SUMMARY
Background
HIV incidence among young women in sub-Saharan Africa remains high and their inclusion in vaccine and cure efforts is crucial. We aimed to establish a cohort of young women detected during Fiebig stage I acute HIV infection in whom treatment was initiated immediately after diagnosis to advance research in this high-risk group.
Methods
945 women 18–23 years in KwaZulu-Natal, South Africa, who were HIV uninfected and sexually active consented to HIV-1 RNA testing twice a week and biological sampling and risk assessment every 3 months during participation in a 48–96 week life-skills and job-readiness programme. We analysed the effect of immediate combination antiretroviral therapy (ART) on viraemia and immune responses, sexual risk behaviour, and the effect of the socioeconomic intervention.
Findings
42 women were diagnosed with acute HIV infection between Dec 1, 2012, and June 30, 2016, (incidence 8·2 per 100 person-years, 95% CI 5·9–11·1), of whom 36 (86%) were diagnosed in Fiebig stage I infection with a median initial viral load of 2·97 log10 copies per mL (IQR 2·42–3·85). 23 of these 36 women started ART at a median of 1 day (1–1) after detection, which limited the median peak viral load to 4·22 log10 copies per mL (3·27–4·83) and the CD4 nadir to 685 cells per μL (561–802). ART also suppressed viral load (to <20 copies per mL) within a median of 16 days (12–26) and, in 20 (87%) of 23 women, prevented seroconversion, as shown with western blotting. 385 women completed the 48 week socioeconomic intervention, of whom 231 were followed up for 1 year. 202 (87%) of these 231 women were placed in jobs, returned to school, or started a business.
Interpretation
Frequent HIV screening combined with a socioeconomic intervention facilitated sampling and risk assessment before and after infection. In addition to detection of acute infection and immediate treatment, we established a cohort optimised for prevention and cure research.
Funding
Bill & Melinda Gates Foundation, National Institute of Allergy and Infectious Diseases, International AIDS Vaccine Initiative, Wellcome Trust, Howard Hughes Medical Institute.
INTRODUCTION
Despite advances in prevention methods and expanded access to combination antiretroviral therapy (ART), HIV incidence remains high in key populations, particularly young women in sub-Saharan Africa.1 Therefore, development of effective interventions targeting this risk group is imperative.2 Improved understanding of the behavioural factors and biological mechanisms underlying increased HIV acquisition risk in young women is important for prevention strategies.3 Furthermore, because of inherent adverse effects of lifelong therapy, and persistent immune activation, an HIV cure is needed 4–7 Investigation of the earliest immunological and virological events after HIV infection is essential to vaccine and cure research. ART initiation during the earliest stages of acute HIV infection can reduce transmission risk, viral reservoir size, and T-cell activation and, in some individuals, lead to long-term viral remission after withdrawal of treatment.8–11 However, detection and treatment during Fiebig stage I infection, when viremia first becomes detectable, is challenging.
One approach to the detection of acute HIV infection during Fiebig stage I is to test uninfected individuals who are at high risk of infection, which requires a cohort capable of adhering to a frequent surveillance schedule. In South Africa, gender inequalities and poor access to education and economic opportunities place young women at disproportionately high risk of HIV infection.12,13 In the South African province KwaZulu-Natal, HIV incidence in young women is around 10·0 per 100 person years.14 Surveillance testing in this group could enable detection during acute HIV infection, even if combined with interventions to reduce HIV incidence. Collection of blood and mucosal tissue before infection would permit identification of factors associated with acquisition and early interactions between host and virus that determine disease progression. Treatment initiated immediately after detection would be expected to limit viremia and establishment of viral reservoirs. Moreover, because research on acute HIV infection tends to be done mainly in men who have sex with men, serodiscordant couples, and blood donors,15 inclusion of women, who bear a disproportinate burden of HIV globally, is vitally important.
With the aim of detecting acute HIV infection during Fiebig stage I, we established the Females Rising through Education, Support, and Health (FRESH) study in KwaZulu-Natal to enrol young, HIV-uninfected women at risk of infection. We hypothesised that HIV RNA testing twice a week in a high-risk cohort, coupled with frequent, longitudinal collection of biological specimens before and immediately after infection, would provide the opportunity for HIV prevention and cure research in this at-risk population. To encourage participant enrolment and adherence to the protocol, we designed a socioeconomic intervention, scheduled to coincide with study visits, which delivered tangible and sustained benefit to participants.
METHODS
Study design and participants
FRESH is an ongoing prospective study in KwaZulu-Natal, South Africa, designed to diagnose acute HIV infection in young at-risk women, obtain blood and mucosal samples before and after infection, and enable study of HIV pathogenesis and biological and behavioural risk factors for HIV acquisition, as well as research on vaccine and cure strategies. Eligible women were HIV uninfected, aged 18–23 years, sexually active, not pregnant, non-anemic (haemoglobin ≥10 g/L), without other barriers to participation (serious chronic illness, enrolment in another study, or family responsibilities), and gave written consent to enrolment. We targeted disadvantaged, at-risk women, favouring the enrolment of those who were unemployed and not attending school. Participants were recruited at local sites frequented by young people, including cafes, nightclubs, and shopping malls. To promote peer support, women were enrolled in groups of 30–35 participants, forming cohorts whose visits were scheduled together througout the surveillance period. Surveillance duration was initially 96 weeks, shortened to 48 weeks on May 1, 2015, 29 months after study launch. HIV-1 RNA testing during surveillance was done by finger prick twice a week and was modeled on a previously described study.16 Unique to the FRESH study design was the incorporation of a socioeconomic intervention developed to address challenges affecting young women that might contribute to HIV acquisition risk.17,18 The intervention’s twice-weekly classes, done at the study site, coincided with the HIV surveillance schedule. The study was approved by the biomedical research ethics committee of the University of KwaZulu-Natal and the institutional review board of Massachusetts General Hospital.
