Decreased Time to Viral Suppression After Implementation of Targeted Testing and Immediate Initiation of Treatment of Acute Human Immunodeficiency Virus Infection Among Men Who Have Sex With Men in Amsterdam

Abstract

Background

Men who have sex with men (MSM) with acute human immunodeficiency virus (HIV) infection (AHI) are a key source of new infections. To curb transmission, we implemented a strategy for rapid AHI diagnosis and immediate initiation of combination antiretroviral therapy (cART) in Amsterdam MSM. We assessed its effectiveness in diagnosing AHI and decreasing the time to viral suppression.

Methods

We included 63 278 HIV testing visits in 2008–2017, during which 1013 MSM were diagnosed. Standard of care (SOC) included HIV diagnosis confirmation in < 1 week and cART initiation in < 1 month. The AHI strategy comprised same-visit diagnosis confirmation and immediate cART. Time from diagnosis to viral suppression was assessed for 3 cART initiation periods: (1) 2008–2011: cART initiation if CD4 < 500 cells/μL (SOC); (2) January 2012–July 2015: cART initiation if CD4 < 500 cells/μL, or if AHI or early HIV infection (SOC); and (3a) August 2015–June 2017: universal cART initiation (SOC) or (3b) August 2015–June 2017 (the AHI strategy).

Results

Before implementation of the AHI strategy, the proportion of AHI among HIV diagnoses was 0.6% (5/876); after implementation this was 11.0% (15/137). Median time (in days) to viral suppression during periods 1, 2, 3a, and 3b was 584 (interquartile range [IQR], 267–1065), 230 (IQR, 132–480), 95 (IQR, 63–136), and 55 (IQR, 31–72), respectively (P < .001).

Conclusions

Implementing the AHI strategy was successful in diagnosing AHI and significantly decreasing the time between HIV diagnosis and viral suppression.

acute HIV infection, men who have sex with men, viral suppression, targeted testing, same-day start

The estimated number of new human immunodeficiency virus (HIV) infections among men who have sex with men (MSM) in Amsterdam has continuously but slowly decreased since 2007 [1]. In 2016, a phylogenetic modeling analysis among MSM in the Netherlands estimated that 43% of onward HIV transmissions occur within 12 months of infection, and 71% from undiagnosed MSM [2]. Thus, identifying MSM with HIV as early as possible, and immediately initiating combination antiretroviral therapy (cART), is crucial to significantly reduce new HIV infections [3–5].

Acute HIV infection (AHI) can be diagnosed with HIV RNA and fourth-generation antigen/antibody (Ag/Ab) tests [6]. In most places, standard HIV testing occurs with rapid antibody tests or Ag/Ab tests. Confirmation of diagnosis and referral to initiate cART usually comprises at least several days. Therefore, an optimally sensitive testing approach, particularly to be able to rapidly detect and confirm the earliest infection stages, linked to same-day cART initiation, demands additional efforts. AHI awareness campaigns, targeted screening for AHI, and HIV RNA testing have been shown to contribute to earlier diagnosis [7–10], and a strategy combining these approaches may substantially decrease HIV incidence.

Several studies in countries with limited healthcare access have shown that early cART initiation improves retention in care and viral suppression rates [11–24]. However, the majority of these studies included people diagnosed with established HIV infection [12–15, 17–19]. The studies that focus specifically on immediate cART initiation among MSM with AHI are scarce [16, 21–24], and some lack comparison groups not offered immediate cART. None of these studies provided AHI diagnosis at the point of care, and they did not focus on mobilizing MSM for AHI testing.

To the best of our knowledge, there are no studies on the feasibility and effectiveness of a combined strategy to increase awareness of AHI and to offer targeted testing and point-of-care diagnosis linked to same-visit referral and immediate cART initiation. It remains unclear whether such a strategy would be of additional value in countries like the Netherlands with universal healthcare. We implemented a strategy for rapid diagnosis of AHI in MSM using point-of-care HIV RNA testing at the Amsterdam sexually transmitted infection (STI) clinic as of August 2015. This was linked to immediate cART initiation, along with a media campaign to increase awareness of AHI among MSM [25]. Amsterdam has 1 large public STI clinic, which performs > 50 000 annual consultations and accounts for approximately 46% of new HIV diagnoses among MSM in Amsterdam (van Sighem, personal communication). The objective of this study was to assess the feasibility and effectiveness of a strategy in diagnosing AHI and decreasing the time to viral suppression in Amsterdam, and historically relating this to changes in the time between diagnosis, cART initiation, and viral suppression during prior calendar periods reflective of cART initiation guidelines.

METHODS

Standard Testing

The standard HIV testing policy was as follows: From January 2008 through April 2014, MSM were tested with a rapid antibody test (Alere Determine HIV 1/2, Alere Inc), and a combined Ag/Ab test if AHI was clinically suspected (through April 2013: AxSYM, Abbott Laboratories; from May 2013: LIAISON XL Murex, Diasorin). From May 2014, rapid antibody and Ag/Ab tests were used. Positive or indeterminate samples were confirmed with line immunoassay (Inno-Lia HIV I/II Score; Fujirebio). Rapid antibody results were delivered at the same visit, and Ag/Ab results within 1 week. MSM with positive or indeterminate test results returned to the STI clinic 1 week after their initial visit for confirmation with a rapid test, were subsequently referred to an HIV treatment center and had an intake at an HIV treatment center within 5 business days.

