Short Communication: Early Antiretroviral Therapy Is Associated with Better Viral Suppression and Less HIV Drug Resistance After Implementation of Universal Treatment in South Africa

Abstract

All people living with HIV should receive antiretroviral therapy (ART), but those with CD4 counts >500 cells/mm³ at ART initiation (“early initiators”) may be less motivated to adhere to treatment, compared with those with CD4 counts <200 cells/mm³ (“late initiators”). We performed a cross-sectional analysis among HIV-positive adults who had a viral load taken at 6 months after first-line ART initiation in a South African public clinic. Retrospective HIV drug resistance testing was performed on all samples with a viral load >1,000 copies/mL. We used Poisson regression models with robust variance to evaluate associations between early ART initiation and viral suppression <40 copies/mL. We assessed HIV drug resistance using descriptive statistics. Of 390 participants enrolled between February and August 2017, 60% were women and median age was 32 years [interquartile range (IQR) 27–38]. At ART initiation, median CD4 count was 366 cells/mm³ (IQR 204–546), and 30% were early initiators with CD4 > 500 cells/mm³. In multivariable analysis, early initiators were more likely to be virally suppressed compared with late initiators (adjusted risk ratio: 1.29, 95% confidence interval: 1.13–1.46). All 18 participants with viral load >1,000 copies/mL had successful genotyping, which identified drug resistance in 14/18 (77.8%). Among early initiators, drug resistance was detected in only 1/117 (0.9%), compared with 11/93 (11.8%) among late initiators. In conclusion, among people receiving ART in a South African public clinic, early initiators had better viral suppression after 6 months and less drug resistance than late initiators, which further supports universal treatment. Clinical trials registration: NCT03066128.

The World Health Organization (WHO) recommends that all people living with HIV (PLHIV) should be initiated on antiretroviral therapy (ART), irrespective of CD4 count.¹ However, there is little evidence regarding viral load suppression among PLHIV initiating ART at CD4 counts >500 cells/mm³ (“early initiators”) in public sector programs in low- and middle-income countries (LMICs).² Since many early initiators are asymptomatic, they may be less motivated to adhere to ART.³ We aimed to assess adherence, virological suppression, and HIV drug resistance among early initiators in South Africa, which implemented the WHO universal treatment policy in September 2016.

We performed a cross-sectional analysis among South African patients who had received 6 months of ART, and were being enrolled in a randomized trial of point-of-care HIV viral load testing.⁴ Eligible participants were nonpregnant clinically stable HIV-positive adults who were due a viral load test 6 months after ART initiation. Before enrolment, all patients had received care under routine programmatic conditions at the Prince Cyril Zulu Clinic, a large urban public sector clinic located in central Durban, South Africa. At enrolment, we obtained data from clinical records and administered three previously validated adherence questionnaires.^5,6 We performed point-of-care viral load testing using the Xpert^® HIV-1 VL (Cepheid, Sunnyvale) for participants in the intervention arm, and laboratory viral load testing with the Abbott Real-Time HIV-1 (m2000rt; Abbott Molecular, Chicago) for participants in the standard-of-care arm. Samples with a viral load ≥1,000 copies/mL had retrospective HIV drug resistance testing using Sanger consensus sequencing.

All participants enrolled in the parent trial were included in this analysis. The primary outcome was a suppressed viral load <40 copies/mL at 6 months after ART initiation. We used bivariable and multivariable Poisson regression models with robust variance to evaluate associations with initiation CD4 count, and potential confounders, including demographic and clinical variables (e.g., age, gender, education, income, alcohol use, previous ART exposure, time from HIV diagnosis to ART initiation, and previous tuberculosis). Covariates with a p value <.2 in bivariable analysis were included in a multivariable analysis, also adjusted for study arm. We used chi-squared tests to evaluate associations between CD4 count at initiation and the secondary outcome of self-reported adherence. We used descriptive statistics to assess HIV drug resistance by CD4 count strata, because numbers with resistance were small. We analyzed data using SAS version 9.4 (SAS Institute, Inc., Cary, NC). Ethical approval was obtained from the Biomedical Research Ethics Committee of the University of KwaZulu-Natal and the Institutional Review Board of the University of Washington.