Procedures
Surveillance before infection included collection of peripheral blood and specimens from the female genital tract every 3 months, as previously described.3,19 After detection of acute HIV infection, weekly prospective blood and female genital tract sampling was done. Testing for sexually transmitted infections was done with GeneXpert (Cepheid, Sunnyvale, CA, USA) for diagnosis of Neisseria gonorrhoea and Chlamydia trachomatis and with real-time PCR (Roche, Basel, Switzerland) for diagnosis of Trichomonas vaginalis, HSV-1, and HSV-2.3,19 Participants with positive results for sexually transmitted infections were referred for treatment.
We designed the socioeconomic programme to coincide with the frequency and duration of HIV surveillance and to address—through provision of life-skills and job-skills training needed to enter and succeed in the workforce—challenges common to young women living in poverty. Topics covered included women’s health, HIV prevention and treatment, gender-based violence, relationships, stress management, self-esteem, and communication. The curriculum also included sections on career exploration, starting a small business, resumé and cover letter preparation, workplace etiquette, how to dress and prepare for an interview, basic computer training, and field trips to local businesses. Participants interested in returning to school, either to complete high school or to enrol at a tertiary institution, were assisted accordingly. Placement in work, starting a business, or enrolment in school was assessed routinely during the 12 months after completion of the curriculum.
Surveillance for acute HIV infection included HIV-1 RNA testing (NucliSens EasyQ v2.0 assay; bioMérieux, Marcy l’Etoile, Switzerland) twice a week, done with 500 μL blood samples obtained by finger prick.16 Results were available within 24 h, prompting confirmatory testing, blood and female genital tract sampling, and ART initation. Staging by Fiebig criteria20 was based on the presence or absence of plasma HIV RNA and p24 antigen (p24Ag; Roche) and the results of a fourth-generation HIV EIA (HIV Cobas Combi; Roche) and western blotting (Bio-Rad GS HIV-1 Western Blot kit; Bio-Rad Laboratories, Redmond, WA, USA): Fiebig stage 1 (RNA positive, p24Ag negative, EIA negative, western blot negative); Fiebig stage II (RNA positive, p24Ag positive, EIA positive or negative, western blot negative); Fiebig stage III (RNA positive, p24Ag positive or negative, EIA positive, western blot negative or indeterminate); Fiebig stage IV (RNA positive, p24Ag positive or negative, EIA positive, western blot indeterminate or positive); and Fiebig stage V (RNA positive, p24Ag positive or negative, EIA positive, western blot positive without p31 band). Western blot interpretation followed Centers for Disease Control and Prevention (CDC) criteria: a positive result had at least two of the major brands (gp160 or gp120 and gp41 and p24).19 When only one or two protein bands were observed, without meeting the CDC criteria, or when the band intensity was weaker than the positive control included in the assay, the result was interpreted as indeterminate. Negative results exhibited no reactive bands.21 Rapid immunoassays (Determine; Alere, Waltham, MA, USA; Unigold; Trinity Biotech. Wicklow, Ireland) were also done. CD4 cell counts were enumerated using Tru-Count technology (Becton Dickinson, San Jose, CA, USA) and analysed on a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, USA). Viral Load and CD4 cell counts were monitored during the 156 week follow-up (appendix pp 1–2). We defined acute HIV infection as a new detection of HIV RNA, positive on repeat testing, and with an evolving HIV western-blot pattern (before development of a p31 band).
During the first 19 months of the study (Dec 1, 2012, to June 30, 2014), participants diagnosed with acute HIV infection were not immediately treated; they were monitored and referred for treatment when they met national eligibility criteria. During this period, the eligibility threshold in South Africa was changed from 350 cells per μL or fewer to 500 cells per μL. In July, 2014, we introduced immediate initiation of ART after detection of acute HIV infection, using a three-drug oral regimen of 300 mg tenofovir disoproxil fumarate, 200 mg emtricitabine, and 600 mg efavirenz. In July, 2015, raltegravir (400 mg twice a day) was introduced as a fourth drug and was continued for 90 days after viral suppression (<20 copies per mL).
Statistical analysis
Sample size was established on the basis of the 10% HIV infection rate in the region and an 80% retention rate. Participants who had available data (non-missing values) for a particular variable were included when comparing the variable between the study groups.
We used descriptive measures to summarise data. Continuous variables were summarised with median (IQR) or mean (SD); categorical variables were summarised with frequency (%). Fisher’s exact test was used to compare categorical variables. We used the Mann-Whitney U test, Wilcoxon signed-rank test, and Kruskal-Wallis test with Dunn’s post-hoc analyses to compare continuous variables. Spearman’s rank correlation coefficient was used to examine bivariate associations between study variables. All p values are two-sided and a p value of less than 0.05 was considered significant. Statistical analyses and graphing were done using GraphPad Prism 5·0 and R package.
Role of the funding source
The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
RESULTS
Between Dec 1, 2012, and June 30, 2016, 1290 women aged 18–23 years were screened for HIV-1 infection, 945 (73%) of whom were eligible and enrolled (figure 1; table 1). Median age was 21 years. Most women had regular or steady partners; however, cohabitation was uncommon (table 1). At enrolment, no participants were employed or attending school full time.
Figure 1. Study profile.
ART=combination antiretroviral therapy. *Early ART was defined as treatment initiated during acute HIV infection, immediately after detection of HIV RNA.
Table 1.