Staging of Infection

In line with prior publications, we defined AHI as Fiebig stages I–II, early HIV infection as Fiebig stages III–V, and acute or early HIV infection (AEHI) as Fiebig stages I–V (Supplementary Table 1) [6, 26–28].

Standard cART Initiation

We considered 3 historical standard of care (SOC) cART initiation periods: (1) In 2008–2011, Dutch guidelines recommended to initiate cART if CD4 < 500 cells/μL; (2) in 2012–2014, cART initiation was recommended if CD4 < 500 cells/μL or irrespective of CD4 count in case of AEHI; and (3) since 2015, guidelines recommend cART initiation as soon as HIV genotype results are available, irrespective of CD4 count or stage of infection, and first-line treatment with integrase strand transfer inhibitor (INSTI)–based regimens [29].

AHI Search and Treat-to-Suppression Strategy

In August 2015, we launched a media campaign promoting an online intervention specifically designed to increase AHI detection and immediately treat individuals with AHI (ie, “AHI search and treat-to-suppression strategy” or in short, the “AHI strategy”). The online intervention aimed to increase awareness for AHI among Amsterdam MSM, help them self-identify AHI symptoms, and increase their behavioral motivation to test [25]. This included a dedicated website (www.hebikhiv.nl/en) with a self-administered screening tool based on a pragmatic AHI risk score [30]. MSM who reported condomless anal intercourse in the previous 3 months and AHI symptoms could download a referral letter to the AHI strategy. Next to this online referral, MSM could also be referred by their general practitioner or during routine STI screening at the STI clinic; eligibility was assessed with the pragmatic AHI risk score at health providers’ discretion [30]. Although the campaign was specifically designed for the AHI strategy, wider dissemination took place, and MSM in SOC and health providers likely have also been exposed to the campaign part of the strategy.

The AHI strategy’s testing procedures included simultaneous rapid antibody (Alere Determine HIV 1/2, Alere Inc), Ag/Ab (LIAISON XL Murex, Diasorin), and qualitative point-of-care HIV RNA testing (GeneXpert, Cepheid), and same-visit delivery of results. MSM identified with AEHI were immediately referred to an HIV treatment center to initiate cART within 24 hours. To summarize, SOC procedures included diagnosis confirmation within 1 week and referral to an HIV treatment center within 5 days of confirmed diagnosis, and cART initiation varied per period (within 1 month of referral for period 3, 2015–2017). AHI strategy procedures included same-visit confirmation of diagnosis, same-day referral to an HIV treatment center, and immediate cART initiation.

Data Linkage

The STI clinic and HIV treatment centers have separate databases. To assess time to viral suppression, we aimed to link records from the STI clinic to medical records from HIV treatment centers, retrieved from the ongoing AIDS Therapy Evaluation in the Netherlands (ATHENA) National HIV cohort [31]. Clinical data in the ATHENA cohort are collected prospectively and encompass data of 98% of all people living with HIV in care in the Netherlands. A trusted third party (ZorgTTP, Utrecht) conducted the linkage based on date of birth and postal code. The Medical Ethics Committee of Amsterdam University Medical Center approved this study (W17_248 number 17.290).

Statistical Analyses

We assessed AHI as a proportion of HIV testing visits yielding a diagnosis of AHI (AHI yield), and as the proportion of AHI among all HIV diagnoses in MSM at the STI clinic, before (January 2008–July 2015) and after (August 2015–June 2017) implementation of the AHI strategy. The latter period was further stratified by testing policy: MSM diagnosed via standard testing vs the AHI strategy.

Time from diagnosis to viral suppression was defined as the number of days between the first STI clinic visit with a positive or indeterminate test result and the first HIV treatment center visit with a viral load < 50 copies/mL. Follow-up HIV treatment center visits were included up to 2 January 2018; data were censored on that date, or on the last HIV treatment center visit date if MSM were lost to follow-up (LTFU). Virological failure was defined as 2 consecutive measures of a plasma viral load ≥ 200 copies/mL within 6 months after viral suppression [32].

We assessed time from diagnosis to intake at an HIV treatment center, cART initiation, and viral suppression for the following periods: (1) 2008–2011: cART if CD4 < 500 cells/μL; (2) 2012–July 2015: cART if CD4 < 500 cells/μL or if AEHI; and (3a) August 2015–June 2017: universal cART in SOC, or (3b) August 2015–June 2017: the AHI strategy. We estimated median time-to-event with the Kaplan-Meier method, and compared periods with the log-rank test. Two-sided P values < .05 were considered statistically significant. Analyses were performed using Stata software (version 15.1; StataCorp).

