RESEARCH NOTE VIROLOGY

Timing of antiretroviral therapy initiation

Corresponding author: N. He, Department of Epidemiology,
after diagnosis of recent human School of Public Health, Fudan University, Shanghai, 200032, China
immunodeficiency virus infection and CD4+ E-mail: nhe@shmu.edu.cn
Y. Ding and S. Duan contributed equally to the study.
T-cell recovery

Y. Ding1, S. Duan2, Z. Wu3, R. Ye2, Y. Yang2, S. Yao2, J. Wang2,

L. Xiang2, Y. Jiang3, L. Lu4, M. Jia4, R. Detels5 and N. He1
1) Department of Epidemiology, School of Public Health, and the Key Introduction
Laboratory of Public Health Safety of Ministry of Education, Fudan University,
Shanghai, 2) Dehong Prefecture Centre for Disease Control and Prevention,
Research shows that early initiation of antiretroviral therapy
Mangshi, Yunnan Province, 3) National Centre for AIDS/STD Control and
(ART) in asymptomatic human immunodeficiency virus (HIV)-
Prevention (NCAIDS), Chinese Centre for Disease Control and Prevention
infected patients with higher CD4+ counts provided immuno-
(China CDC), Beijing, 4) Yunnan Centre for Disease Control and Prevention,
logical and clinical benefits [1,2]. Studies also reveal that ART
Kuming, Yunnan Province, China and 5) Department of Epidemiology, School
initiation in acute or early infection could lead to rapid CD4+ T-
of Public Health, University of California, Los Angeles, Los Angeles, USA
cell recovery even among those with CD4+ counts 500 cells/
μL [3,4]. However, in many of these studies the date of infec-
tion was either estimated through a negative Western blot
result followed by a positive Western blot result, or through
Abstract the staging method of primary HIV infection developed by
Fiebig et al. [5], so limiting its use in many resource-constrained
countries. There is a need for inexpensive, easy-to-use assays to
We retrospectively examined the timing of antiretroviral therapy
identify recent infections for early ART initiation.
(ART) initiation and CD4+ T-cell recovery over 36 months
The BED (HIV-1 subtypes B, E and D) immunoglobulin G
among recent human immunodeficiency virus (HIV) infections
capture enzyme immunoassay (BED-CEIA) is most commonly
using BED (HIV-1 subtypes B, E and D) immunoglobulin G
used for the purpose of HIV incidence estimation [6], and BED has
capture enzyme immunoassay (BED-CEIA). Regardless of baseline
proven sensitive for detecting recent infection within the past
CD4+ counts, individuals (n = 393) who initiated ART >2 months
155 days (95% CI146–165 days) [7,8]. BED was first adopted in
after diagnosis had significantly decreased probability and rate of
China in 2005 for HIV-1 incidence surveillance, and is now being
achieving CD4+ counts 900 cells/μL or 600 cells/μL than
used in all 31 provinces [9]. This inexpensive technology affords us
those individuals (n = 135) who started ART earlier
a unique opportunity to evaluate the potential benefit of early ART
(2 months). But the mean CD4+ counts in two groups
initiation for CD4+ T-cell recovery among recently infected
converged after 30 months of treatment. Early ART initiation
individuals.
leads to accelerated CD4+ recovery, but does not offer a long-
term advantage in CD4+ counts.
Clinical Microbiology and Infection © 2015 European Society of Methods
Clinical Microbiology and Infectious Diseases. Published by
Elsevier Ltd. All rights reserved.
Data sources of our study came from the Dehong Prefecture,
Yunnan Province, China. From January 2005 to December
Keywords: Antiretroviral therapy, CD4+T cells, China, human
2013, 1261 individuals were categorized as recently infected
immunodeficiency virus, immune recovery, recent infection
with HIV-1 using the BED assay (see Supplementary material,
Original Submission: 20 August 2015; Revised Submission:
Data S1). Participants were observed through May 2015. We
9 October 2015; Accepted: 15 November 2015
restricted our analysis to 528 participants aged 16 years, who
Editor: G. Antonelli
started ART before February 2014, had a CD4+ count mea-
Article published online: 25 November 2015
surement taken close to initiation of ART, and had at least two
CD4+ count measurements during the first 36 months of ART,
including one during the first 12 months. Details about follow