Demographic, behavioural and clinical information
| Baseline | All participants | Uninfected | HIV Infected (AHI) | p value of Uninfected vs. AHI |
|---|---|---|---|---|
| General | ||||
| Median Age (IQR) | 21 (20–22) | 21 (20–22) | 21 (20–22) | 0·8072 |
| Did not graduate from high school | 375/945 (40%) | 360/903 (40%) | 15/42 (36%) | 0·4256 |
| Unemployed (not working full-time) | 945/945 (100%) | 903/903 (100%) | 42/42 (100%) | 1·000 |
| Median age at sexual debut (IQR) | 18 (17–19) | 18 (17–19) | 17 (17–18) | 0·1885 |
| Pregnancies, prior to enrollment | 0·0546 | |||
| 0 pregancies | 377/945 (40%) | 363/903 (40%) | 14/42 (33%) | ·· |
| 1 pregnancy | 405/945 (43%) | 386/903 (43%) | 19/42 (45%) | ·· |
| 2 pregnancies | 78/945 (8%) | 69/903 (8%) | 9/42 (21%) | ·· |
| 3 pregnancies | 8/945 (<1%) | 8/903 (1%) | 0/42 (0%) | ·· |
| Not applicable | 77/945 (8%) | 77/903 (9%) | 0/42 (0%) | ·· |
| Sexual Partners | ||||
| Median no. of sexual partners, lifetime (IQR) | 2 (1–3) | 2 (1–3) | 2 (2–4) | 0·0047 |
| Participants with a “regular/steady” partner | 909/945 (96%) | 869/903 (96%) | 40/42 (95%) | 0·6549 |
| Participants living with “regular/steady” partner | 33/945 (3%) | 30/903 (3%) | 3/42 (7%) | 0·1823 |
| Median age of current/most recent partner (IQR | 24 (22–26) | 23.5 (22–26) | 24.5 (22–27) | 0·2964 |
| Sexual Activity Type (last 30 days) | ||||
| Participants having receptive vaginal intercours | 640/945 (68%) | 610/903 (68%) | 30/42 (71%) | 0·8478 |
| Participants having receptive anal intercourse | 29/945 (3%) | 28/903 (3%) | 1/42 (2%) | 1·000 |
| Participants having receptive oral sex | 86/945 (9%) | 82/903 (9%) | 4/42 (10%) | 1·000 |
| Condom Use | 0·7415 | |||
| Always | 223/945 (24%) | 211/903 (23%) | 12/42 (29%) | ·· |
| Sometimes | 429/945 (45%) | 411/903 (46%) | 18/42 (43%) | ·· |
| Never | 139/945 (15%) | 132/903 (15%) | 7/42 (17%) | ·· |
| No answer | 154/945 (16%) | 149/903 (17%) | 5/42 (12%) | ·· |
| 3-month Survey & Testing | All participants | Uninfected | HIV Infected (AHI) | p value ofUninfected vs. AHI |
| Recent sexual activity (last 30 days) | ||||
| Participants sexually active | 509/693 (73.4%) | 469/651 (72%) | 40/42 (95%) | <0·0001 |
| Median age of oldest partner (IQR) | 25 (23–28) | 25 (23–27) | 27 (25–29) | <0·0001 |
Data are n (%) or median (IQR).
The total number of participants at enrolment was 945, which decreased to 693 (651 in the uninfected group and 42 in the HIV-infected group) by the first 3 month survey and sampling timepoint because of study dropout.
42 (4%) participants had acute HIV infection after enrolment, reflecting an incidence of 8.2 per 100 person-years (95% CI 5.9–11.1). The median duration of monitoring until detection was 134 days (range 0–684) for the first 29 months of the study (Dec 1, 2012, to April 30, 2015) and 48 weeks for the subsequent 14 months (May 1, 2015, to June 30, 2016). The surveillance period was shortened after observing that 21 (95%) of the first 22 acute infections were detected within 48 weeks of enrolment (appendix p 3.)
HIV acquistion was not associated with age of sexual debut, partner age, type of sexual activity, frequency of condom use (table 1), or prevalence of an active sexually transmitted infection.19 HIV acquisition was associated with the number of lifetime sexual partners (p=0.0047). Engagement in sexual activity during the 30 days before the 3 month risk survey was higher among infected participants than among uninfected participants (95% vs 72%; p<0·0001; table 1).
40 (95%) of the 42 women diagnosed with acute HIV infection are still in follow-up (figure 1). 625 (66%) of the 945 women enrolled in the study completed pre-infection surveillance. 320 uninfected women withdrew during this period (figure 1). During pre-infection surveillance, adherence to the 3 monthly sampling scheduled was 98% for blood sampling and 85% for vaginal mucosal sampling.
The first 14 participants to be diagnosed with HIV infection were untreated during acute infection, 13 (93%) of whom were diagnosed during Fiebig stage I infection (table 2). The median viral load at detection for these 13 individuals was 3·29 log10 copies per mL (IQR 2·954·57), reaching a peak of 7·11 log10 copies per mL (6·98–7·48) within 8 days (7–11). The median viral load set-point was 4·68 log10 copies per mL (4·65–5·18; appendix p 4). Set-point was defined as the mean viral load from all samples collected before initiation of ART between 42 days and 330 days after infection.
Table 2.