RESULTS

Trends in AHI Diagnosis Over Time

Between 2008 and 2017, 19 728 MSM without prior known HIV infection visited the STI clinic, and during 63 278 visits, 1013 MSM were diagnosed. The AHI yield per visit was 0.01% (5/47 008) before and 0.09% (15/16 270) after implementation of the AHI strategy (Table 1). When we stratified for testing policy during the period 2015–2017, AHI yield was 0.03% (5/16 021) via standard testing and 4.0% (10/249) via the AHI strategy. The proportion of AHI among HIV diagnoses was 0.6% (5/876) before and 11.0% (15/137) after implementation of the AHI strategy. Stratified for testing policy (2015–2017), this was 4.2% (5/118) via standard testing and 52.6% (10/19) via the AHI strategy. After implementation of the AHI strategy, AEHI yield decreased; however, the proportion of AEHI among HIV diagnoses increased. Fiebig stages are displayed in Figure 1; referral source and test results of MSM in the AHI strategy are shown in Supplementary Figure 1. Of the 127 MSM who had been referred through the campaign’s online referral, none were diagnosed with AHI, 1 with early HIV infection, and 1 with established HIV infection. Characteristics of MSM diagnosed with HIV in 2015–2017 are shown in Supplementary Table 2. Median time from intake at the STI clinic to delivery of confirmed HIV diagnosis was 6.9 (interquartile range [IQR], 5.9–7.1) days in the standard testing setting and 3.5 (IQR, 1.7–3.8) hours in the AHI strategy; this did not differ by infection stage.

Table 1.

Human Immunodeficiency Virus (HIV) Diagnoses and Fiebig Stages Among 63 278 HIV Testing Visits of 19 728 Unique Men Who Have Sex With Men at the Amsterdam Sexually Transmitted Infection Clinic, Before and After Implementation of the Acute HIV Infection Search and Treat-to-Suppression Strategy

Characteristic	Preimplementation Standard	Postimplementation AHI Strategy + Standard	Postimplementation Standard	Postimplementation AHI strategy
Period of diagnosis	January 2008–July 2015	August 2015–June 2017	August 2015– June 2017	August 2015–June 2017
Diagnostic strategy
Point-of-care RNA test	No	If eligiblea	No	If eligiblea
Fourth-generation Ag/Ab test	If symptoms of AHIb/Yesc	Yes	Yes	Yes
Rapid antibody test	Yes	Yes	Yes	Yes
Visits, No.	47 008	16 270	16 021	249
Diagnosed with HIV, no.	876	137	118	19
HIV yieldd	1.9%	0.8%	0.7%	7.6%
Diagnosed with AHI,e no.	5	15	5	10
AHI yieldf	0.01%	0.09%	0.03%	4.0%
Proportion of AHI among diagnosed with HIV	0.6%	11.0%	4.2%	52.6%
Diagnosed with AEHI,g no.	237	52	35	17
AEHI yieldh	0.5%	0.3%	0.2%	6.8%
Proportion of AEHI among diagnosed with HIV	27.1%	38.0%	29.7%	89.5%
Fiebig stage, No.
I	NA	2	NA	2
IIi	5	13	5	8
II/IIIi	22	NA	NA	NA
IIIi	9	6	4	2
IV	19	4	2	2
V	182	27	24	3
VI	639	85	83	2

Abbreviations: AEHI, acute or early human immunodeficiency virus infection; AHI, acute human immunodeficiency virus infection; Ag/Ab, antigen/antibody test; HIV, human immunodeficiency virus; NA, not assessed.

^aEligibility was assessed using a pragmatic AHI risk score.

^bIf AHI was suspected (eg, because of AHI symptoms, through April 2014).

^cFrom May 2014 onward.

^dVisits with a new HIV diagnosis as a proportion of all visits.

^eDefined as Fiebig stages I–II.

^fVisits with an AHI diagnosis as a proportion of all visits.

^gDefined as Fiebig stages I–V.

^hVisits with an AEHI diagnosis as a proportion of all visits.

ⁱUntil May 2013, the AxSYM Ag/Ab test was used, and no distinction between Fiebig stage II and III could be made; from May 2013 onward, the LIAISON Ag/Ab test was used, allowing for distinction between Fiebig stages II and III.

Figure 1.

Fiebig (F) stages as a proportion of new human immunodeficiency virus (HIV) diagnoses for 1013 men who have sex with men diagnosed with HIV at the Amsterdam sexually transmitted infection clinic, January 2008–June 2017. A, Visits from January 2008 through June 2017 before and after implementation of the acute HIV infection (AHI) strategy; diagnostic strategies included standard testing (through April 2014: universal rapid antibody test, antigen/antibody [Ag/Ab] test if AHI was suspected; May 2014 onward: universal rapid antibody test and universal Ag/Ab test) and the AHI strategy (August 2015 onward: rapid antibody, Ag/Ab, and point-of-care HIV RNA test if eligible). Until May 2013, the AxSYM Ag/Ab test was used, and no distinction between Fiebig stages II and III could be made. From May 2013 onward, the LIAISON Ag/Ab test was used, allowing for distinction between Fiebig stages II and III. B, Visits from August 2015 through June 2017, after implementation of the AHI strategy; stratified by diagnostic strategy (standard testing and the AHI strategy).

