
  despite ongoing risk reduction counseling
• Biomedical interventions (vaccines, PREP, STD rx,
  microbicides) being developed, but still years
  away
• Behavioral interventions needed both as
  stand-alone and to complement biomedical
  interventions

9
EXPLORE

• First behavioral intervention powered to address
  impact of intensive intervention on HIV
  seroincidence
• Cadillac version of behavioral intervention
  intended to maximize effects, likelihood for
  success
• Use of ACASI to get most accurate measures of
  risk, correlate change in risk with change in
  seroincidence

10
Study Design

• Multi-site RCT (randomized controlled trial)
  efficacy trial
• Eligibility (range of risk behaviors)
• Male, gt 16 yo
• Any anal sex with man in last year
• Not in mutually monogamous relationship gt 2 yrs
• Intervention
• 10 individualized sessions w/ boosters q 3 mos
• Control
• Project Respect risk reduction counseling q 6 mos

11
Statistical analyses

• Intent-to-treat analysis
• Primary outcome
• HIV infection
• Secondary outcomes behaviors
• unprotected anal (UA)
• unprotected anal with /unknown status partners
  (SDUA)
• unprotected receptive anal with /unknown status
  partners (SDURA)

12
Intervention modules
13
Measurements

• HIV antibody q 6 months
• Interviewer administered
• Demographics
• STD history
• ACASI (audio computer-assisted self interview)
• Sexual risk
• Drug use

14
Outcomes

• Primary endpoint HIV infection rates
• Phase IIB or screening efficacy trial
• If efficacy lt 10, discard or reformulate
• If 10-35, plausibly efficacious, more study
• If gt 35, efficacious implement
• Because Phase IIB are smaller than full efficacy
  trial, less precision in measure

15
Enrollment by site
16
Recruitment Sources
17
Baseline Demographics
18
Baseline Partners
19
Baseline Sexual Practices
20
Adherence to initial sessions

• No. of initial session-modules
  completing
• 0 1
• 1-3 12
• 4-6 5
• 7-9 7
• 10 75

21
Retention
22
Retention

• Group Final visit retention ()
• Race/ethnicity
• White 89.5
• Af-Am/Latino/API/NAm/oth 83.9
• Age (years)
• lt 25 80.0
• 26 89.8
• Female partners
• No 88.5
• Yes 74.2
• Unprotected anal
• No 89.0
• Yes 86.7
• Intervention sessions completed
• lt9 63.6
• 9 92.2

Plt0.05 for all comparisons
23
HIV SeroincidenceOverall seroincidence 2.1
(1.9, 2.4) per 100 py
1.00


Intervention
Control Efficacy 18.2 (-4.7,
36.0) Adj Efficacy 15.7 (-8.4, 34.4)
0.98
0.96
Percent free of HIV
0.94
0.92
0.90
Months
6
12
18
24
30
36
42
48
OR 0.67 0.61 0.83 1.17 0.73 1.32 0.75
1.05
24
Pre-set cutpoints for efficacy

• Efficacy 18.2 (95 CI -4.7 to 36)
• Adjusted efficacy 15.7 (95 CI -8.4 to 34.4)
• If lower bound 95 CI gt 10 declare efficacious
• Didnt meet this cutpoint
• If upper bound 95 CI lt 35 no substantial
  efficacy
• Meet cutpoint to say no efficacy?

25
Sexual behavior outcomes UA, SDUA SDURA
26
Time
Surrogate True Clinical Endpoint
Outcome
Disease
True Clinical Outcome
Disease
Surrogate Endpoint
27
Time
Intervention
Surrogate HIV Endpoint Infection
Risk
Intervention
HIV Infection
Risk
Surrogate Endpoint
28
Time
Intervention
Self reported HIV 6 mo. UA
effects Infection
Risk

• Potential Differences between
• self reported risk behaviors and true
  risk behaviors
• Alternative Pathways for risk of HIV infection
• Durability of effect

29
Conclusions

• First study of impact of behavioral intervention
  for MSM on HIV seroincidence
• Recruitment of large cohort, excellent retention
• Modest reduction in HIV seroincidence
• Can rule out substantial efficacy
• At cusp between discarding and pursuing further
• Significant reduction in self-reported risk
  behaviors
• Implications for using self-reported risk as
  endpoint in intervention trials?

30
Future directions

• Rationale for further analysis and modification
• Significant reduction in risk
• Possibility of early effects
• Likelihood that control condition exceeds usual
  care
• Precautions
• Problems retaining young, diverse, risky MSM
• Unwieldy intervention
• Plans for exploratory analyses
• Subgroup analyses
• Focus groups w/ men of color
• Many other analyses (HSV2, HHV8, risk factors for
  infection)

31
Explore Study Team

• Co-chairs
• Margaret Chesney
• Thomas Coates
• Beryl Koblin
• Site Principal Investigators
• Susan Buchbinder/Grant Colfax
• Connie Celum
• Frank Judson
• Beryl Koblin
• Ken Mayer
• David McKirnan