Virological suppression is associated with reduction in transmission of HIV infection at both population and individual levels [1,2]. At the individual level, treatment as prevention succeeds by reducing the duration of infectiousness, which is made up of the time between HIV infection and diagnosis and between diagnosis and virological suppression.
The main intervention to reduce the time between HIV infection and diagnosis is frequent testing in the at-risk population. In Australia, where HIV is most commonly transmitted between MSM, the frequency of testing in this risk group has been only gradually increasing [3,4]. Estimation of the time to diagnosis, at the jurisdiction level, is complex because the testing history of the individual is often not part of the notification record . The main intervention to reduce the time between HIV diagnosis and virological suppression is early initiation of antiretroviral therapy (ART). Recent changes in clinical guidelines, recommending treatment in all patients regardless of CD4+ cell count or viral load, have led to treatment being offered at an earlier stage [6,7].
Melbourne Sexual Health Centre (MSHC) in Victoria, Australia, is a large publicly funded sexual health service which performs HIV and sexual health screening for more than 11 000 MSM/year, makes approximately a third of new HIV diagnoses in MSM in the state, and has the state's largest HIV treatment center embedded within it [8–10].
We wanted to examine the combined effect of early diagnosis of HIV infection and early initiation of treatment on the duration of infectiousness, the period of time during which HIV could be potentially transmitted. The aim of this study was to examine the time from HIV infection to virological suppression, based on data currently not available at the jurisdictional level, and to determine whether it has changed over time or is associated with clinical, behaviour, or demographic predictors.
We undertook a retrospective study of the records of MSM newly diagnosed with HIV infection at MSHC from January 2007 until March 2016. Patients were included if they were men, reported sexual contact with other men, had their first ever positive HIV test result at MSHC between 1 January 2007 and 31 March 2016, and had no record of previously having had a positive HIV test.
MSHC uses a customized electronic medical record [Clinical Patient Management System (CPMS)], which collects demographic and clinical information on all patients. All patients complete a self-administered computer-assisted self-interview on sexual history such as number and sex of sexual partners, condom use, IDU, and prior HIV testing. The Melbourne Diagnostic Unit Public Health Laboratory (University of Melbourne) provides onsite laboratory services for microbiological specimens for Neisseria gonorrhoeae [nucleic acid amplification testing (NAAT) and culture] and Chlamydia trachomatis (NAAT). The Victorian Infectious Diseases Reference Laboratory is contracted to perform all off-site laboratory testing, including HIV and syphilis serology, HIV viral load testing, and CD4+ cell counts.
We extracted the following data for the presentation at which HIV was first diagnosed: age, sex, country of birth, year of arrival in Australia, history of IDU, number and sex of sexual partners, and condom use in the past 3 months, and the patient's report of the date of their most recent previous HIV test. Data provided from Melbourne Diagnostic Unit included results of testing for gonorrhea and chlamydia at the consultation in which HIV was first diagnosed. Data provided by Victorian Infectious Diseases Reference Laboratory included prior HIV and syphilis serology and the HIV testing results (screening and confirmatory test) resulting in the diagnosis of HIV infection. HIV viral load and CD4+ cell count results were extracted up until March 2017, so that all patients had at least 1 year of follow-up after diagnosis.
Syphilis serology was performed using rapid plasma reagin (RPR) (Macro-Vue RPR card; Becton Dickinson BD Microbiology Systems, Sparks, Maryland, USA), Treponema pallidum particle agglutination assay (TPPA) (Serodia TPPA, Fujirebio Inc., Tokyo, Japan), a recombinant total antibody enzyme-linked immunosorbent assay (EIA) (Trepanostika TP recombinant EIA; BioMerieux, Marcy l’Etoile, France). From January 2016 the BioMerieux EIA was replaced by an automated chemiluminescence immunoassay (CLIA) (LIASON Treponema screen; Diasorin, Saluggia, Italy). Prior to March 2015, chlamydia NAAT testing was performed using the Becton Dickinson ProbeTec strand displacement amplification assay and gonorrhea testing was performed by microscopy and culture (direct inoculation on selective and nonselective media and incubation in CO2-enriched atmosphere at 37°C). From March 2015, NAAT testing for both gonorrhea and chlamydia were performed using the Aptima Combo 2 (Hologic Gen-Probe, San Diego, California, USA) transmission-mediated amplification assay.