Data Linkage

Of the 1013 STI clinic records of diagnosed MSM (2008–2017), a subset could be linked to records of HIV treatment centers and were used to assess time to viral suppression: 66.1% in period 1 (2008–2011), 65.8% in period 2 (2012–2015), 61.0% in period 3a (2015–2017; SOC), and 79.0% in 3b (2015–2017; AHI strategy) (Table 2). Compared with MSM with unlinked records, median age was higher among MSM with linked records (36 [IQR, 30–43] years vs 31 [IQR, 26–39] years), and a higher proportion of MSM were born in the Netherlands (48.6% vs 28.6%). Of the 348 records that could not be linked, 31.9% contained an invalid or no postal code, and 0.3% an invalid date of birth.

Table 2.

Time to Intake at a Human Immunodeficiency Virus (HIV) Treatment Center, Combination Antiretroviral Therapy Initiation, and Viral Suppression for 665 Men Who Have Sex With Men Diagnosed With HIV at the Amsterdam Sexually Transmitted Infection (STI) Clinic of Whom Records From the STI Clinic Were Successfully Linked With Records From HIV Clinics, January 2008–June 2017

Characteristic	(1) Standarda		(2) Standarda		(3a) Standarda		(3b) AHI Strategya		P Value
									Overallb	(3a) vs (3b)c
Period of diagnosis	January 2008–December 2011		January 2012–July 2015		August 2015–June 2017		August 2015–June 2017
cART initiation policyd	CD4 count < 500e		CD4 count < 500e or AEHI		Universal		Immediate
STI clinic record linked with HIV treatment center recordf	355	(66.1%)	223	(65.8%)	72	(61.0%)	15	(79.0%)	.80g	.13h
Baseline viral load (log10 copies/mL)i	4.6	(4.0–5.1)	4.7	(4.0–5.1)	4.7	(4.1–5.2)	6.0	(4.7–7.0)	.0034	.0019
Baseline CD4 count (cells/μL)j	455	(330–600)	520	(390–700)	480	(400–600)	480	(380–520)	.0021	.63
CD4 nadir (cells/μL)k	330	(260–410)	440	(310–570)	478	(400–600)	480	(380–520)	.0001	.79
Lost to follow-up before cART initiationl	16	(4.5%)	5	(2.2%)	5	(6.9%)	0	…	.77g	.29h
Initial cART regimenm									< .0001g	< .0001g
NNRTI use	225	(70.5%)	71	(33.2%)	0	…	0	…
PI use	52	(16.3%)	33	(15.4%)	1	(1.6%)	0	…
INSTI use	42	(13.2%)	106	(49.5%)	49	(75.6%)	3	(20.0%)
INSTI + PI usen	0	…	4	(1.9%)	14	(21.9%)	12	(80.0%)
cART initiation after diagnosiso,p									< .0001g	< .0001g
Same-day	0	…	0	…	0	…	2	(13.3%)
After 1 day	0	…	1	(0.5%)	1	(1.5%)	7	(46.7%)
From 2 through 7 days	1	(0.3%)	4	(1.8%)	11	(15.9%)	5	(33.3%)
Lost to follow-up before viral suppressionq,r	5	(1.5%)	3	(1.4%)	2	(3.0%)	0	…	.95g	.498h
Virological failures	7	(2.3%)	2	(1.0%)	1	(2.4%)	0	…	.45g	.57h
Days from diagnosis tot:
Intake at HIV clinicu	14	(9–21)	13	(9–17)	9	(6–14)	1	(0–2)	< .0001v	< .0001v
cART initiationo	439	(102–899)	99	(43–339)	29	(15–43)	1	(1–3)	< .0001v	< .0001v
Viral suppressionq	584	(267–1065)	230	(132–480)	95	(63–136)	55	(31–72)	< .0001v	.0007v
Days from cART initiation too,t:
Viral suppressionq	125	(77–172)	91	(39–152)	59	(28–97)	54	(28–66)	< .0001v	.18v
Days between viral load measurementsw	125	(78–172)	91	(39–152)	61	(28–97)	54	(28–66)	.0001	.30

Abbreviations: AEHI, acute or early human immunodeficiency virus infection; AHI, acute human immunodeficiency virus infection; cART, combination antiretroviral therapy; HIV, human immunodeficiency virus infection; INSTI, integrase strand transfer inhibitor; MSM, men who have sex with men; NNRTI, nonnucleoside reverse-transcriptase inhibitor; PI, protease inhibitor; STI; sexually transmitted infection.

^ano. (%) or median (interquartile range).

^bKruskal-Wallis rank test, unless stated otherwise.

^cWilcoxon rank-sum test, unless stated otherwise.

^dIn routine strategies, cART was initiated as per national treatment guidelines; men who have sex with men (MSM) diagnosed in the AHI strategy were offered cART within 24 hours of diagnosis.

^eIn cells per microliter.

^fProportion of diagnosed MSM at Amsterdam STI clinic.

^gNonparametric trend test.

^hχ ² test.

ⁱTwo missing values.

^jTwenty-one missing values.

^kThirteen missing values.

^lProportion of STI clinic records that were linked to HIV clinic records.

^mTwenty-seven missing values, proportion of MSM who were not lost to follow-up before cART initiation.