Clin Microbiol Infect 2016; 22: 290.e5–290.e8

Clinical Microbiology and Infection © 2015 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved
http://dx.doi.org/10.1016/j.cmi.2015.11.007
CMI Ding et al. Timing of ART initiation and CD4 recovery 290.e6

TABLE 1. Charateristics of study participants by the time from HIV diagnosis to ART initiation (n [ 528)

Early-ART group Later-ART group

(ART (ART initiated
initiated £ 2 months) >2 months)
(n [ 135) (n [ 393) p

Baseline (i.e. at ART initiation) variables

Age (years) 31.5 (24.1–40.9) 34.1 (26.8–41.9) 0.088
16–29 62 45.9 144 36.6
30–44 45 33.3 173 44.0
45 28 20.7 76 19.3
Male sex 69 51.1 167 42.5 0.082
Han ethnicity 57 42.2 170 43.3 0.834
Had at least junior middle school education 54 40.0 154 82.6 0.279
CD4+ count (cells/μL)a 376 (289–549) 352 (252–502) 0.153
<350 59 49.9 196 43.7
350–499 30 24.7 97 22.2
500 46 25.4 100 34.1
Plasma viral load (log10 copies/mL), n = 85 4.4 (3.5–4.7) 4.3 (3.9–4.8) 0.806
Time from HIV diagnosis to ART initiation (months) 0.4 (0.2–0.8) 28.2 (11.1–48.7)
Outcome variables during first 36 months of treatment
Duration of treatment (months)b 25.3 (19.1–36.0) 30.5 (20.0–36.0) 0.005
CD4+ count peak (cells/μL) 767 (607–1010) 679 (521–910) <0.001
Time to CD4+ count peak (months) 13.6 (6.1–20.7) 17.3 (8.9–26.4) <0.001
Difference between baseline and peak CD4+ counts (cells/μL) 344 (169–575) 299 (182–465) 0.169
Attainment of CD4+ count 900 cells/μL 53 39.3 100 25.4 0.002
Time to first CD4+ count 900 cells/μL (months) 8.2 (3.9–15.7) 11.8 (5.6–19.8) 0.056
Attainment of CD4+ count 600 cells/μL 102 75.6 246 62.6 0.006
Time to first CD4+ count 600 cells/μL (months) 4.5 (2.9–9.1) 6.1 (3.0–14.0) 0.002
Attainment of viral load suppression, (n = 358) 78 95.1 266 96.4 0.890
Time to viral load suppression (months)c 7.5 (5.9–11.8) 8.7 (5.9–12.0) 0.241

Data are shown as N (%) or median (IQR).

Abbreviations: ART, antiretroviral therapy; HIV, human immunodeficiency virus; IQR, interquartile range.
Subjects with missing data were excluded from analysis.
a
Since 2012, all new reported HIV infections are eligible for free cART regardless of CD4+ counts in the study area.
b
Maximum duration of treatment was up to 36 months.
c
Refers to undetectable plasma HIV viral loads (i.e. <50 copies/mL).

up and ART are presented in the Supplementary material (Data
S1). The study was approved by institutional ethics committees.
Participants were classified by the timing of ART initiation
after HIV diagnosis: 135 in the early-ART group (2 months),
and 393 in the later-ART group (>2 months). The primary and
secondary outcomes were the attainment of CD4+ count
900 cells/μL and 600 cells/μL during the first 36 months of
treatment, respectively. Chi-squared tests, Wilcoxon–Mann–
Whitney tests, logistic regression models, Cox proportional-
hazard models, and mixed-effect spline models with random
intercept and slope (square-root CD4+ count) were used
where appropriate. We also assessed the sensitivity of our
results for alternative ways to classify treatment groups ac-
cording to the timing of ART initiation after diagnosis of recent

TABLE 2. Primary and sensitivity analysis for the association between timing of ART and CD4+ T-cell recovery during the first
36 months of ART

