Figures
Abstract
Background
Point of-care (POC) HIV-1 RNA tests which are accurate and easy to use with limited infrastructure are needed in resource-limited settings (RLS). We systematically reviewed evidence of POC test performance compared to laboratory-based HIV-1 RNA assays and the potential utility of these tests for diagnosis and care in RLS.
Methods
Studies published up to July 2018 were identified by a search of PUBMED, EMBASE, Web of Science, CINAHL and Cochrane Central Register of Controlled Trials. Studies evaluating the use of POC HIV-1 RNA testing for early infant diagnosis (EID), acute HIV infection (AHI) diagnosis, or viral load monitoring (VL), compared to centralized testing, were included. Separate search strategies were used for each testing objective.
Results
197 abstracts were screened and 34 full-text articles were assessed, of which 32 met inclusion criteria. Thirty studies evaluated performance and diagnostic accuracy of POC tests compared to standard reference tests. Two of the thirty and two additional studies with no comparative testing reported on clinical utility of POC results. Five different POC tests (Cepheid GeneXpert HIV-1 Quantitative and Qualitative assays, Alere q HIV‐1/2 Detect, SAMBA, Liat HIV Quant and Aptima HIV‐1 Quant) were used in 21 studies of VL, 11 of EID and 2 of AHI. POC tests were easy to use, had rapid turnaround times, and comparable accuracy and precision to reference technologies. Sensitivity and specificity were high for EID and AHI but lower for VL. For VL, lower sensitivity was reported for whole blood and dried blood spots compared to plasma samples. Reported error rates for Cepheid GeneXpert Qual (2.0%-5.0%), GeneXpert Quant (2.5%-17.0%) and Alere q HIV‐1/2 Detect (3.1%-11.0%) were higher than in WHO prequalification reports. Most errors resolved with retesting; however, inadequate sample volumes often precluded repeat testing. Only two studies used POC results for clinical management, one for EID and another for VL. POC EID resulted in shorter time-to-result, rapid ART initiation, and better retention in care compared to centralised testing.
Citation: Agutu CA, Ngetsa CJ, Price MA, Rinke de Wit TF, Omosa-Manyonyi G, Sanders EJ, et al. (2019) Systematic review of the performance and clinical utility of point of care HIV-1 RNA testing for diagnosis and care. PLoS ONE 14(6): e0218369. https://doi.org/10.1371/journal.pone.0218369
Editor: Kwasi Torpey, University of Ghana College of Health Sciences, GHANA
Received: November 8, 2018; Accepted: May 31, 2019; Published: June 27, 2019
Copyright: © 2019 Agutu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results presented in the study are available from publicly accessible peer reviewed publications. The search terms and databases used to generate said publications have been provided as supplementary material.
Funding: This work was supported through the Sub-Saharan African Network for TB/HIV Research Excellence (SANTHE), a DELTAS Africa Initiative [grant # DEL-15-006]. The DELTAS Africa Initiative is an independent funding scheme of the African Academy of Sciences (AAS)’s Alliance for Accelerating Excellence in Science in Africa (AESA) and supported by the New Partnership for Africa’s Development Planning and Coordinating Agency (NEPAD Agency) with funding from the Wellcome Trust [grant # 107752/Z/15/Z] and the UK government. SMG was supported by the Robert W. Anderson Endowed Professorship in Medicine. The views expressed in this publication are those of the authors and not necessarily those of AAS, NEPAD Agency, Wellcome Trust or the UK government’. This manuscript was submitted for publication with the permission from the Director of the Kenya Medical Research Institute (KEMRI).
Competing interests: The authors have declared that no competing interests exist.
Introduction
Clinical point of care (POC) testing has evolved for situations requiring fast turnaround times and those in which a centralized lab approach faces other limitations, such as difficult or costly transportation of samples [1]. POC testing is defined as near-patient testing in a hospital, doctor’s office, clinic or home, with the advantage of providing a rapid answer[1]and thus resulting in fewer patients being lost to follow up. Testing occurs while patients are on-site and reduces the burden on patients by circumventing the need for a return visit [2]. POC testing can significantly expand access to clinical laboratory testing for rural populations by eliminating the need for sample transport, laboratory and data management infrastructure, and highly trained staff [3].