ⁿMSM diagnosed in acute or early infection stage were offered to immediately start with a 4-drug cART regimen (ritonavir-boosted darunavir, emtricitabine/tenofovir, dolutegravir) because of unknown HIV resistance; after 4 weeks ritonavir-boosted darunavir was stopped.

^oFive missing values.

^pProportion of STI clinic records that were linked to HIV clinic records.

^qDefined as the first documented viral load < 50 copies/mL.

^rProportion of MSM who were not lost to follow-up before cART initiation.

^sTwo consecutive measures of viral load ≥ 200 copies/mL within 6 months after viral suppression for the 629 MSM who achieved viral suppression. For 84 MSM, no viral load data were available after 6 months of initial viral suppression.

^tEstimated with Kaplan-Meier estimators.

^uTwelve missing values.

^vLog-rank test.

^wMedian number of days between cART initiation visit and follow-up viral load measurements; MSM who did not achieve viral suppression and visits after the first viral load measurement < 50 copies/mL were excluded from these estimations.

Time to Viral Suppression

Median time between diagnosis and intake at an HIV treatment center was similar in the first 2 periods (14 and 13 days, respectively), decreased to 9 days in period 3a (SOC), and was reduced to 1 day in the AHI strategy (Table 2). LTFU before cART initiation was similar in SOC (2.2%–6.9%) and 0% in the AHI strategy among MSM with linked records. Median time from diagnosis to cART initiation was 439 days, 99 days, 29 days, and 1 day in periods 1, 2, 3a, and 3b, respectively (Figure 2). Median time from diagnosis to viral suppression was 584 (IQR, 267–1065) days in period 1; 230 (IQR, 132–480) days in period 2; 95 (IQR, 63–136) days in period 3a; and 55 (IQR, 31–72) days in 3b (the AHI strategy).

Figure 2.

Kaplan-Meier curves with days from human immunodeficiency virus (HIV) diagnosis to combination antiretroviral therapy (cART) initiation and viral suppression, respectively, among men who have sex with men (MSM) diagnosed with HIV at the Amsterdam sexually transmitted infection clinic, stratified by cART initiation period, January 2008–June 2017. A, Estimated time from HIV diagnosis to cART initiation. B, Estimated time from HIV diagnosis to viral suppression. C, Estimated time from diagnosis to viral suppression among MSM with a median of ≤ 60 days between viral load measurements. Stratification for all panels was done by cART initiation period, based on the Dutch HIV treatment guidelines: (1) standard of care (cART if CD4 count < 500 cells/μL, January 2008–December 2011); (2) standard of care (cART if CD4 count < 500 cells/μL or if in acute or early HIV infection [AEHI], January 2012–July 2015); and (3a) standard of care (universal cART, January 2015–June 2017) or (3b) acute HIV infection (AHI) strategy (immediate cART, January 2015–June 2017). The black hashes represent censored MSM.

As MSM with AEHI may have initiated cART earlier than MSM with established infection, we created separate Kaplan-Meier graphs for MSM with AEHI and those with established infection. These indicate that the time from diagnosis to viral suppression decreased from period 1 to period 3a for MSM with AEHI and also for those with established infection (Supplementary Figure 2). Median time from cART initiation to viral suppression was 125, 91, 59, and 54 days in periods 1, 2, 3a, and 3b, respectively. As there was no standardized study visit schedule for viral load measurements, we assessed median interval between the cART initiation visit and the follow-up viral load measurements until viral suppression was achieved. This decreased from 125 days in period 1 to 54 days in period 3b, which may explain the observed decline in time from cART initiation to viral suppression. All MSM achieved viral suppression at the first follow-up visit after cART initiation. In a subgroup analysis of MSM with similar intervals between viral load measurements (ie, a median of ≤ 60 days), the median time from diagnosis to viral suppression was 799, 129, 63, and 34 days in periods 1, 2, 3a, and 3b, respectively (P < .0001; Figure 2C).

DISCUSSION

The implementation of a strategy for rapid AHI diagnosis in MSM linked to immediate cART initiation proved feasible at the Amsterdam STI clinic, and was temporally associated with a higher AHI yield and a significant reduction in the time from HIV diagnosis to viral suppression, in relation to SOC. The increase in AHI diagnoses may have resulted from increased awareness of AHI among MSM and healthcare providers, as well as more frequent testing among MSM. This assumption is supported by recent citywide data from MSM entering HIV care in Amsterdam, showing an increase in the proportion of MSM with a negative HIV test in the 6 months before diagnosis (as a proxy for early infection): from 18% in 2010–2015 to 34% in 2016, and 27% in 2017 [1]. In the AHI strategy, 53% of MSM diagnosed with HIV had AHI, and 89% had AEHI. The recommendation for universal use of cART dramatically decreased time from diagnosis to viral suppression: from 584 days in 2008–2011 to 95 days in 2015–2017. The interventions of the AHI strategy decreased the time from diagnosis to viral suppression even further to 55 days.

The addition of the point-of-care HIV RNA test resulted in 2 MSM diagnosed with Fiebig stage I, who otherwise would not have been diagnosed. More importantly, it ensured same-visit results and confirmation of an indeterminate Ag/Ab test result (Fiebig II), as described by others [33]. In the standard testing setting, MSM with indeterminate Ag/Ab tests were requested to return to the STI clinic after 1–3 weeks for follow-up testing, thus delaying cART initiation.