Detection, staging, and treatment of acute HIV infection
| Viral load at detection | Fiebig stage | Viral load ramp-up * | Peak viral load | Treatment | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Days since negative viral load | Log copies10 per mL | Days after detection until seen | Number of visits | Number of samples (blood, FGT, stool) | Log copies10 per mL | Days to peak viral load | Started cART | Days to treatment initiation | Time to <20 copies per mL (days) | ART regimen | ||
| Untreated acute HIV infection (n=14) | ||||||||||||
| 1 | 4 | 2·67 | 2 | I | 4 | 1 | 7·04 | 7 | ·· | ·· | ·· | ·· |
| 2 | 4 | 4·85 | 1 | I | 4 | 1 | 7·76 | 7 | Yes | 724 | 120 | 3 drug |
| 3 | 4 | 3·29 | 1 | I | 4 | 2 | 6·91 | 10 | Yes | 310 | 174 | 3 drug |
| 4 | 3 | 4·57 | 4 | I | 3 | 2 | 7·48 | 11 | Yes | 347 | Withdrew | 3 drug |
| 5 | 4 | 2·20 | 2 | I | 5 | 2 | 7·72 | 10 | Yes | 809 | 109 | ·· |
| 6† | 4 | 5·03 | 1 | I | 4 | 2 | 7·11 | 7 | ·· | ·· | ·· | ·· |
| 7 | 4 | 1·99 | 2 | I | 7 | 3 | 6·60 | 17 | Yes | 456 | 156 | 3 drug |
| 8 | 3 | 4·57 | 1 | I | 6 | 3 | 6·98 | 15 | Yes | ·· | 98 | 3 drug |
| 9 | 3 | 3·23 | 1 | I | 4 | 2 | 6·99 | 8 | ·· | ·· | ·· | ·· |
| 10 | 21 | 5·76 | 1 | III | 3 | 1 | 8·29 | 4 | ·· | ·· | ·· | ·· |
| 11 | 3 | 2·97 | 1 | I | 5 | 2 | 7·15 | 7 | Yes | 427 | 70 | 3 drug |
| 12 | 4 | 4·51 | 1 | I | 4 | 2 | 7·15 | 7 | ·· | ·· | ·· | ·· |
| 13 | 3 | 2·95 | 1 | I | 5 | 2 | 5·89 | 11 | ·· | ·· | ·· | ·· |
| 14 | 4 | 4·45 | 1 | I | 4 | 2 | 8·00 | 7 | Yes | 528 | ·· | 3 drug |
| Median (IQR) | 4·00 (3·00–4·00) | 3·87 (2·96–4·57) | 1·00 (1·00–1·75) | ·· | 4·00 (4·00–5·00) | 2·00 (2·00–2·00) | 7·13 (6·98–7·66) | 7·50 (7·00–10·75) | ·· | ·· | 114·50 (100·75–147·00) | ·· |
| Treated acute HIV infection (n=28) | ||||||||||||
| 15 | 7 | 3·78 | 1 | I | 2 | 2 | 4·37 | 1 | Yes | 1 | 27 | 3 drug |
| 16 | 7 | 1·99 | 1 | I | 3 | 1 | 4·53 | 2 | Yes | 1 | 18 | 3 drug |
| 17 | 45 | 4·92 | 1 | V | Not detected‡ | Not detected‡ | 4·92 | 0 | Yes | 1 | 72 | 3 drug |
| 18 | 42 | 1·99 | 1 | V | 2 | 0 | 2·11 | 0 | Yes | 3 | 9 | 3 drug |
| 19 | 4 | 4·04 | 1 | I | 3 | 1 | 5·55 | 3 | Yes | 1 | 62 | 3 drug |
| 20 | 3 | 2·28 | 1 | I | 4 | 2 | 3·34 | 7 | Yes | 1 | 25 | 3 drug |
| 21 | 78 | 5·37 | 1 | III | 2 | 2 | 5·88 | 1 | Yes | 1 | Withdrew | 3 drug |
| 22 | 3 | 3·72 | 1 | I | 3 | 1 | 5·32 | 4 | Yes | 1 | 38 | 3 drug |
| 23 | 5 | 2·91 | 3 | I | 3 | 1 | 4·61 | 7 | Yes | 3 | 21 | 3 drug |
| 24 | 4 | 2·75 | 1 | I | 2 | 1 | 3·77 | 1 | Yes | 1 | 13 | 3 drug |
| 25 | 3 | 2·43 | 2 | I | 3 | 1 | 4·95 | 3 | Yes | 3 | 78 | 3 drug |
| 26 | NA§ | 5·32 | 1 | V | Not detected‡ | Not detected‡ | 5·32 | 0 | Yes | 1 | 22 | 4 drug |
| 27 | 3 | 3·64 | 1 | I | 2 | 1 | 4·72 | 1 | Yes | 1 | 13 | 4 drug |
| 28 | 3 | 3·65 | 2 | I | 3 | 1 | 6·41 | 4 | Yes | 2 | 33 | 4 drug |
| 29 | 3 | 2·40 | 1 | I | 2 | 1 | 4·15 | 1 | Yes | 1 | 17 | 4 drug |
| 30 | 10 | 7·57 | 1 | III | 1 | 0 | 7·57 | 0 | Yes | 1 | 27 | 4 drug |
| 31 | 3 | 1·99 | 1 | I | 1 | 1 | 4·22 | 1 | Yes | 1 | 10 | 4 drug |
| 32 | 4 | 2·56 | 1 | I | 4 | 3 | 3·86 | 7 | Yes | 1 | 13 | 4 drug |
| 33 | 4 | 2·64 | 1 | I | 1 | 0 | 2·64 | 0 | Yes | 1 | 6 | 4 drug |
| 34 | 7 | 3·76 | 2 | I | 3 | 2 | 4·93 | 4 | Yes | 2 | 16 | 4 drug |
| 35 | 4 | 2·23 | 1 | I | 3 | 1 | 2·89 | 3 | Yes | 1 | 13 | 4 drug |
| 36 | 4 | 2·75 | 1 | I | 3 | 1 | 3·20 | 3 | Yes | 1 | 15 | 4 drug |
| 37 | 5 | 4·88 | 1 | I | 2 | 1 | 5·45 | 1 | Yes | 1 | 41 | 4 drug |
| 38 | 4 | 2·97 | 1 | I | 1 | 0 | 2·97 | 0 | Yes | 1 | 9 | 4 drug |
| 39 | 4 | 3·23 | 1 | I | 3 | 1 | 3·75 | 3 | Yes | 1 | 20 | 4 drug |
| 40 | 3 | 1·99 | 1 | I | 2 | 1 | 2·96 | 1 | Yes | 1 | 11 | 4 drug |
| 41 | 3 | 1·99 | 1 | I | 2 | 1 | 3·19 | 1 | Yes | 1 | 6 | 4 drug |
| 42 | 4 | 4·27 | 1 | I | Not detected‡ | Not detected‡ | 4·27 | 0 | Yes | 1 | 9 | 4 drug |
| Median (IQR) | 4·0 (3·00–6·00) | 2·94 (2·37–3·85) | 1·0 (1·00–1·00) | ·· | 2·00 (2·00–3·00) | 1·00 (1·00–1·00) | 4·3 (3·31–5·04) | 1·0 (0·75–3·00) | ·· | 1·0 (1·00–1·00) | 17·0 (12·00–27·00) | ·· |
Of 42 individuals diagnosed with acute HIV-1 infection after study enrolment, 14 were untreated and 28 received immediate ART after detection. 36 were diagnosed during Fiebig stage I, including 13 in the untreated group and 23 in the treated group. Of the 28 participants who received immediate treatment, 11 received a standard three-drug regimen of tenofovir, emtricitabine, and efavirenz (given once per day as a fixed-dose combination pill) and 17 received a four-drug regimen that also included raltegravir. FGT=female genital tract. ART=combination antiretroviral therapy. NA=not applicable.