Many currently available diagnostic assays for human immunodeficiency virus type 1 (HIV-1) infection, including polymerase chain reaction (PCR), enzyme-linked immunosorbent assay (ELISA) and Western blot (WB), are limited to centralized laboratories due to requirements for infrastructure and trained personnel. These laboratory-based assays are complex, expensive and time-consuming, limiting their accessibility in developing countries where the challenge of the HIV/AIDS pandemic is most severe [4]. Although rapid antibody tests have greatly expanded access to HIV diagnosis, the inability of these tests to detect HIV-1 RNA and the poor performance of current fourth generation rapid antigen/antibody assays in many high-prevalence settings [5, 6] means that resource-limited settings face challenges in the detection of infection in infants and in patients who were recently infected. In addition, HIV-1 RNA quantitation or semi-quantitative viral load (VL) testing is needed to monitor ART [4]. POC HIV-1 RNA testing could contribute to timely HIV diagnosis and improve detection of treatment failure, resulting in improved clinical outcomes and reduced HIV transmission [4].
In 2016, two POC HIV-1 RNA assays received World Health Organization (WHO) prequalification for early infant diagnosis (EID): the Alere q HIV-1/2 Detect (Alere Technologies GmbH, Jena, Germany) and the Xpert HIV-1 Qual Assay (Cepheid AB, Solna, Sweden)[7]. Additional nucleic acid amplification testing (NAAT) assays for HIV-1 RNA detection or quantitation that have received approval include the Aptima HIV‐1 Quant Dx Assay (Hologic, Inc., San Diego, USA) in 2017, Xpert HIV‐1 Viral Load (Cepheid AB, Solna, Sweden) in 2017 [7], and the Alere q HIV-1/2 VL plasma assay (“m-PIMA HIV-1/2 VL”) (Alere Technologies GmbH, Jena, Germany) in 2019 [8]. Several additional assays of this type have active applications for WHO prequalification, including the SAMBA I & II HIV-1 Semi-Q test (Diagnostics for the Real World Ltd, San Jose, USA and Cambridge, United Kingdom) and the SAMBA I & II HIV-1 Qual Whole blood test (Diagnostics for the Real World Ltd, San Jose, USA and Cambridge, United Kingdom) [9]. Other newer assays such as the Liat HIV Quant POC VL assay (Iquum, Inc., Marlborough, MA) have not yet applied for WHO prequalification [10].
In 2018, Nash et al conducted a review of the performance of the Xpert HIV‐1 Viral Load assay, reporting high correlation between POC Xpert results and those of laboratory-based reference assays (pooled Pearson correlation 0.94; pooled Spearman correlation 0.96). Bland-Altman analyses pooled from 11 identified studies were within 0.35 log/copies ml of perfect agreement [11]. Numerous field evaluations of other POC HIV-1 RNA assays have since been conducted in different countries and settings. However, to date, there is no systematic review on the performance of all currently available POC HIV-1 RNA testing assays and assessment of their uses in HIV care.
As POC HIV-1 RNA testing has become increasingly available and has been shown to be accurate and valid for multiple clinical uses, this systematic review aimed to synthesize evidence on the performance and clinical utility of POC quantitative (i.e., continuous) or qualitative (i.e., dichotomous) HIV-1 RNA testing assays for different purposes, and to identify barriers and facilitators to their scale up in resource-limited settings.
Methods
Eligibility criteria
Studies evaluating the use of POC HIV-1 RNA testing for EID, acute HIV infection (AHI) diagnosis, or VL monitoring, as well as studies comparing the POC HIV-1 RNA testing to centralized HIV-1 RNA testing were included. We excluded studies testing commercially prepared sample panels [12]. Publications on laboratory-based HIV-1 RNA diagnostics, POC HIV-2 RNA assays, rapid antibody HIV tests, combined rapid antibody/antigen tests, and POC assays used for diagnosis of infectious diseases other than HIV were also excluded.