Following Australian guidelines, HIV is diagnosed using a two-step algorithm: a screening assay which, if reactive, is followed by a confirmatory western blot assay [11,12]. From the beginning of the study period until 29 April 2014 the screening assay was the Abbott Murex HIV-1 2.0 EIA (third generation). From 29 April 2014 until the end of the study period the DiaSorin LIASON XL Murex HIV Ab/Ag CLIA (fourth generation) assay was used as the screening assay. The MP Diagnostics HIV1/2 Blot 2.0 western blot assay was used as the confirmatory test for the entire study period. Until November 2013, the western blot was reported as positive if there were one or more glycoprotein bands and three or more viral protein-specific bands. After that date, the western blot was reported as positive if there were two or more glycoprotein bands and three or more viral protein-specific bands. The assay was reported as indeterminate if specific viral bands were present but the above criteria for a positive result were not met. If no viral-specific bands were present the assay was reported as negative. All indeterminate western blot results after November 2013 were manually reviewed to determine if the change in reporting criteria led to a change in reported result.
After HIV infection, the western blot evolves from negative to indeterminate to positive over a period of up to 90 days . We defined serological evidence of recent HIV infection as a reactive screening test with a negative or indeterminate western blot in a patient in whom HIV diagnosis was confirmed through ancillary or repeat testing. No serological evidence of recent HIV infection was defined as a reactive screening test with a positive western blot.
We used the method of midpoint imputation to assign a date of HIV infection in individuals with either serological evidence of recent infection or a known date of previous negative HIV test . In this method, the individual is inferred to have been infected midway between their most recent negative and their first positive HIV test. However, serological evidence of recent infection provides additional information for this calculation. For individuals with no serological evidence of recent infection, it is unlikely that they were infected within the past 90 days; the date of infection was imputed as the midpoint between the most recent negative test and 90 days prior to diagnosis. For individuals with serological evidence of recent infection, whose previous HIV test was greater than 90 days prior or unknown, the date of HIV infection was imputed as 45 days prior to diagnosis. In patients with serological evidence of recent infection and a negative test within the last 90 days, the date of infection was imputed as the midpoint between the two tests. In any patients with no serological evidence of recent infection but with a negative test within the past 90 days, the date of HIV infection was imputed as the date of the previous HIV test. For those with no previous known test, a duration of HIV infection based on the first CD4+ cell count was imputed using a method previously described by Jansson et al..
We defined the time to diagnosis as the time in months between the imputed date of infection and the date of the diagnosis of HIV infection. We felt that it was important to include the CD4+-based calculation on patients with unknown or no prior HIV test because they may have been infected for a longer period prior to diagnosis. However, to avoid bias because of potential insensitivity of that method, we conducted all analyses both with and without these individuals included in the data set. We defined the time to suppression as the time in months between the date of diagnosis and the date of the first ever viral load test less than 200 copies/ml after initiation of ART, in those patients in whom any virological data were available (event). Subsequent virological rebounds in these patients were not examined. Those patients who never commenced ART or never achieved virological suppression to less than 200 copies/ml were censored at the date of their most recent viral load (censor). We defined the duration of infectiousness as the sum of the time to diagnosis and time to suppression.
Factors that were associated with the odds of diagnosis with serological evidence of recent infection were examined using logistic regression. Factors associated with time to suppression and duration of infectiousness were examined using Cox regression. The following variables were examined in each: time period of diagnosis (2007–2009, 2010–2012, 2013–2014, 2015 to March 2016), age, history of injecting, having been born in a non-English speaking country and arrival in Australia within the last 3 years, symptomatic gonorrhea, chlamydia, or syphilis at the time of HIV diagnosis, any diagnosis of gonorrhea or chlamydia at time of HIV diagnosis (symptomatic or asymptomatic), diagnosis of recent syphilis infection at time of HIV diagnosis (as defined by rise in RPR of greater than four times most recent previous RPR in patients previously treated for syphilis, seroconversion from negative to positive syphilis serology or, in patients with no prior syphilis serology, clinical evidence of primary or secondary syphilis and/or RPR greater than 1 : 16), sexual behaviour in the past 3 months categorized as low risk (consistent condom use or no reported sexual partners), high risk (inconsistent condom use and one to three reported sexual partners), or very high risk (inconsistent condom use and four or more reported sexual partners). For individuals born in a non-English speaking country, we chose to examine those who had arrived within 3 years because most temporary or early visa holders in Australia are not eligible for the universal health insurance scheme, Medicare, during that time. This culturally and linguistically diverse group is of particular concern for healthcare access in Australia . Kaplan–Meier survival curves and median survival time were also generated.