Viral load ramp-up describes the period between HIV detection and peak viraemia, and thus is indicative of rapid viral replication.
This individual was assigned Fiebig stage I based on a 4 day gap since the last negative result for HIV RNA, although no p24 antigen result was available on the day of detection and the initial viral load was high (5·03 log 10 copies per mL).
Three participants were not detected during viral load ramp-up because the second viral load was equal to or less than the viral load at detection.
Days since the last negative viral load was NA for this individual because HIV RNA was detected on the first day of screening.
23 (82%) of the 28 participants who received immediate ART after detection were diagnosed during Fiebig stage I acute HIV infection (table 2), with a median viral load at detection of 2·75 log10 copies per mL (IQR 2·34–3·69). ART was initiatied within a median of 1 day (1–1) after detection and reduced the median peak viral load to 4·22 log10 copies per mL (3·27–4·83) within 2 days (1–3), compared with 7·11 log10 copies per mL (6·98–7·70; p<0·0001) within 8 days (7–11) for the 13 untreated individuals with Fiebig stage I HIV infection. Viral suppression (<20 copies per mL) was achieved within a median of 16 days (12–26) in the 23 immediately treated participants; 26 days (20–44) for the eight participants who received the three-drug regimen and 13 days (10–13) for the 15 participants who received the four-drug regimen (p<0.018; figure 2). A positive correlation was seen between viral load at treatment initation and peak viral load (r=0·73; p<0·0001) and time to suppression (r=0·53; p=0·005; appendix p 5).
Figure 2. Effect of early treatment on viral load and CD4 cell dynamics in acute HIV infection.
(A) The plasma viral load for 14 participants untreated during acute HIV infection is plotted against days after detection of HIV-1 RNA, with day 0 being the day of detection. (B) Effect of ART on viral load for 28 participants initiated on treatment immediately after detection of HIV-1 RNA; the horizontal boxes show median time to viral suppression (<20 copies per mL) for Fiebig stage I (16 days, IQR 12–26, range 6–78) versus Fiebig stage II-V (25 days, 19–38, 9–72). (C) Effect of immediate ART on viral-load dynamics in 23 participants with Fiebig stage I acute HIV infection compared with 13 untreated women with Fiebig stage I infection; median viral load is plotted against longitudinal sampling timepoints; the vertical bars on each curve represent the range for viral load at each timepoint. (D) Reduction of peak HIV viraemia by immeidate initiation of ART in participants detected during Fiebig stage I acute HIV infection; median peak viral load (bold lines) and IQR (thin lines) are shown for 13 untreated (median 7.11 log10 copies per mL, IQR 6.98–7.48) and 23 early treated (4.22 log10 copies per mL, 3.27–4.83) participants (p<0.0001). (E) Dynamics of the CD4 cell count for 13 untreated and 23 early treated participants with Fiebig stage I acute HIV infection; pre-infection CD4 was the absolute CD4 cell count from the sampling point preceding detection of acute HIV infection; nadir CD4 was the lowest absolute CD4 cell count measured after detection of infection; rebound CD4 cell count was the absolute count collected at the scheduled 42 days sampling point. ART=combination antiretroviral therapy.
Immediate treatment minimised the reductions in CD4 cell count and resulted in higher CD4 rebound than observed for untreated infection (figure 2E). For the 13 women with untreated Fiebig stage I acute HIV infection, the median CD4 count before infection was 812 cells per μL (IQR 662–993), which decreased to a nadir of 304 cells per μL (280–457; p=0·0002) within a median of 7 days (7–11) after infection (figure 2E) and recovered to a median of 526 cells per μL (IQR 491–590; p=0·002) within a median of 43 days (42–45). For the 23 women with Fiebig stage I acute HIV infection who were treated with ART, the median CD4 count before infection was 926 cells per μL (740–1206), which was reduced to a median nadir of 685 cells per μL (561–802; p<0·0001) within 14 days (3–21) after infection (figure 2E) and recovered to a median of 847 cells per μL (695–1136; p<0·0001) within 42 days (42–46). Immediate treatment preserved the CD4 count: the median nadir was 304 cells per μL for the untreated women (p<0·0001).
Of 23 women diagnosed with Fiebig stage I acute HIV infection who received immediate treatment, only three (13%) became positive in the western blot; nine (39%) had indeterminate results and 11 (48%) remained consistently negative up to 340 days of follow-up (figure 3). By contrast, all 13 untreated women with Fiebig stage I acute HIV infection developed positive western blot patterns. Heterogeneity in seroconversion was also observed with rapid immunoassays, ELISA, and p24Ag testing (appendix p 6).
Figure 3. Effect of ART initiated during acute HIV infection on western blot development.
Longitudinal western-blot testing is shown for 36 participants with Fiebig stage I HIV infection, of whom 13 were untreated and 23 received immediate ART. *Fiebig stage III. †Fiebig stage V.
Of 385 participants who completed the 48 week socioeconomic intervention, 202 (87%) of 231 women who completed 1 year follow-up were successfully placed within 1 year after intervention: 78 (34%) in full-time employment, 57 (25%) in a work internship or learnership, 37 (16%) in a tertiary education programme, 27 (12%) in a high-school completion course, and three (1%) had started their own business.