Search strategy
The search was carried out in March 2017 and updated in July 2018. PubMed, EMBASE, Web of Science, CINAHL and the Cochrane controlled trial register were searched using the following terms: (“point of care HIV-1 viral load” OR “point-of-care HIV-1 viral load” OR “Xpert HIV-1” OR “GeneXpert” OR Alere OR SAMBA) in combination with (EID OR “early infant diagnosis” OR “infant HIV infection” OR MTCT OR “mother to child transmission”) or (“ART monitoring” OR “HIV antiretroviral therapy monitoring “OR “viral load monitoring “OR “HIV-1 viral load monitoring” OR “treatment failure”) or (“acute HIV diagnosis” OR “early HIV diagnosis”). References of included studies were searched for additional relevant literature. Conference abstracts were excluded due to insufficient detail.
Data extraction
Study titles and abstracts were checked for eligibility according to the inclusion criteria detailed above. Full-text articles were retrieved for potentially eligible studies, and the final set of included studies was agreed upon by all researchers. Data were abstracted by two independent reviewers (CAA and CJN) using a standard data abstraction form that recorded study characteristics, sample characteristics, POC assay evaluated, comparator assay (if any), test accuracy (correlation or concordance, sensitivity, specificity), and error rate. In addition, factors related to barriers or facilitators to the scale-up of POC assays were noted, including those related to human resources, supply chain management, and patient and provider attitudes.
Quality assessment
The quality of the studies comparing POC to reference assays was assessed by two independent reviewers using eighteen of twenty-four criteria selected from the STARD guidelines for reporting diagnostic studies [13]; these criteria were selected based on relevance to the literature reviewed. The eighteen parameters appraised covered six main categories including the title; abstract and key words; introduction; methods (participant eligibility criteria and sampling, reference and index test methods and statistical methods for comparing measures of diagnostic accuracy); results (flow chart of participant sampling, turnaround time to test results for reference and index tests and reported estimates of diagnostic accuracy) and discussion (clinical applicability of the results). Each reviewer scored the publications, with disagreements resolved by discussion with all authors.
Results
Study selection
The search produced 313 references. Following removal of duplicates, 197 titles and abstracts were screened and 34 references were identified for full-text review. An additional 8 studies were identified through searches of the references of included articles. After screening was completed and any discrepancies resolved by the study team, 32 full-text articles met inclusion criteria and were included in the final review. Fig 1 details the study selection flow chart.
Characteristics of included studies
Table 1 outlines the characteristics of the 32 included studies. All studies were published between 2014 and 2018. Thirty studies were field reviews comparing the performance of POC HIV-1 RNA tests to a reference standard. Two of the thirty and two additional studies with no comparative testing included reported on the clinical utility of POC results [14–17]. Of the two field reviews that reported on clinical utility [15, 17], one was a field feasibility evaluation study [14] and the other a cluster-randomized trial [16]. POC results were used for clinical management in two studies [14, 15]. The studies enrolled 12,535 infants for EID, 846 adults for AHI screening, 222 adults for HIV detection and 6,975 HIV patients for VL monitoring, and were conducted in sub-Saharan Africa (n = 16 studies, N = 14,744 participants), Europe (n = 5, N = 811), India (n = 3, N = 561), Israel (n = 4, N = 1,633), or multiple sites (n = 4, N = 2,627).
Of the 32 included studies, 11 focused on EID, 2 on AHI diagnosis, and 21 on VL monitoring. Two studies assessed both EID and VL monitoring [14],[18]. The studies evaluated the following POC assays: Cepheid GeneXpert Qual (5 for EID, 2 for AHI) and Cepheid GeneXpert Quantitative (Quant) assay (13 for VL), Alere q HIV‐1/2 Detect (6 for EID, 1 for VL), Aptima HIV‐1 Quant Dx Assay (4 for VL), Liat HIV Quant (1 for VL), and SAMBA I/II (1 for EID, 3 for VL). Three studies used multiple POC assays. References assays used to compare POC to laboratory-based assays are detailed in Table 1. Six studies used multiple comparator assays. In addition, the 2 studies of AHI diagnosis both used Architect HIV Ag/Ab Combo assay (AR, Abbott Diagnostics, Abbott Park, IL, USA), followed by Vidas HIV DUO ULTRA (VD, Biomérieux, Marcy-l'Etoile, France) if reactive; confirmatory testing was performed with the Geenius HIV-1/2 differentiation assay (GS, Bio-Rad Laboratories, 68 Hercules, California).