All analyses were conducted using the STATA Statistics software package (version 14: Stata Corp., College Station, Texas, USA). This study was approved by the Alfred Hospital Ethics Committee (approval number 130/16).
In total, 437 MSM newly diagnosed with HIV infection at MSHC between January 2007 and March 2016 were included. The median age of participants was 30.3 years. Of these 437 men, 30.9% were also diagnosed with gonorrhea, chlamydia, or recently acquired syphilis on that day. In total, 31.1% reported inconsistent condom use and more than three partners in the past 3 months (classified as very high risk). The proportion of patients from non-English speaking countries who had arrived in Australia within the past 3 years increased from 11.0% in 2007–2009 to 26.0% in 2015–2016 (P = 0.007). The proportion of patients diagnosed with serological evidence of recent HIV infection increased from 15.6 to 34.3% (P < 0.0001). Those patients had a higher CD4+ cell count and viral load than those diagnosed with no serological evidence of recent infection (P < 0.0001; See Table 1).
After the introduction of the fourth generation screening assay in April 2014, six patients were antibody negative and antigen positive at diagnosis, whose diagnosis may have been delayed if the third generation assay was used. Every patient with reported indeterminate western blot after November 2013, when there was a change in reporting, also had an indeterminate western blot when the previous criteria were applied.
During the study period, the median time to diagnosis in patients with a known date of prior negative HIV or with serological evidence of recent infection fell significantly from 5.4 [interquartile range (IQR) 3.3–17.1] to 4.2 months (IQR 1.5–9.3, P = 0.0054). After inclusion of 48 patients without history of previous testing or serological evidence of recent infection (in whom time to diagnosis was based on CD4+ cell count), the median time to diagnosis fell from 6.8 (IQR 3.2–21.2) to 4.3 months (IQR 1.5–9.6, P = 0.0004). The median time to suppression fell from 22.7 (IQR 10.5–47.0) to 3.2 months (IQR 2.3–5.9, P < 0.0001). The median duration of infectiousness declined from 49.0 months (IQR 30.3–73.5) for patients diagnosed between 2007 and 2009 to 9.6 months (IQR 5.7–24.3) for those diagnosed between 2015 and 2016 (P < 0.0001; See Table 2 and Fig. 1). Kaplan–Meier survival curves are in Fig. 2a and b.
In the multivariate analyses, only being diagnosed with HIV in a more recent time period was associated with a higher odds of being diagnosed with serological evidence of recent infection, a shorter time to suppression and a shorter duration of infectiousness (Table 3). In the subanalysis excluding 48 patients with no known date of previous HIV testing and no serological evidence of recent infection, more recent time period of HIV diagnosis, younger age, and higher sexual risk were all associated with a shorter duration of infectiousness. Recently acquired syphilis was associated with a longer duration of infectiousness. See Supplementary Table 1, http://links.lww.com/QAD/B154.
The median estimated duration of infectiousness (time between HIV infection and virological suppression) among MSM at a sexual health service in Melbourne, Australia fell more than five-fold from 49.0 months for patients diagnosed between 2007 and 2009 to 9.6 months in those diagnosed between 2015 and March 2016. This is the first study to examine the combined effect of early diagnosis of HIV infection and early initiation of ART using data on prior HIV testing history, serological evidence of recent HIV infection, and monitoring of viral load after diagnosis. If replicated across the community, this reduction in the period of infectiousness could be expected to have a substantial impact on HIV transmission and incidence.
What have been the drivers of this change? The reduction in time to suppression has been the greater contributor. From Fig. 2a, we can see that proportion of individuals virologically suppressed within 3 months of diagnosis rose from less than 5% to almost 50% over the time period of our study. Several factors will have influenced this dramatic decline in time to viral suppression. First, awareness of the effect of HIV treatment on transmission has steadily grown since the release of the HIV Prevention Trials Network 052 study in 2011 and other subsequent studies [2,17–19]. Second, after the release of the strategic timing of antiretroviral treatment (START) and Early Antiretrovirals and Isoniazid Preventive Therapy in Africa (TEMPRANO) studies in 2015, international and local treatment recommendations changed to include the recommendation to treat all [6,20–22]. Third, in addition to earlier initiation of ART, integrase strand transfer inhibitor-based regimens are known to result in earlier virological suppression . Use of integrase strand transfer inhibitors has grown steadily since raltegravir was first recommended in first-line treatment in 2009 .