DISCUSSION
Young, socioeconomically disadvantaged women at high risk of HIV infection are able to adhere to frequent follow-up and collection of biological specimens, thereby allowing very early detection of acute HIV infection. 42 women were diagnosed with acute HIV infection, many of whom had initial viral loads of less than 100 RNA copies per mL of plasma (1.99 log10 copies per mL). 36 (86%) of the 42 women were in Fiebig stage I HIV infection, of whom 23 were started on immediate ART. These women showed a substantial reduction in peak viraemia, relative sparing of CD4-positive T cells, and rapid clearance of plasma viraemia compared with the 13 untreated women with Fiebig stage I HIV infection. Time to clearance of plasma viraemia was rapid and shortened by addition of raltegravir. Immediate treatment had a substantial effect on HIV seroconversion, with 20 (87%) of 23 treated Fiebig stage I infections not developing a positive western blot. By contrast, those treated in Fiebig stages III-IV all seroconverted, as did all untreated participants. Some of the third-generation and fourth-generation antibody tests were positive in the absence of western-blot positivity, perhaps indicative of continued low-level virus replication despite immediate ART or long-term memory responses to major structural proteins. These results suggested that some individuals who initiate ART during Fiebig stage I acute HIV infection might not seroconvert, rendering antibody tests for confirmation of HIV infection unreliable. Further research is required to establish confirmatory algorithms for HIV infection in this situation.
Of the 385 women who completed the integrated 48 week empowerment and job skills programme, 202 (87%) obtained jobs or internships, started a business, or resumed their education within 1 year after the intervention. This achievement is noteworthy in South Africa, where 45% of young black women aged 15–35 years are unemployed22, and where women are 20 times more likely to drop out of school than men. 23 Although we did not aim to assess the effect of the intervention on HIV prevention, our study allowed insight into the challenges of poverty and absence of opportunity in this community. Furthermore, this study provides a model for combining basic science and research with a social support programme to directly benefit participants and facilitate aherence to rigorous follow-up schedules. For those who became HIV-infected, intensive counselling on disclosure to partners was provided; access to counselling and testing was also provided to partners. Future studies should address whether such an approach has an effect on HIV acquisition and transmission.
Collection of demographic and behavioural data allowed assessment of factors that might predispose this population to HIV infection. All women enrolled in the study were considered high risk on the basis of living in poverty in a highly HIV-endemic area. Unlike other studies that have attributed high HIV incidence among young women to multiple or intergenerational sexual partners, neither of these behaviours were observed in the FRESH cohort. Both uninfected and infected participants reported few current and lifetime sexual partners, and the median age gap between participants and their current sexual partner was only 2·8 years; that between participants and their oldest sexual partner was only 3.8 years. 24,25 Only the number of lifetime sexual partners and sexual intercourse during the 30 days before detection was significantly associated with HIV acquisition risk. Information about the patterns of sexual relationships, whether concurrent or sequential, might be important to better understand behavioural risk. These findings suggest that other factors, still to be determined, contributed to the high HIV incidence in our cohort and emphasise the need for additional studies of similar design to our study (ie, with regular mucosal and blood sampling before and after HIV infection). This study highlighted some factors that might affect the high HIV incidence among women in South Africa, such as reproductive health practices and the microbiome of the female genital tract.3,19,26
The availability of samples before infection and during Fiebig stage I enabled characterisation of early events during acute HIV infection and identification of factors leading to immune dysfunction that might enable the virus to persist and become chronic.27,28 These early events included massive HIV-specific CD8 cell activation, followed by rapid apoptosis, depletion of innate lymphoid cells,27 and biological factors that underlie inflammation of the female genital tract and risk of HIV-1 acquisition.3,19
Our experience also highlights the need for flexibility in the setting of evolving ART options and treatment guidelines. At study commencement, immediate ART was not included in the context of HIV treatment guidelines for South Africa, which restricted free ART to individuals with CD4 counts of 350 cells per μL or fewer. However, success in identification of acute infections and observation of rapid viral replication with high peak and prolonged viraemia prompted us to introduce immediate ART with a standard three-drug regimen from July, 2014. This strategy resulted in significant reduction of peak and duration of viraemia. In January, 2015, we introduced an intensified regimen containing raltegravir twice daily, which further decreased in time to viral suppression. In a future study, we plan to assess the effect of immediate ART on acute HIV infection and that of innate and adaptive immune responses on the establishment and persistence of viral reservoirs.
Immediate ART during acute HIV infection, as done in this study, might have relevance for functional cure research and long-term remission of HIV. Changes in WHO guidelines call for a combination prevention approach, with commencement of ART for all HIV-infected individuals, and the use of daily oral pre-exposure prophylaxis for those at substantial risk of HIV infection.30 Therefore, future intervention research, including vaccine and passive immunisation studies, is likely to be done in the context of ART and pre-exposure prophylaxis. Our study offers a platform to do HIV cure research, making it possible to address important questions about reservoir establishment and persistance, and the effect of immediate ART on antiviral immune responses. Further analyses of the immediately treated cohort in this study are underway to examine viral reservoir and dissemination, toward development of interventions to achieve functional cure. The successful incorporation of a socioeconomic intervention to facilitate adherence to a complex study protocol, while delivering direct benefit to participants, offers a model for other investigators doing biomoedical research within vulnerable and economically disadvantaged communities.
Limitations of our study included a small sample size that might have affected the detection of more significant differences between the untreated and early treated groups. Another limitation is that our study was not designed to assess the effectiveness of the life-skills and job-readiness programme in prevention of HIV acquisition. Additionally, the study did not include men and did not involve sexual partners of study participants.
In conclusion, to our knowledge, this study is the first to show in a resource-limited, HIV-hyperendemic African setting, that frequent monitoring and collection of samples before and during the earliest stages of acute HIV infection can be achieved successfully with initiation of ART immediately after detection. We have also shown that immediate initiation of ART during Fiebig stage I infection can lead to rapid and sustained viral suppression. This cohort is ideal for testing of new prevention and eradication strategies in a key HIV high-risk group in the global region of greatest need.