Three studies assessed the performance of POC HIV-1 RNA assays compared to laboratory-based NAAT assays for VL on specimens other than blood plasma [31, 34, 44]. One evaluated VL results using whole blood, plasma, and dry blood spots (DBS) [31]; another whole blood only [34]; and a third used leuko-depleted whole blood (i.e., filtered whole blood) [44].
Accuracy and precision
Where correlation with quantitative reference test results was reported in the included studies, this was found to be high across all POC assays assessed: Alere q NAT [34], Cepheid GeneXpert Quant [18, 27–29, 32, 35–39, 43] and Qual assays [18], Aptima HIV‐1 Quant Dx assay [33, 37, 42], and Liat HIV Quant [10]. No information on correlation was reported for the studies performed with SAMBA HIV-1 Qual and SAMBA HIV-1 Semiquantitative tests [24, 30, 40, 44]. Most studies reported Spearman [18, 27, 32] or Pearson’s correlation coefficients (ρ and r, respectively) for quantitative HIV-1 RNA results [18, 28, 35–39, 42, 43], while seven studies reported Cohen’s Kappa coefficient (κ) for both quantitative and qualitative HIV-1 RNA results [15, 17, 22, 33, 35, 41, 42]. Percentage concordance or agreement with dichotomous reference test results (i.e., results above or below a threshold) was reported by sixteen studies [10, 18, 20, 22, 23, 30, 31, 33, 35–38, 40–42, 44], while sensitivity and specificity were reported in 17 and 18 studies respectively [10, 15, 17, 19–27, 30–32, 34–36, 40]. Eleven studies on VL monitoring reported Bland-Altman analysis results (i.e., mean difference vs. the comparator) for log10-transformed viral loads across the wide range of HIV-1 RNA levels [18, 27, 29, 32, 36–39, 41–43]. For all studies, an inter-assay difference of <0.5 log copies/ml, the accepted clinically relevant difference between two viral load measurements, was observed [45, 46].
Early infant diagnosis
Cepheid GeneXpert HIV-1 Qual (Xpert Qual).
Five studies evaluated the use of Xpert Qual for EID [14, 15, 18, 21, 23]. The Xpert Qual provides a qualitative result (HIV detectable or undetectable). A high correlation and agreement compared with reference tests was reported in two studies (r = 0.95 and k = 0.97, respectively) [15, 18]. Concordance compared to the reference test was reported by one study, and was found to be high at 90.9% [18]. Overall, sensitivity and specificity compared to reference assays ranged from 93.3%-100% and 99.5%-100%, respectively [15, 21, 23]. Xpert Qual performed well on DBS samples from infants, with 93.3% sensitivity and 100% specificity [21]. Fig 2 presents a Forest Plot of the sensitivity of Cepheid GeneXpert Qual assay compared to reference tests for EID.
Alere q HIV‐1/2 Detect (Qual).
Six studies evaluated Alere Detect for EID compared to reference tests, with a qualitative result provided (HIV detected or undetected) [16, 17, 19, 20, 22, 23]. Sensitivities and specificities were high, ranging from 95.5%-100% and 99.5% -100% [17, 19, 20, 22, 23]. In two studies, sensitivity was reported to be lower in birth testing (infants <7 days old) compared to older children (routine EID at ages 6–14 weeks), at 90.0% and 93.3% compared to 100% and 95.5% respectively [19, 20]. Three studies reported high percentage agreement compared with reference tests, ranging from 97.8%-99.8% [20, 22, 23], and two studies reported on agreement by Cohen’s Kappa coefficient (κ) ranging from 0.98–1.0 [17, 22]. Fig 3 presents a Forest Plot of the sensitivity of Alere Detect compared to reference tests for EID.