Immediate cART initiation has been shown to decrease time to viral suppression in different settings with limited to moderate access to healthcare [11–24], including for people with AHI. The Rapid ART Program Initiative for HIV Diagnoses (RAPID) intervention in San Francisco decreased time from diagnosis to viral suppression from 170 to 65 days [22]. This included, among other interventions, accelerated insurance approval and 5-day cART starter packs. A study among MSM in London showed that cART initiation after a median of 20 days of diagnosis resulted in 99% being virally suppressed at 24 weeks [24]. In San Diego, MSM with AHI initiated cART slightly later after diagnosis than MSM with established infection (14 vs 9 days, respectively), because AHI diagnosis was not provided at the point of care [21], but there were no significant differences in median time from diagnosis to viral suppression (12 weeks in both groups). Our study suggests that, even in the Dutch context, where health insurance is mandatory and includes coverage of cART, the interventions of the AHI strategy could have potential additional value in preventing onward transmission. In our experience, the accelerated confirmation of diagnosis and linkage to care were the most valuable and easiest-to-implement components of the strategy. Therefore, these interventions have now been implemented for all people newly diagnosed with HIV at the Amsterdam STI clinic. Although none of the MSM with linked records were LTFU in the AHI strategy, a small but worrisome proportion (4%) of MSM in SOC were LTFU before cART initiation. This needs careful attention in HIV elimination programming.

This study has some limitations. First, MSM tested in SOC and the AHI strategy may not have been independent groups, as wider dissemination of the campaign may have reached both groups. The campaign referrals yielded no direct AHI diagnoses, which warrants critical evaluation of the benefits of the campaign. It could be hypothesized that the campaign increased general awareness of AHI among MSM and healthcare providers, which should be assessed in future studies. Furthermore, one of the referral sources of the AHI strategy was the STI clinic itself. This has overestimated the difference in AHI yield when we compared standard testing with the AHI strategy in 2015–2017, as MSM with Fiebig stage II would otherwise have been diagnosed via routine testing. However, the high AHI yield (4%) in the AHI strategy suggests that MSM with high risk of AHI were rightly targeted. Second, we were unable to link 35% of STI clinic records to HIV treatment center records. These records were excluded from the analyses of time to viral suppression. The proportion of record linkages was similar across all years; therefore, we expect that limited bias was introduced. A possible explanation for failure to link could be that the STI clinic allows anonymous free-of-charge consultations for MSM residing in Amsterdam. Thirty-two percent of MSM with unlinked records submitted an invalid postal code; other MSM might have moved or submitted a valid postal code which was not theirs, as they did not want to be traceable or lived outside the region. Since Amsterdam has exceeded the Joint United Nations Programme on HIV/AIDS (UNAIDS) 90-90-90 targets [34], we expect that the majority of MSM with unlinked records are in care, but we were unable to confirm this due to privacy regulations. Third, median time between viral load measurements decreased over time, which may partly explain the decreased time from cART initiation to viral suppression, but not the decreased time from diagnosis to cART initiation. Due to the differences in viral load measurement intervals, we were unable to assess the potential effect of treatment regimen on the time from cART initiation to viral suppression. As INSTI-based regimens decrease the time to viral suppression compared with other cART regimens [21, 35], we expect that the introduction of INSTI-based regimens has also contributed to the decreased time from cART initiation to viral suppression observed in this study.

CONCLUSIONS

We found that targeted testing for AHI is feasible and resulted in a high AHI yield and uptake of immediate cART. Changes in HIV treatment guidelines over time, increased testing rates, and awareness of AHI among MSM, along with the combined interventions of the AHI strategy, decreased the time between HIV infection, subsequent diagnosis, and viral suppression. Future studies should assess the impact of early diagnosis and immediate cART initiation on HIV incidence in Amsterdam, as well as experiences of MSM with immediate cART initiation, long-term retention in care, and cost-effectiveness.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. M. D., M. S. R., M. M. H., M. F. S. L., and G. J. B. designed the study. M. D., A. H., T. H., U. D., E. H., M. P., J. M. P., P. R., M. F. S. L., and G. J. B. contributed to design of the acute human immunodeficiency virus (HIV) infection strategy. M. D., A. H., T. H., E. H., and G. J. B. had oversight in data collection and procedures at the sexually transmitted infection (STI) clinic. M. M. H., A. S., C. S., P. R., and M. V. had oversight in data collection and procedures at the ATHENA cohort. M. D., M. S. R., M. M. H., A. S., and C. S. conducted data cleaning. M. D. conducted the statistical analyses and drafted the manuscript. All authors critically reviewed and revised the manuscript and approved the final version for publication.

Acknowledgments. The authors acknowledge STI clinic and HIV treatment center staff who collected the data; Annebel Tahey, Justin Luidens, and Dewi Usmany who helped with study logistics; and Michelle Kroone, Gerben Rienk Visser, and Dominique Loomans for developing and maintaining the study database. The authors thank David Kwa, Alje van Dam, and Sylvia Bruisten for their advice in interpretation of laboratory results; Arjan van Bijnen, Jan van Bergen, Fred Verdult, and Wim Zuilhof for developing the media campaign; and Lindsey Bauer and Sharon Dorrestijn for their support in linking the data. Last, the authors thank Marije Groot Bruinderink, Nina Schat, and Judith van der Schaaf for their continuous support and financial advice.