Of 390 participants enrolled between February and August 2017, 235 (60%) were women and median age was 32 years [interquartile range (IQR) 27–38]. At ART initiation, median CD4 count was 366 cells/mm³ (IQR 204–546). There were 117 (30%) early initiators with a CD4 count >500 cells/mm³, and 93 (24%) late initiators with a CD4 count ≤200 cells/mm³. Most patients (387/390, 99%) received a once-daily fixed dose combination of tenofovir, emtricitabine, and efavirenz, and only 18/390 (5%) reported previous ART exposure.

Median time from ART initiation to viral load testing at study enrolment was 25 weeks (IQR 24–28). Viral load was <40 copies/mL in 336 (86.2%), 40–999 copies/mL in 36 (9.2%), and ≥1,000 copies/mL in 18 (4.6%) participants. In bivariable analysis, early initiators were more likely to be virally suppressed compared with late initiators with CD4 counts ≤200 cells/mm³ [risk ratio (RR): 1.32, 95% confidence interval (CI): 1.16–1.50, Table 1]. In addition, PLHIV with previous tuberculosis were less likely to be virally suppressed (RR: 0.84, 95% CI: 0.73–0.98). In multivariable analysis adjusting for age, previous tuberculosis, and previous ART exposure, early initiators remained more likely to be virally suppressed compared with late initiators (adjusted RR: 1.29, 95% CI: 1.13–1.46).

Table 1.

Higher CD4 Count at Antiretroviral Therapy Initiation Is Associated with Viral Suppression (<40 Copies/mL) After 6 Months, N = 390

Characteristic	Viral load <40 copies/mL n/N (%)	Bivariable analysis^a		Multivariable analysis
Characteristic	Viral load <40 copies/mL n/N (%)	RR ^b (95% CI)	p	Adjusted RR ^c (95% CI)	p
Initiation CD4 count (cells/mm³)
≤200	67/93 (72.8)	1		1
201–350	72/88 (81.8)	1.12 (0.96–1.32)	.154	1.10 (0.94–1.28)	.245
351–500	85/92 (92.4)	1.27 (1.10–1.46)	.001	1.24 (1.08–1.42)	.002
>500	112/117 (95.7)	1.32 (1.16–1.50)	.001	1.29 (1.13–1.46)	<.001
Previous tuberculosis
No	291/330 (88.4)	1		1
Yes	45/60 (74.6)	0.84 (0.73–0.98)	.026	0.89 (0.77–1.02)	.094

In bivariable analysis there was no evidence of an association between viral suppression and potential confounders of gender, income, education, alcohol use, time from diagnosis to ART initiation, and travel time to clinic.

Adjusted for study arm.

Adjusted for study arm, age, and previous self-reported ART exposure.

ART, antiretroviral therapy; CI, confidence interval; RR, risk ratio.

Regarding secondary outcomes, 235/390 (60%) reported very good to excellent adherence in the past month, 283/390 (73%) reported their last missed dose to be >4 weeks ago, and 314/390 (81%) reported no missed doses in the last 4 days. There was no evidence that any of the three self-reported adherence measures varied by CD4 count at ART initiation (all 95% CIs crossed 1 and p values >.1). All 18 participants with viral load >1,000 copies/mL had successful genotyping, which identified drug resistance in 14/18 (77.8%). All 14 had dual-class resistance against nucleoside reverse transcriptase inhibitors (NRTIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs), but no protease inhibitor resistance. The commonest mutations were M184I/V (n = 13), K65R (n = 5), and L74I/V (n = 5) against NRTIs, and K103N (n = 11) and V106M (n = 5) against NNRTIs. Among early initiators, drug resistance was detected in only 1/117 (0.9%), compared with 11/93 (11.8%) among late initiators.

After implementation of universal treatment in this public sector clinic, early initiators with CD4 counts >500 cells/mm³ had better viral suppression, less HIV drug resistance, and similar levels of self-reported adherence after 6 months compared with late initiators.