SAMBA HIV-1 Qual.
One study assessed the performance of the SAMBA HIV-1 Qualitative test for EID compared to a reference test. High sensitivity (98.5%) and high specificity (99.8%) were reported [24].
Acute and chronic HIV diagnosis
Cepheid GeneXpert HIV-1 Qual.
Two studies by Michaeli et al [25] and Rakovsky et al [26] evaluated the use of Cepheid GeneXpert HIV Qual for detection of HIV-1 RNA in stored pre-seroconversion samples collected from individuals with recent, documented HIV-1 acquisition. All samples tested reactive on Architect HIV Ag/Ab Combo and Vidas HIV DUO ULTRA 4th generation assays and had an indeterminate or negative result when tested by the Bio-Rad Geenius HIV-1/2 Confirmatory assay. A high sensitivity (94.9%-100%) and specificity (92.6%-100%) for the detection of HIV-1 RNA was reported in the two studies [25, 26]. Of note, Cepheid GeneXpert HIV-1 Qual failed to detect HIV-1 RNA in samples from patients with VL suppressed on antiretroviral therapy, as expected [26].
Garrett et al [27] reported on the use of the Xpert Qual for the detection of HIV-1 RNA on whole blood samples from adult known HIV positive women, likely with chronic infection. Of the twenty samples collected, 13 were from patients with detectable viral loads and 7 were virologically suppressed. All except one of the participants (95% sensitivity) with low level viremia (VL 523 copies/ml) were correctly identified by the assay.
SAMBA HIV-1 Qual.
Ondiek [24] et al evaluated the use of SAMBA HIV-1 Qual for the detection of HIV using 202 whole blood adult samples of previously known and unknown HIV status, including ART-naïve patients. It was not specified whether participants were acutely or chronically infected with HIV. They reported a sensitivity and specificity of 100% and 99.2% respectively.
Viral load monitoring
Cepheid GeneXpert HIV-1 Quant.
Thirteen studies evaluated the performance of the Xpert Quant for viral load monitoring compared to reference assays [14, 18, 27–29, 31, 32, 35–39, 43]. The linear detection range of the Xpert Quant is 40−107 copies/ml, with results given as a quantitative value within the analytical measurement range [47]. Correlation of HIV-1 RNA levels between Xpert Quant and reference test results was reported in eleven studies, ranging from 0.81 to 0.99 [18, 27–29, 32, 35–39, 43]. Nine studies reported a mean difference in viral load vales of -0.01 to 0.34 log copies/ml for Xpert Quant compared to reference assays [18, 27, 29, 32, 36–39, 43]. Where concordance and agreement above a threshold was reported, it was found to be high, ranging from 87.2%-100% compared to reference tests [18, 31] [35–38]. In one study, agreement was 97.1% at 40 copies/ml, but only 90.6% at 1000 copies/ml, the WHO threshold for clinical decision-making for virological failure [38]. Overall, sensitivity and specificity for HIV-1 RNA detection ranged between 84%-97% and 95.9%-100% respectively, when POC assays were compared to reference assays [31, 32, 35, 36, 38]. Sensitivity compared to reference tests at the clinically relevant threshold of 1000 copies/ml ranged from 92.9%-100% [31, 36, 38], with specificity ranging from 95.9%-98% [31, 36, 38]. Low sensitivity of 60.7% and 50.0% for the detection of HIV-1 RNA >1000 copies/ml was reported for whole blood and DBS samples respectively. Specificity was comparable to plasma samples at 91.6% and 96.6% for whole blood and DBS samples respectively [31].
Alere q NAT (Quant).
One study evaluated the performance of Alere q NAT compared to a reference test in whole blood samples, using several potential thresholds for treatment failure. Sensitivity for identifying treatment failure was 96.8% at 1,000 copies/ml and 84.0% at 10,000 copies/ml; however, specificity was 47.8% at 1,000 copies/ml compared to 90.3% at 10,000 copies/ml [34]. Correlation of HIV-1 RNA levels by the two methods used was r2 = 0.361.