Factors which may have contributed to the fall in time from infection to diagnosis are less clear. A study of testing frequency in MSM at the MSHC showed only small increases between 2003 and 2013, although a greater increase was observed at the end of that study period and may have continued after it . The fourth generation HIV antibody/antigen screening immunoassay introduced in April 2014 detected very early HIV infection that may otherwise have been delayed in six individuals, five of which occurred in the most recent time period. It is possible, however, that individuals are driven to earlier testing by other factors including either perception of very recent increased risk or awareness of symptoms of acute HIV infection, which our study was not able to capture.
Although the fall in time to diagnosis was less in absolute terms than the fall in time from diagnosis to suppression, it was both statistically and clinically significant. It is widely recognized that secondary transmission of HIV is more likely to occur in the first weeks after infection, although the magnitude of that increase and its effect on the epidemic are still debated [24,25]. More than one-third of new diagnoses at this center are now within the acute phase of infection. These individuals have higher viral loads and are very sexually active (see Table 1). Although few patients achieved virological suppression within 3 months of infection, research elsewhere in urban MSM populations has demonstrated a change in sexual behaviour after diagnosis to less condomless anal intercourse with men of negative or unknown HIV status .
Effective control of HIV depends on universal access to testing and treatment including marginalized groups and the uninsured. In response, MSHC has a high volume, highly accessible model of testing that does not require advance appointments, payments, or insurance coverage. IDUs, and MSM newly arrived from non-English speaking countries, who are unlikely to be eligible for Medicare, Australia's universal health insurance scheme, were no less likely to be diagnosed with recent infection or to wait longer to achieve virological suppression. This is significant because these groups have been identified as likely to experience delayed care .
About one-third of HIV diagnoses in MSM in the state of Victoria are made at MSHC [8,10]. If replicated across the community, the reduction in duration of infectiousness and the increase in the proportion diagnosed with recent infection can be expected to have had a profound influence on HIV incidence. Additionally, a government program now provides pre-exposure prophylaxis free of charge for 3600 individuals in the state. A reduction in both the number of infectious and susceptible individuals can be expected to be complimentary or synergistic.
Our study is subject to certain limitations. First, it is not possible to be certain when a person has been infected with HIV prior to being diagnosed. However, midpoint imputation is an accepted method for converting testing history into estimated duration of infection and assumes that the individual has been at a constant level of risk since they tested negative. Second, we used the baseline CD4+ cell count to estimate the duration of infection in 48 (11%) patients of the population of 437 with no other data from which to impute it. Either inclusion or exclusion of this group of patients and the different method used to impute duration of infection in them has the potential to bias the results of the study. The median time to diagnosis was longer when they were included, suggesting that these patients had indeed been infected for a longer period of time than regularly tested individuals. Also the number of individuals without prior testing fell from 11.9 to 4.3% over the study period. Results of analyses including and excluding this group were qualitatively similar and unopposite.
The duration of infectiousness in MSM diagnosed with HIV infection at MSHC in Victoria has fallen dramatically between 2007 and 2016 and the proportion diagnosed with serological evidence of recent infection has significantly increased. This effect is across all population subgroups and marks a very positive milestone for the treatment as prevention paradigm.
N.A.M.: study conceptualization, data curation, formal analysis, investigation, methodology, project administration, validation, visualization, writing – original draft preparation and review and editing.
S.N.: Methodology, data curation, interpretation, and presentation of data, writing – review and editing.
E.P.F.C.: Formal analysis and writing – review and editing.
T.R.H.R.: C.S.B.: Writing – review and editing.
I.D.: Data curation, writing – review and edition.
C.K.F.: Supervision, formal analysis, and writing – review and editing.
The work was supported by the Australian National Health and Medical Research Council (PGS 1094518 to N.A.M., ECF 1091226 to E.P.F.C., and ECF 1091536 to T.R.H.R.). The other authors were supported in this work through their institutional salaries, only.
N.A.M., E.P.F.C., and T.R.H.R. received funding from the Australian Government National Health and Medical Research Council.
Conflicts of interest
There are no conflicts of interest.
1. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. HPTN 052 Study Team. Antiretroviral therapy for the prevention of HIV-1 transmission
. N Engl J Med
2. Tanser F, Bärnighausen T, Grapsa E, Zaidi J, Newell ML. High coverage of ART associated with decline in risk of HIV acquisition in rural KwaZulu-Natal, South Africa
3. The Kirby Institute. HIV, viral heptatitis and sexually transmissive infection in Australia Annual Surveillance Report 2016
. New South Wales: The Kirby Institute; 2016.