Results from clinical trials in LMICs have also suggested that early initiators had acceptable adherence⁷ and good viral suppression^2,8

–12 in research settings. In our study, all patients received care under programmatic conditions from ART initiation until the 6-month viral load, providing early confirmation of these findings in routine clinical settings in LMICs. We hypothesize that patients who engage with HIV services earlier and, therefore, have high CD4 counts are also more likely to continue engaging with treatment, resulting in better viral suppression. Furthermore, late initiators with advanced HIV infection may be more likely to have previous unreported ART exposure, resulting in HIV drug resistance,² as found in our study. Although our study provides initial data on successful treatment outcomes among early ART initiators in an LMIC setting, it is limited by the single site and cross-sectional design. In addition, different assays were used to measure the 6 month viral load at enrolment into this study, although we adjusted for this in our analysis. Overall, further research will be needed to assess longer term outcomes in a wider range of public sector clinics that have implemented universal treatment.

In conclusion, we demonstrate that PLHIV initiated early on ART at CD4 counts >500 cells/mm³ had good viral outcomes and less HIV drug resistance, which supports the implementation of universal treatment in nonresearch LMIC settings.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by the National Institute for Health, grant number AI124719-01. HIV drug resistance testing was funded by the South African National Health Laboratory Service Research Trust, grant number 2018-1DEV-PMO01. The funders had no role in the conception, drafting, or submission of this manuscript.

References

Ford

, Vitoria

, Doherty

: Providing antiretroviral therapy to all who are HIV positive: The clinical, public health and programmatic benefits of Treat All . J Int AIDS Soc, 2018; 21:e25078.

Fatti

, Grimwood

, Nachega

, et al.: Better virological outcomes amongst people living with HIV initiating early antiretroviral treatment (CD4 counts ≥500 cells/μL) in the HPTN 071 (PopART) trial in South Africa. Clin Infect Dis, 2019 [Epub ahead of print]; DOI: 10.1093/cid/ciz214.

Fox

, Rosen

: A new cascade of HIV care for the era of “treat all.”. PLoS Med, 2017; 14:e1002268.

Dorward

, Garrett

, Quame-Amaglo

, et al.: Protocol for a randomised controlled implementation trial of point-of-care viral load testing and task shifting: The Simplifying HIV TREAtment and Monitoring (STREAM) study. BMJ Open, 2017; 7:e017507.

Reynolds

, Sun

, Nagaraja

, Gifford

, Wu

, Chesney

: Optimizing measurement of self-reported adherence with the ACTG adherence questionnaire. J Acquir Immune Defic Syndr, 2007; 46:402–409.

Chaiyachati

, Hirschhorn

, Tanser

, Newell

M-L

, Bärnighausen

: Validating five questions of antiretroviral nonadherence in a public-sector treatment program in rural South Africa. AIDS Patient Care STDS, 2011; 25:110126193341078.

Iwuji

, McGrath

, Calmy

, et al.: Universal test and treat is not associated with sub-optimal antiretroviral therapy adherence in rural South Africa: The ANRS 12249 TasP trial. J Int AIDS Soc, 2018; 21:e25112.

Kwarisiima

, Kamya

, Owaraganise

, et al.: High rates of viral suppression in adults and children with high CD4+ counts using a streamlined ART delivery model in the SEARCH trial in rural Uganda and Kenya. J Int AIDS Soc, 2017; 20:21673.

Garrett

, Norman

, Leask

, et al.: Acceptability of early antiretroviral therapy among South African women. AIDS Behav, 2018; 22:1018–1024.

10.

The TEMPRANO ANRS 12136 Study Group. A trial of early antiretrovirals and isoniazid preventive therapy in Africa. N Engl J Med, 2015; 373:808–822.

11.

The INSIGHT START Study Group. Initiation of antiretroviral therapy in early asymptomatic HIV infection. N Engl J Med, 2015; 373:795–807.

12.

Haberer

, Bwana

, Orrell

, et al.: ART adherence and viral suppression are high among most non-pregnant individuals with early-stage, asymptomatic HIV infection: An observational study from Uganda and South Africa. J Int AIDS Soc, 2019; 22:e25232.