Financial support. The HIV Transmission Elimination Amsterdam (H-TEAM) Initiative is supported by Aids Fonds (grant number 2013169); the AmsterdamDiner Foundation, Bristol-Myers Squibb International Corporation (study number AI424-541); Gilead Sciences Europe Ltd (grant number PA-HIV-PREP-16–0024); Gilead Sciences (protocol numbers CO-NL-276–4222 and CO-US-276–1712); Janssen Pharmaceuticals (reference number PHNL/JAN/0714/0005b/1912fde); the M.A.C. AIDS Fund, ViiV Healthcare (PO numbers 3000268822 and 3000747780); and ZonMw (grant number 522002003). The ATHENA cohort is managed by Stichting HIV Monitoring and is supported by a grant from the Dutch Ministry of Health, Welfare and Sport through the Centre for Infectious Disease Control of the National Institute for Public Health and the Environment.

Potential conflicts of interest. G. J. B. has received grants through her institution from Bristol-Meyers Squibb and the M.A.C. AIDS Fund; honoraria to her institution for scientific advisory board participation with Gilead Sciences; and speaker’s fees from Gilead Sciences and Takeda. A. S. has received grants from the Dutch Ministry of Health, Welfare and Sport, paid to his institution, and grants from the European Centre for Disease Prevention and Control, paid to his institution, during the conduct of the study. P. R. has received grants through his institution from Gilead Sciences, Janssen Pharmaceutica, ViiV Healthcare, and Merck, and honoraria to his institution for scientific advisory board participation from Gilead Sciences, ViiV Healthcare, Merck, and Teva. M. P. has received unrestricted research grants and speaker’s fees from Gilead Sciences, Roche, AbbVie, and Merck Sharp and Dohme, all of which were paid to her institution. M. V. has received grants through his institution from Gilead Sciences, Janssen Pharmaceutica, ViiV Healthcare, Merck, and AbbVie, and honoraria to his institution for scientific advisory board participation from Gilead Sciences, ViiV Healthcare, Merck, AbbVie, and Janssen Pharmaceutica. E. H. obtained study medication and a research grant from Gilead Sciences through her institution. U. D. obtained unrestricted research grants and speaker’s fees from Gilead Sciences through his institution. All other authors report no potential conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

APPENDIX

HIV Transmission Elimination Amsterdam (H-TEAM). Members are listed in alphabetical order:

H-TEAM Steering Committee. J. E. A. M. van Bergen,^1,2,3 P. Brokx,⁴ F. Deug,¹ M. Heidenrijk,⁵ M. Prins,^6,7 P. Reiss,^5,8 (chair) M. van der Valk.⁷

H-TEAM Core Project Group. J. E. A. M. van Bergen,^1,2,3 G.J. de Bree,^5,7 (chair) P. Brokx,⁵ U. Davidovich,⁶ S. E. Geerlings,⁷ E. Hoornenborg,⁶ A. Oomen,¹ A. van Sighem,⁸ W. Zuilhof.¹

H-TEAM Project Management. N. Schat.⁵

H-TEAM additional collaborators. R. C. A. Achterbergh,⁶ M. van Agtmael,⁹ J. Ananworanich,¹⁰ D. Van de Beek,¹¹ G. E. L. van den Berk,¹² D. Bezemer,⁸ A. van Bijnen,¹ W. L. Blok,¹² S. Bogers,⁷ M. Bomers,⁹ C. A. B. Boucher,¹³ W. Brokking,¹⁴ D. Burger,¹⁵ K. Brinkman,¹² N. Brinkman,¹⁶ M. de Bruin,¹⁷ S. Bruisten,⁶ L. Coyer,⁶ R. van Crevel,¹⁸ C. G. Daans,⁶,¹⁹ L. Dellemann,¹ M. Dijkstra,⁶ Y. T. van Duijnhoven,⁶ A. van Eeden,¹⁴ L. Elsenburg,¹⁴ M. A. M. van den Elshout,⁶ C. Ester,⁸ E. Ersan,² P. E. V. Felipa,² P. H. J. Frissen,¹² T. B. H. Geijtenbeek,²⁰ M. H. Godfried,⁷ J. van Goo,^l6 A. Goorhuis,⁷ M. Groot, ¹⁴ C. A. Hankins,⁵ A. Heijnen,^21,22 M. M. J. Hillebregt, ⁸ A. Hogewoning,⁶ M. Hommenga,⁶ J. W. Hovius,⁷ Y. Janssen,¹⁶ K. de Jong,⁶ V. Jongen,⁶ N. A. Kootstra,²³ R. A. Koup,²⁴ F. P. Kroon,²⁵ T. J. W. van de Laar,^26,27 F. Lauw,²⁸ M. M. van Leeuwen,⁶ K. Lettinga,²⁹ I. Linde,⁶ D. S. E. Loomans,⁶ J. T. van der Meer,⁷ T. Mouhebati,¹ B. J. Mulder,⁶ J. Mulder,³⁰ F. J. Nellen,⁷ A. Nijsters,1 H. Nobel,⁷ P. Oostvogel,⁶ E. L. M. Op de Coul,³ E. Peters,⁹ I. S. Peters,⁶ T. van der Poll,⁷ O. Ratmann,³¹ C. Rokx,³² M. S. van Rooijen,⁶ M. F. Schim van der Loeff,^6,33 W. E. M. Schoute,¹² G. J. Sonder,⁶ J. Veenstra,²⁹ A. Verbon,³² F. Verdult,⁴ J. de Vocht,⁹ H. J. de Vries,^6,33,34 S. Vrouenraets,²⁹ M. van Vugt,⁷ W. J. Wiersinga,⁷ F. W. Wit,^7,8 L. R. Woittiez,⁷ S. Zaheri,⁸ P. Zantkuijl,¹ M. C. van Zelm,³⁵ A. Żakowicz,³⁶ H. M. L. Zimmermann.⁶