4. Lin AC, Fairley CK, Dutt K, Klassen KM, Chen MY, Fehler G, et al. Testing for HIV among men who have sex with men needs a paradigm shift in Australia, given the minimal increase between 2003 and 2013 in Melbourne, Australia
. Sex Health
5. Supervie V, Marty L, Lacombe JM, Dray-Spira R, Costagliola D. FHDH-ANRS CO4 study group. Looking Beyond the Cascade of HIV Care to End the AIDS Epidemic: Estimation of the Time Interval From HIV Infection to Viral Suppression
. J Acquir Immune Defic Syndr
6. Lundgren JD, Babiker AG, Gordin F, Emery S, Grund B, Sharma S, et al. INSIGHT START Study Group. Initiation of antiretroviral therapy in early asymptomatic HIV infection
. N Engl J Med
7. De Cock KM, El-Sadr WM. From START to finish: implications of the START study
. Lancet Infect Dis
8. The Centre for Population Health Burnet Institute, HIV/AIDS Monthly Surveillance Report December 2016. Melbourne, Australia: Victoria State Government Department of Health and Human Services, 2017.
9. McMahon JH, Moore R, Eu B, Tee BK, Chen M, El-Hayek C, et al. Victorian Initiative for Patient Engagement and Retention VIPER Study Group. Clinic network collaboration and patient tracing to maximize retention in HIV care
. PLoS One
10. Melbourne Sexual Health Centre Annual Report 2016. 2017
11. National Pathology Accreditation Advisory Council, Requirements for Laboratory Testing for Human Immunodeficiency Virus (HIV) and Hepatitis C Virus HCV Canberra, Australia: Australian Government of Health; 2013.
12. Australasian Society of HIV Medicine Expert Reference Committee, National HIV Testing Policy. 2017
13. Hecht FM, Wellman R, Busch MP, Pilcher CD, Norris PJ, Margolick JB, et al. Acute Infection Early Disease Research Program. Identifying the early post-HIV antibody seroconversion period
. J Infect Dis
14. Geskus RB. Methods for estimating the AIDS incubation time distribution when date of seroconversion is censored
. Stat Med
15. Jansson J, Kerr CC, Mallitt KA, Wu J, Gray RT, Wilson DP. Inferring HIV incidence from case surveillance with CD4+ cell counts
16. Medland NA, Mao L, Crooks L, de Wit J. Obstacles to prescribers’ initiation early antiretroviral therapy: a barrier to achieving 90-90-90 goals
. Lancet HIV
17. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Prevention of HIV-1 infection with early antiretroviral therapy
. N Engl J Med
18. Grulich A.E., et al. HIV Transmission in male serodiscordant couples in Australia, Thailand and Brazil,Conferences on Retroviruses and Opportunistic Infections
. 2016: Boston
19. Rodger A., et al. HIV transmission risk through condomless sex if HIV+ partner on suppressive ART: PARTNER Study
, Conferences on Retroviruses and Opportunistic Infections
. 2016: Boston
20. Danel C, Moh R, Gabillard D, Badje A, Le Carrou J, Ouassa T, et al. TEMPRANO Study Group. A Trial of early antiretrovirals and isoniazid preventive therapy in Africa
. N Engl J Med
21. Panel on Antiretroviral Guidelines for Adults and Adolescents, Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents
. 2015, Department of Health and Human Services
22. Australasian Society of HIV Medicine. Antiretroviral Guidelines: US DHSS Guidelines with Australian Commentary. 2015 [cited 05 August 2017]; Available from: http://arv.ashm.org.au/
[Accessed 28 August 2017]
23. Messiaen P, Wensing AM, Fun A, Nijhuis M, Brusselaers N, Vandekerckhove L. Clinical use of HIV integrase inhibitors: a systematic review and meta-analysis
. PLoS One
24. Abu-Raddad LJ. Role of acute HIV infection in driving HIV transmission: implications for HIV treatment as prevention
. PLoS Med
25. Cohen MS, Dye C, Fraser C, Miller WC, Powers KA, Williams BG. HIV treatment as prevention: debate and commentary: will early infection compromise treatment-as-prevention strategies?
. PLoS Med
26. Khosropour CM, Dombrowski JC, Kerani RP, Katz DA, Barbee LA, Golden MR, et al. Changes in condomless sex and serosorting among men who have sex with men after HIV diagnosis
. J Acquir Immune Defic Syndr