Sarah Hawkes, Paz Bailey G, Sternberg M, Lewis DA and Puren A

1
Acute HIV infections among men with genital ulcer
disease in South Africa

• Sarah Hawkes, Paz Bailey G, Sternberg M, Lewis DA
  and Puren A
• 1. London School of Hygiene and Tropical
  Medicine 2. NCHHSTP, CDC US 3. STI Reference
  Centre, NICD 4. Specialized Molecular Diagnostic
  Unit, NICD, South Africa

2
Rationale for study

• Genital ulcer disease (GUD) increases the risk of
  HIV acquisition and transmission
• Cohort studies and mathematical modeling show
  that the rate of HIV transmission is higher after
  acute HIV infection
• Most common method to diagnose acute HIV
  infection is documenting seroconversion
  (diagnosis in retrospect) or diagnosis in real
  time among HIV seronegative individuals using HIV
  nucleic acid amplification

3
Objectives of the trial

• To determine the prevalence of acute HIV among
  men with genital ulcer disease
• To evaluate risk factors for acute HIV infection

4
Methods

• Baseline and follow-up data from a double blind
  randomized placebo controlled trial of acyclovir
  400mgs TID for 5 days versus placebo
• Men with GUD presenting at three clinics in South
  Africa were enrolled after consent
• All patients received syndromic management for
  syphilis and chancroid
• Behavioral questionnaire was administered
• At enrolment tested for HSV2, HIV, syphilis, GUD
  aetiology

5
Methods (continued)

• HIV-negative men tested again for HIV at day 28
  to detect HIV seroconversions
• HIV tested for CD4, HIV RNA in plasma and
  ulcers, and HSV2 in ulcers
• HIV-1 antibody negative serum were tested by HIV
  RNA PCR using the COBAS AmpliScreen HIV-1 Test
  v.1.5 (lower limit of detection 100 copies/ml) to
  detect acute HIV
• Assay was used on initial pools of 6 samples, if
  positive individual samples tested

6
Definitions and analysis

• HIV-1 seroconverters HIV antibody negative at
  baseline and HIV antibody positive by two rapid
  tests at follow-up on day 28
• HIV prevalent 1) two concordant positive rapid
  antibody tests or 2) discordant rapid antibody
  tests with positive ELISA
• Acute HIV 1) Antibody rapid test was negative
  and HIV RNA results were positive or 2) two rapid
  tests discordant, ELISA negative and HIV RNA
  positive and 3) HIV seroconverters as described
  above
• HIV prevalent cases and HIV acute cases were
  compared to HIV negative participants to detect
  risk factors

7
Results

• 615 men enrolled in the trial
• 63 HIV positive
• 36.5 had been tested for HIV before and only
  20.7 HIV-positive knew their status
• No signs or symptoms of HIV seroconversion
• Of 228 HIV-negative men at baseline, there were
  8 seroconversions at day 28 and 20 HIV RNA
  positive specimens
• Prevalence of acute HIV overall 4.6 (95 CI 2.9
  6.2, 28/615)
• Prevalence of acute HIV among HIV-negative 12.2
  (95 CI 8.0 16.5, 28/228)

8
Study flowchart
615 men with GUD
228 negative by HIV rapid tests
387 positive by HIV rapid tests
20 HIV RNA positive
8 sero-conversions at day 28
200 HIV negative
387 prevalent HIV infections
28 acute HIV infections
9
Results-Bivariate Analysis
10
Acute and Prevalent HIV versus HIV
negative-Multinomial regression
Note Adjusted by age, prevalent HIV associated
with older age (plt0.001)
11
Conclusions

• High prevalence of acute HIV among men with GUD
  in South Africa
• The proportion of acute HIV was 4.6 among all
  study participants and 12.2 among persons
  without established HIV infection, higher to what
  has been reported for other groups
• HIV prevalent infections were associated with
  older age and HSV-2 seropositivity
• Acute HIV was associated with casual partners and
  ulcers of unknown etiology
• GUD patients represent an opportunity to identify
  acute infections and initiate interventions that
  may reduce the chances of HIV-1 transmission

12
Collaborators

• National Institute of Communicable Diseases in
  South Africa
• US Centers for Disease Control and Prevention
• London School of Hygiene and Tropical Medicine
• Reproductive Health and HIV Research Unit
• Special thanks to participating clinics and study
  participants