SAMBA HIV Semiquantitative assay.
High agreement was reported between the SAMBA HIV Semiquantitative assay and reference tests at a threshold of 1,000 copies/ml, ranging from 96.5%-98.1% [30, 40, 44]. Two studies reported only specificity at a threshold of 1,000 copies/ml, which was 100% in both studies [30, 40].
Liat HIV Quant.
Scott et al evaluated the performance of Liat HIV Quant compared to reference tests for VL monitoring. A high Pc concordance (Pearson’s correlation x a bias correlation factor) of 0.96 with the reference test was reported. Sensitivity was 100% at a threshold of 1000 copies/ml for both plasma and whole blood assays, but specificity at this threshold was only 41.2% for whole blood, compared to 88.2% for plasma assays [10]. Downward misclassification compared to reference results at the 1000 copies/ml threshold, falsely suggesting virologic suppression, was more frequent with whole blood compared to plasma samples [10].
Aptima HIV‐1 Quant Dx Assay.
High agreement and concordance were reported between Aptima Quant Dx assay and reference tests, ranging from 83.9%-95% [33, 37, 41, 42]. At a threshold of 50 copies/ml, agreement between the Aptima Quant Dx Assay and three reference assays (Abbott M2000 HIV-1 RealTime, Artus and Roche CAP CTMv2.0) ranged from 88.0%-95.0%. Agreement was highest with Abbott M2000 HIV-1 RealTime (k = 0.74) and lowest with Roche CAP CTMv2.0 (k = 0.50) [33]. Mor et al reported concordance at a threshold of 40 copies/ml between Aptima Quant Dx Assay, Xpert Quant, and two reference assays (NucliSENS EasyQ HIV‐1 v2.0 and Abbott M2000 HIV-1 RealTime), which ranged from 83.9%-89.8%. Concordance was highest between Aptima Quant Dx assay and Xpert Quant (91.4%) [37]. Two additional studies reported high agreement between the Aptima Quant Dx Assay and reference assays, at 30 copies/ml [41] and 50 copies/ml and 200 copies/ml (details in Table 1). No studies of Aptima Quant Dx reported sensitivity or specificity. In two studies of quantitative HIV-1 RNA levels, the Aptima Quant Dx assay yielded significantly higher results than Abbott M2000 HIV-1 RealTime [33, 37]; in another, a trend was observed for higher Aptima Quant Dx results relative to Roche CAP CTMv2.0 results [42]. In three studies of the mean difference between Aptima Quant Dx and reference test results, Aptima Quant Dx viral load values were 0.13 to 0.36 log copies/ml higher than the reference assays [37, 41, 42].
Error rates
Error rates were reported for four POC assays; Xpert Quant, Xpert Qual, Alere Detect and Liat HIV Quant. Error rates ranged between 2.0%–5.0% for Xpert Qual [14, 15, 18, 23] and between 2.5%–17.0% for Xpert Quant [14, 18, 31, 32, 35, 39], and were associated with inadequate sample volume (“error” result) [14, 15, 18, 23], incorrect sample processing or PCR inhibition (“invalid” result) [18], insufficient data collection (“no result”) [18], mechanical or cartridge errors resulting from faulty modules requiring replacement [15], incorrect pipette supply [15], device optic errors [23], probe check failures [31], and power outages [23]. Repeat testing resolved Xpert Qual errors in two studies [15, 23], with 10% of errors persisting in another [15]. In three studies using Xpert Quant [18, 32, 35], samples for which errors occurred could not be retested due to lack of extra plasma aliquots. In the study with the highest reported invalid rate (17%) for Xpert Quant [39], the error rate was attributed primarily to cartridges that were broken during shipment or were defective. For Alere Detect, error rates ranged from 3.1%–11.0% [16, 17, 19, 20, 23], including operator errors such as no or too little sample being detected, cartridge not properly locked or misaligned, and device errors, such as a connection error between the controller and processor and failure to read the barcode [23]. Specimens once retested were reported to produce valid results[23] or resolve the majority of the errors [20]. Error rates were observed to be higher in birth testing (infants<7 days) compared to testing in older infants [17, 20]. For the Liat HIV Quant, a 1.6% error rate was reported, all due to scanning errors [10].