Supplementary Material
RESEARCH IN CONTEXT.
Evidence before this study
We searched PubMed between Jan 1, 2008, and Jan 31, 2017, for original and review English language articles on acute HIV infection, HIV in young women in sub-Saharan Africa, HIV susceptibility and incidence in young women, HIV cure strategies, and antiretroviral therapy (ART) for acute HIV infection using the search terms “acute HIV,” “infection,” “HIV eradication,” “HIV cure strategies,” “immediate ART,” “early ART,” “HIV cure,” “HIV vaccine,” “HIV reservoirs,” and “Fiebig stage I HIV infection”. We found several studies of acute HIV infection, but these studies had substantial limitations in that, in almost all studies, no mucosal or peripheral blood samples were taken before infection that could help unravel biological and behavioural risk factors for HIV acquisition. Furthemore, previous studies of acute HIV infection did not have frequent sampling during the critical stage before peak viremia, and therefore, missed the opportunity to explore virological and host events associated with subsequent control of HIV. Although early initiation of ART has been proposed as a strategy to achieve a functional cure and to understand the mechanisms of viral reservoir formation, we found no studies of this strategy in sub-Saharan Africa, and, even in the most resource-rich settings, this strategy has only been attempted after seroconversion when the reservoir might already be well established.
Added value of this study
We showed that Fiebig stage I acute HIV infection could be identified in a high-incidence, resource-poor setting by frequent sampling before and immediately after infection. Our observations of CD4 cell preservation and reduction of peak viraemia with combination ART might have important implications for attempts to inhibit reservoir formation and subsequently purge the virus to achieve a functional cure. A socioeconomic empowerment programme, designed to coincide with sampling before and after infection, might facilitate HIV prevention and pathogenesis research.
Implications of all the available evidence
We have established a unique cohort optimised for prevention, vaccine, and cure research in a highly endemic, resource-poor setting.
Acknowledgements
We thank the participants, the FRESH clinical team, the HPP GCLP Lab staff, and Global Laboratories for their excellent technical support and sample processing. We also thank the United States Military HIV Research Program for their helpful discussions. This study was supported in part by grants from the Bill & Melinda Gates Foundation, the International AIDS Vaccine Initiative (UKZNRSA1001), the National Institute of Allergy and Infectious Disesaes (R37AI067073), the Harvard University CFAR grant (P30 AI060354), the Witten Family Foundation, Dan and Marjorie Sullivan, the Mark and Lisa Schwartz Foundation, Ursula Brunner, AIDS Healthcare Foundation, and the Howard Hughes Medical Institute. Raltegravir was donated by Merck & Co. This work was also partially supported through the Sub-Saharan African Network for TB/HIV Research Excellence, a Developing Excellence in Leadership, Training and Science (DELTAS) Africa Initiative (DEL-15-006). The DELTAS Africa Initiative is an independent funding scheme of the African Academy of Sciences (AAS)’s Alliance for Accelerating Excellence in Science in Africa (AESA) and is supported by the New Partnership for Africa’s Development Planning and Coordinating Agency (NEPAD Agency) with funding from the Wellcome Trust (107752/Z/15/Z) and the UK government. The views expressed in this publication are those of the author(s) and not necessarily those of AAS, NEPAD Agency, Wellcome Trust, or the UK government.
Footnotes
Declaration of interests
We declare no competing interests.
Contributor Information
Krista L Dong, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA; Infectious Disease Division, Massachusetts General Hospital, Boston, MA, USA.
Amber Moodley, HIV Pathogenesis Programme, Doris Duke Medical Research Institute.
Douglas S Kwon, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA; Infectious Disease Division, Massachusetts General Hospital, Boston, MA, USA.
Musie S. Ghebremichael, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA.
Mary Dong, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA.
Nasreen Ismail, HIV Pathogenesis Programme, Doris Duke Medical Research Institute.
Zaza M Ndhlovu, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA; HIV Pathogenesis Programme, Doris Duke Medical Research Institute.
Jenniffer M Mabuka, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA; HIV Pathogenesis Programme, Doris Duke Medical Research Institute; Africa Health Research Institute.
Daniel M Muema, HIV Pathogenesis Programme, Doris Duke Medical Research Institute; Africa Health Research Institute.
Karyn Pretorius, HIV Pathogenesis Programme, Doris Duke Medical Research Institute.
Nina Lin, Infectious Disease Division, Massachusetts General Hospital, Boston, MA, USA.
Bruce D Walker, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA; Infectious Disease Division, Massachusetts General Hospital, Boston, MA, USA; HIV Pathogenesis Programme, Doris Duke Medical Research Institute; University of KwaZulu-Natal, Durban, South Africa; Institute for Medical Sciences and Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA; Howard Hughes Medical Institute, Chevy Chase, MD, USA.
Prof Thumbi Ndung’u, Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA; HIV Pathogenesis Programme, Doris Duke Medical Research Institute; Africa Health Research Institute; Max Planck Instiute for Infection Biology, Berlin, Germany.