CD4+ count ‡900 cells/μL CD4+ count ‡600 cells/μL

Adjusted ORa Adjusted RRa Adjusted ORa Adjusted RRa

(95% CI) p (95% CI) p (95% CI) p (95% CI) p

Primary analysis
Time from HIV diagnosis to ART 0.46 (0.28– 0.75) 0.002 0.66 (0.47–0.93) 0.017 0.56 (0.34–0.92) 0.022 0.71 (0.56–0.90) 0.004
initiation (>2 versus 2 months)
Age at ART initiation (versus 16–29 years)
30–44 years 0.57 (0.35–0.95) 0.030 0.69 (0.48–0.99) 0.042 0.56 (0.34–0.91) 0.018 0.74 (0.58–0.94) 0.013
45 years 0.26 (0.12–0.56) <0.001 0.44 (0.25–0.77) 0.004 0.30 (0.17–0.54) <0.001 0.53 (0.38–0.74) <0.001
CD4+ count at ART initiation (versus <350 cells/μL)
350–499 cells/μL 2.48 (1.35–4.56) <0.001 2.31 (1.43–3.74) <0.001 3.50 (2.14–5.71) <0.001 2.47 (1.87–3.25) <0.001
500 cells/μL 9.50 (5.53–16.33) <0.001 7.10 (4.70–10.72) <0.001 9.70 (5.16–18.26) <0.001 4.82 (3.69–6.30) <0.001
Sensitivity analysis
Model 1: Time from HIV diagnosis to
ART initiation (versus 2 months)
>2–12 months 0.59 (0.30–1.13) 0.113 0.80 (0.51–1.24) 0.323 0.67 (0.35–1.29) 0.232 0.78 (0.57–1.06) 0.110
>12 months 0.42 (0.25–0.71) 0.001 0.61 (0.42–0.88) 0.008 0.52 (0.31–0.88) 0.014 0.69 (0.54–0.78) 0.003
Model 2: Time from HIV diagnosis to 0.48 (0.30– 0.76) 0.002 0.68 (0.49–0.94) 0.020 0.60 (0.38–0.94) 0.027 0.75 (0.61–0.94) 0.013
ART initiation (>4 versus 4 months)
Model 3: Time from HIV diagnosis to 0.99 (0.98–0.99) 0.007 0.99 (0.98–0.99) 0.017 0.99 (0.98–0.99) 0.012 0.99 (0.99–0.99) 0.015
ART initiation as continuous (months)

Abbreviations: ART, antiretroviral therapy; HIV, human immunodeficiency virus; OR, odds ratio; RR, rate ratio.
a
All analyses adjusted simultaneously for age, gender, ethnicity, education and baseline CD4+ count as indicated in Table 1.