¹Soa Aids Nederland, Amsterdam, the Netherlands; ²Department of General Practice, Amsterdam UMC – location AMC, University of Amsterdam, Amsterdam, the Netherlands; ³Epidemiology and Surveillance Unit, Center for Infectious Disease Control, National Institute of Public Health and the Environment, the Netherlands; ⁴Dutch Association of PLHIV, Amsterdam, the Netherlands; ⁵Department of Global Health, Amsterdam UMC – location AMC, and Amsterdam Institute for Global Health and Development, Amsterdam, the Netherlands; ⁶Department of Infectious Diseases, Public Health Service of Amsterdam, Amsterdam, the Netherlands; ⁷Department of Internal Medicine, Division of Infectious Diseases, Amsterdam UMC – location AMC, Amsterdam, the Netherlands; ⁸Stichting HIV Monitoring, Amsterdam, the Netherlands; ⁹Department of Internal Medicine, Amsterdam UMC – location VUMC, Amsterdam, the Netherlands; ¹⁰US Military HIV Research Program and the Henry M. Jackson Foundation for the Advancement of Military Medicine, Bethesda, United States; ¹¹Center of Infection and Immunity Amsterdam (CINIMA), Department of Neurology, Amsterdam UMC – location AMC, Amsterdam, the Netherlands; ¹²Department of internal medicine, Onze Lieve Vrouwe Gasthuis – location East, Amsterdam, the Netherlands; ¹³Department of viro-science, Erasmus Medical Center Rotterdam, Rotterdam, the Netherlands; ¹⁴DC Klinieken, Amsterdam, the Netherlands; ¹⁵Department of Pharmacy, Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands; ¹⁶Primary Care Amsterdam and Almere (Elaa), Amsterdam, the Netherlands; ¹⁷Aberdeen Health Psychology Group, Institute of Applied Health Sciences, University of Aberdeen, Aberdeen, United Kingdom; ¹⁸Department of Internal Medicine, Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands; ¹⁹Center of Expertise on Gender Dysphoria, Amsterdam UMC – location VUMC, Amsterdam, the Netherlands; ²⁰Laboratory of Experimental Immunology, Amsterdam UMC – location AMC Amsterdam, the Netherlands; ²¹Sexology Center Amsterdam, Amsterdam, the Netherlands; ²²GP practice Heijnen & de Meij, Amsterdam, the Netherlands; ²³Laboratory for Viral Immune Pathogenesis, Amsterdam UMC – location AMC Amsterdam, the Netherlands; ²⁴Immunology Laboratory, Vaccine Research Center, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Rockville, Maryland, USA; ²⁵Department of Infectious Diseases, Leiden University Medical Center, Leiden, the Netherlands; ²⁶Department of Medical Microbiology, Onze Lieve Vrouwe Gasthuis, Amsterdam, the Netherlands; ²⁷Department of Donor Medicine Research, Laboratory of Blood-borne Infections, Sanquin Research, Amsterdam, the Netherlands; ²⁸Department of Internal Medicine, Medical Center Jan van Goyen, Amsterdam, the Netherlands; ²⁹Department of Internal Medicine, Onze Lieve Vrouwe Gasthuis – location West, Amsterdam, the Netherlands; ³⁰Department of Internal Medicine, Slotervaart Hospital (former), Amsterdam, the Netherlands; ³¹School of Public Health, Faculty of Medicine, Imperial College London, London, United Kingdom; ³²Department of Internal Medicine and Infectious Diseases, Erasmus Medical Center, Rotterdam, the Netherlands; ³³Center for Infection and Immunology, Amsterdam (CINIMA), Amsterdam UMC – location AMC, University of Amsterdam, Amsterdam, the Netherlands; ³⁴Department of Dermatology, Amsterdam UMC – location AMC, University of Amsterdam, Amsterdam, the Netherlands; ³⁵Department of Virology, Erasmus Medical Center, Rotterdam, the Netherlands; ³⁶AIDS Healthcare Foundation, Amsterdam, the Netherlands.

Author notes

Members of the H-TEAM Initiative are listed in the Appendix.