References
- 1.Kharsany ABM, Cawood C, Khanyile D, et al. Strengthening HIV surveillance in the antiretroviral therapy era: rationale and design of a longitudinal study to monitor HIV prevalence and incidence in the uMgungundlovu District, KwaZulu-Natal, South Africa. BMC Public Health 2015; 15: 1149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Abdool Karim Q, Kharsany AB, Frohlich JA, et al. HIV incidence in young girls in KwaZulu-Natal, South Africa—public health imperative for their inclusion in HIV biomedical intervention trials. AIDS Behav 2012; 16: 1870–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Byrne EH, Anahtar MN, Cohen KE, et al. Association between injectable progestinonly contraceptives and HIV acquisition and HIV target cell frequency in the female genital tract in South African women: a prospective cohort study. Lancet Infect Dis 2016: 16: 441–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Excler JL, Robb ML, Kim JH. Prospects for a globally effective HIV-1 vaccine. Am J Prev Med 2015; 49 (suppl 4): S307–18. [DOI] [PubMed] [Google Scholar]
- 5.Hope TJ, Marrazzo JM. A shot in the arm for HIV prevention? Recent successes and critical thresholds. AIDS Res Hum Retroviruses 2015; 31: 1055–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ensoli B, Cafaro A, Monini P, Marcotullio S, Ensoli F. Challenges in HIV vaccine research for treatment and prevention. Front Immunol 2014; 5: 417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Passaes CP, Saez-Cirion A. HIV cure research: advances and prospects. Virology 2014; 454–55: 340–52. [DOI] [PubMed] [Google Scholar]
- 8.Jain V, Hartogensis W, Bacchetti P, et al. Antiretroviral therapy initiated within 6 months of HIV infection is associated with lower T-cell activation and smaller HIV reservoir size. J Infect Dis 2013; 208: 1202–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Saez-Cirion A, Bacchus C, Hocqueloux L, et al. Post-treatment HIV-1 controllers with a long-term virological remission after the interruption of early initiated antiretroviral therapy ANRS VISCONTI Study. PLoS Pathog 2013; 9: e1003211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cohen MS, Smith MK, Muessig KE, Hallett TB, Powers KA, Kashuba AD. Antiretroviral treatment of HIV-1 prevents transmission of HIV-1: where do we go from here? Lancet 2013; 382: 1515–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hill AL, Rosenbloom DI, Fu F, Nowak MA, Siliciano RF. Predicting the outcomes of treatment to eradicate the latent reservoir for HIV-1. Proc Natl Acad Sci USA 2014; 111: 13475–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Harrison A, Colvin CJ, Kuo C, Swartz A, Lurie M. Sustained high HIV incidence in young women in southern Africa: social, behavioral, and structural factors and emerging intervention approaches. Curr HIV/AIDS Rep 2015; 12: 207–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Delva W, Abdool Karim Q. The HIV epidemic in southern Africa—is an AIDS-free generation possible? Curr HIV/AIDS Rep 2014; 11: 99–108. [DOI] [PubMed] [Google Scholar]
- 14.Ramjee G, Wand H, Whitaker C, et al. HIV incidence among non-pregnant women living in selected rural, semi-rural and urban areas in Kwazulu-Natal, South Africa. AIDS Behav 2012; 16: 2062–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shisana O, Zungu N, Evans M, Risher K, Rehle T, Clementano D. The case for expanding the definition of ‘key populations’ to include high-risk groups in the general population to improve targeted HIV prevention efforts. S Afr Med J 2015; 105: 664–69. [DOI] [PubMed] [Google Scholar]
- 16.Robb ML, Eller LA, Kibuuka H, et al. Prospective study of acute HIV-1 infection in adults in east Africa and Thailand. N Eng J Med 2016; 374: 2120–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McKinnon LR, Karim QA. Factors driving the HIV epidemic in southern Africa. Curr HIV/AIDS Rep 2016; 13: 158–69. [DOI] [PubMed] [Google Scholar]
- 18.Santelli JS, Edelstein ZR, Mathur S, et al. Behavioral, biological, and demographic risk and protective factors for new HIV infections among youth in Rakai, Uganda. J Acquir Immune Defic Syndr 2013; 63: 393–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Anahtar MN, Byrne EH, Doherty KE, et al. Cervicovaginal bacteria are a major modulator of host inflammatory responses in the female genital tract. Immunity 2015; 42: 965–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Fiebig EW, Wright DJ, Rawal BD, et al. Dynamics of HIV viremia and antibody seroconversion in plasma donors: implications for diagnosis and staging of primary HIV infection. AIDS 2003; 17: 1871–79. [DOI] [PubMed] [Google Scholar]
- 21.CDC. Interpretation and use of the western blot assay for serodiagnosis of human immunodeficiency virus type 1 infections. MMWR Suppl 1989; 38: 1–7. [PubMed] [Google Scholar]
- 22.Statistics South Africa. National and provincial labour market: youth. Q1: 2008—Q1: 2015. 2015. http://www.statssa.gov.za/publications/P02114.2/P02114.22015.pdf (accessed March 8, 2016).
- 23.Statistics South Africa. General household survey 2014. 2015. http:www.statssa.gov.za/publications/P0318/P03182014.pdf (accesed March 8, 2016).
- 24.Harling G, Newell ML, Tanser F, Kawachi I, Subramanian SV, Barnighausen T. Do age-disparate relationships drive HIV incidence in young women? Evidence from a population cohort in rural KwaZulu-Natal, South Africa. J Acquir Immune Defic Syndr 2014; 66: 443–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ott MQ, Barnighausen T, Tanser F, Lurie MN, Newell ML. Age-gaps in sexual partnerships: seeing beyond ‘sugar daddies’. AIDS 2011; 25: 861–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gosmann C, Anahtar MN, Handley SA, et al. Lactobacillus-deficient cervicovaginal bacterial communities are associated with increased HIV acquisition in young South African women. Immunity 2017; 46: 29–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Ndhlovu ZM, Kamya P, Mewalal N, et al. Magnitude and kinetics of CD8+ T cell activation during hyperacute HIV infection impact viral set point. Immunity 2015; 43: 591–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kloverpris HN, Kazer SW, Mjosberg J, et al. Innate lymphoid cells are depleted irreversibly during acute HIV-1 infection in the absence of viral suppression. Immunity 2016; 44: 391–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Standard treatment guidelines and essential medicines list for South Africa. Hospital level: adults. 3rd edn. Pretoria: The National Department of Health, 2012. [Google Scholar]
- 30.WHO. Guideline on when to start antiretroviral therapy and on pre-exposure prophylaxis for HIV. Geneva: World Health Organization, 2015. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.