Clinical Microbiology and Infection © 2015 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved, CMI, 22, 290.e5–290.e8
290.e7 Clinical Microbiology and Infection, Volume 22 Number 3, March 2016
infection: (1) >12 and 2–12 versus 2 months; (2) >2 versus
2 months; (3) as a continuous variable. All analyses were
performed using SAS 9.11 and SPSS 18.
Results
There were no significant differences between the two groups
for all variables at baseline (Table 1). During the first 36 months
of treatment, the proportions of participants achieving CD4+
count 900 cells/μL or 600 cells/μL were significantly higher
in the early-ART group than in the later-ART group and the
median time from ART initiation to such attainments was also
shorter in the early-ART group (Table 1).
The later-ART group had significantly lower odds and rate of
achieving CD4+ count 900 cells/μL than the early-ART group.
Sensitivity analysis showed a trend of decreased probability and
rate of CD4+ recovery as time from HIV diagnosis to ART
initiation increased. Furthermore, those individuals with base-
line CD4+ counts of 350 and 499 cells/μL or 500 cells/μL and
who initiated ART at an age of <30 years were also indepen-
dently associated with decreased odds and rates of achieving
CD4+ count 900 cells/μL or 600 cells/μL (Table 2).
After treatment, the estimated mean CD4+ counts increased
steeply during the first 6 months of treatment for both groups
but faster in the early-ART group (631 versus 502 cells/μL; p
<0.001), and then it reached a plateau in the early-ART group
during 6 and 36 months of treatment but increased less steeply
in the later-ART group. The estimated mean CD4+ counts
converged after 30 months of treatment (622 versus 585 cells/
μL; p 0.349) (see Supplementary material, Fig. S1 in Data S1).
Discussion
Our findings are consistent with previous studies regarding the
benefits of early treatment [3,4,10]. But our results also indi-
cated that the CD4 response to treatment converged by
3 years in two groups—similarly observed by Hecht et al.
[11]—indicating that immediate ART initiation after infection
may not offer a long-term CD4+ advantage. Nonetheless, an
early favourable immunological response to treatment has
proven to be associated with subsequent delayed disease pro-
gression and vice versa, regardless of baseline CD4+ counts and
plasma viral loads [12,13]. This suggests that early ART initia-
tion may lead to long-term clinical benefits [1]. This study also
demonstrates that baseline CD4+ count is an independent
predictor of CD4+ T-cell recovery, a finding consistent with the
existing literature [4,14]. These findings underscore the joint
Clinical Microbiology and Infection © 2015 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved, CMI, 22, 290.e5–290.e8
CMI
effects of the timing of baseline CD4+ count on CD4+ T-cell
recovery.
Another notable finding is that older age is associated with
decreased odds and rate of CD4+ T-cell recovery, which is
consistent with the existing literature [15,16]. Plasma viral load
suppression was achieved for most participants who were on
ART, with no significant differences between the two groups.
This was also observed in other studies [4,17].
Several limitations should be noted. First, this is an obser-
vational study. Second, the BED assay may misclassify a pro-
portion of individuals as being recently infected, a previous
study conducted in the same study area indicated that the
specificity of BED assay was 82.6% [18], where this level of
specificity suggests that some individuals who initiated ART
shortly after diagnosis may not have been recently infected. In
other words, there is the likelihood of underestimating the
effects of early ART initiation.
In conclusion, early initiation of ART leads to rapid and
better CD4+ T-cell recovery, but it does not offer a long-term
advantage in CD4+ counts. Nonetheless, given that the early
favourable immunological response is associated with better
clinical outcomes, further investigations into the long-term
clinical benefits is warranted. BED assay is a useful tool for
detecting recent HIV infections to allow for early treatment,
particularly in resource-constraint settings.
Acknowledgements
The authors wish to thank all who participated in this study and
Dr Frank Y. Wong for his editorial assistance.
Appendix A. Supporting Information
Additional Supporting Information may be found in the online
version of this article. Data S1, at http://dx.doi.org/10.1016/j.
cmi.2015.11.007.
Funding
This work was supported by the Natural Science Foundation of
China (grant numbers 81373062, 81072345, 30671880), the
National Science and Technology Major Projects on Infectious
Diseases of China Ministry of Health and Ministry of Science &
Technology (grant numbers 2008ZX10001-003, 2008ZX10001-
016, 2012ZX10001001-004, 2012ZX10001007-006), and
Shanghai Municipal Bureau of Health (grant numbers
XBR2011043, 12GWZX0101).
CMI
Transparency Declaration
The authors declare no potential conflicts of interest.
References
[1] INSIGHT START Study Group, Lundgren JD, Babiker AG, Gordin F,
Emery S, Grund B, Sharma S, et al. Initiation of ART in early asymp-
tomatic HIV infection. N Engl J Med 2015;373:795–807.
[2] Collaboration of Observational HIV Epidemiological Research Europe
(COHERE) in EuroCoord, Lewden C, Bouteloup V, De Wit S, Sabin C,
Mocroft A, Wasmuth JC, et al. All-cause mortality in treated HIV-
infected adults with CD4 500/mm3 compared with the general
population: evidence from a large European observational cohort
collaboration. Int J Epidemiol 2012;41:433–45.
[3] Kaufmann GR, Zaunders JJ, Cunningham P, Kelleher AD, Grey P,
Smith D, et al. Rapid restoration of CD4 T cell subsets in subjects
receiving antiretroviral therapy during primary HIV-1 infection. AIDS
2000;14:2643–51.
[4] Le T, Wright EJ, Smith DM, He W, Catano G, Okulicz JF, et al.
Enhanced CD4+ T-cell recovery with early HIV-1 antiretroviral ther-
apy. N Engl J Med 2013;368:218–30.
[5] Fiebig EW, Wright DJ, Rawal BD, Garrett PE, Schumacher RT,
Peddada L, et al. Dynamics of HIV viremia and antibody seroconversion
in plasma donors: implications for diagnosis and staging of primary HIV
infection. AIDS 2003;17:1871–9.
[6] Hargrove JW, Humphrey JH, Mutasa K, Parekh BS, McDougal JS,
Ntozini R, et al. Improved HIV-1 incidence estimates using the BED
capture enzyme immunoassay. AIDS 2008;22:511–8.
[7] Parekh BS, Kennedy MS, Dobbs T, Pau CP, Byers R, Green T, et al.
Quantitative detection of increasing HIV type 1 antibodies after
seroconversion: a simple assay for detecting recent HIV infection and
estimating incidence. AIDS Res Hum Retroviruses 2002;18:295–307.
[8] Laeyendecker O, Piwowar-Manning E, Fiamma A, Kulich M, Donnell D,
Bassuk D, et al. Estimation of HIV incidence in a large, community-
based, randomized clinical trial: NIMH project accept (HIV Preven-
tion Trials Network 043). PLoS One 2013;8:e68349.
[9] Yan H, Yu H, Xing W, Xiao Y, Zhang H, Pei L, et al. Development of a
proficiency testing program for the HIV-1BED incidence assay in
China. Sci Rep 2014;4:4512.
Clinical Microbiology and Infection © 2015 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved, CMI, 22, 290.e5–290.e8
Ding et al. Timing of ART initiation and CD4 recovery 290.e8
[10] Babiker A, Darbyshire J, Pezzotti P, Porter K, Prins M, Sabin C, et al.
CASCADE Collaboration. Short-term CD4 cell response after highly
active antiretroviral therapy initiated at different times from sero-
conversion in 1,500 seroconverters. J Acquir Immune Defic Syndr
2003;32:303–10.
[11] Hecht F, Hartogensis W, Saha S, Little S, Markowitz M, Margolick J,
et al. Long-term effects on CD4 Counts of initiating ART in very early
vs. chronic HIV. Acute Infection and Early Disease Research Program;
2013. http://ari.ucsf.edu/science/posters_2013/croi2013/hecht_croi13.
pdf. Accessed 30 May 2015.
[12] Takuva S, Maskew M, Brennan AT, Long L, Sanne I, Fox MP. Poor CD4
recovery and risk of subsequent progression to AIDS or death despite
viral suppression in a South African cohort. J Int AIDS Soc 2014;17:
18651.
[13] Lewden C, Chene G, Morlat P, Raffi F, Dupon M, Dellamonica P, et al.,
Agence Nationale de Recherches sur le Sida et les Hepatites Virales
(ANRS) CO8 APROCO-COPILOTE Study Group; Agence Nationale
de Recherches sur le Sida et les Hepatites Virales (ANRS) CO3
AQUITAINE Study Group. HIV infected adults with a CD4 cell count
greater than 500 cells/mm3 on long-term combination antiretroviral
therapy reach same mortality rates as the general population. J Acquir
Immune Defic Syndr 2007;46:72–7.
[14] Khanna N, Opravil M, Furrer H, Cavassini M, Vernazza P,
Bernasconi E, et al. CD4+ cell count recovery in HIV type 1-infected
patients is independent of class of antiretroviral therapy. Clin Infect
Dis 2008;47:1093–101.
[15] Balestre E, Eholie SP, Lokossue A, Sow PS, Charurat M, Minga A, et al.,
the International Epidemiological Database to Evaluate AIDS (IeDEA)
West Africa Collaboration. Effect of age on immunological response in
the first year of antiretroviral therapy in HIV-1-infected adults in West
Africa. AIDS 2012;26:951–7.
[16] Vinikoor MJ, Joseph J, Mwale J, Marx MA, Goma FM, Mulenga LB, et al.
Age at antiretroviral therapy initiation predicts immune recovery,
death, and loss to follow-up among HIV-infected adults in urban
Zambia. AIDS Res Hum Retroviruses 2014;30:949–55.
[17] Korten V, Gökengin D, Fincanci M, Yildirmak T, Kes NU, Fişgin NT,
et al. Outcomes of initial antiretroviral treatment (ART) among
recently diagnosed HIV patients in HIV-TR cohort, 2011–2012. J Int
AIDS Soc 2014;17:19678.
[18] Yang J, Duan S, Wang J, Li Y, Gong Y, Xiang L, et al. Sensitivity
and specificity of BED-capture enzyme linked immunoassay in
identifying recent HIV infectors. Chin J Health Lab Tech 2013;23:
114–7.