Clinical utility
Four studies reported on clinical utility of POC results[14–17], and two studies used POC results for clinical management [14, 15]. Technau et al reported results of POC test implementation within an EID programme in Johannesburg, South Africa. Samples were obtained from HIV-exposed infants for POC testing with the Xpert Qual and compared to the local standard of care, which was HIV PCR testing using the Roche COBAS TaqMan HIV-1 Qualitative test (version 2·0, Roche Molecular Systems, Branchburg, NJ, USA). A positive test by either assay prompted ART initiation and confirmatory testing. A total of 30 neonates were diagnosed using the POC Xpert test, all of whom initiated ART. Time to result was reduced by POC testing (median 1 day vs. 10 days for HIV PCR), leading to more rapid ART initiation in infected neonates identified by the POC test [15].
Ndlovu et al evaluated the operational feasibility of integrated HIV VL, EID and MTB/RIF testing using the GeneXpert platform[14]. At three rural health facilities in Zimbabwe, whole blood samples were collected for HIV VL testing and DBS were collected from infants for EID. POC Xpert EID and VL testing had shorter median turnaround time for result delivery (1 day for each), compared to conventional centralised testing (17 and 26 days, respectively), substantially reducing time to ART initiation and decreasing patient loss to follow-up [14].
Jani et al measured the effect of POC EID on ART initiation rates and retention in care among HIV-positive infants in Mozambique [16]. POC EID facilitated rapid diagnosis and treatment of HIV-infected infants: 89.7% of HIV positive infants in the POC arm and 12.8% in the standard of care arm initiated ART within 60 days of sample collection. At 90 days of follow-up, 61.6% of those who initiated ART in POC arm and 42.9% in the standard of care arm were retained in care [16].
Meggi et al evaluated the feasibility, performance and diagnostic yield of rapid POC EID at birth within primary health care maternity wards in Mozambique [17]. Samples obtained from HIV-exposed infants were tested at birth and at 4–6 weeks using both POC Alere q HIV‐1/2 Detect Qual and Roche COBAS TaqMan HIV-1 Qualitative test (version 2·0). Sensitivity and specificity of POC testing compared with laboratory testing at birth were 100% (95% CI 89.4±100.0) and 100% (95% CI 99.8±100.0), respectively. Notable within the study were results of four infants who tested positive for HIV infection with laboratory-based and/or POC EID nucleic acid tests at birth but tested negative at least once during follow-up while on nevirapine (NVP) prophylaxis. For all four infants, results turned positive following NVP cessation, indicating that the diagnosis at birth was correct [17].
Provider experiences with POC assays
Cepheid GeneXpert HIV-1 Qual and HIV-1 Quant.
Table 2 outlines the provider experiences with POC assays as obtained from the included studies.
These assays were found to be simple to use [32, 35, 36, 43], with a rapid turnaround time of 90 minutes [18, 32, 36, 43], resulting in expeditious clinical decision-making [43]. Other advantages included: results that were easy to interpret [23], more efficient patient management compared to batch testing in a central laboratory [43], no requirement for calibrated mechanical pipettes, and the ability for less skilled personnel (potentially including non-laboratory personnel) to perform the test [15, 23, 35]. The GeneXpert platform was found to be compact in comparison to other platforms and more cost effective (≈17 USD per cartridge) [21, 36, 43]. The machine that takes a 4-cartridge module is easy to transport for use in various settings, an advantage for low-income countries [32]. The ability to run assays for tuberculosis and other pathogens in addition to HIV-1 was also advantageous [14, 18, 31, 36, 38], allowing for use of the same procurement chains and service plans for both TB and HIV diagnostics [18]. Of note, the Cepheid GeneXpert platform can be used in large laboratories with high volume, as configurations for up to 80 modules are available [31]. The modular nature of the equipment makes it possible to change one module if not working, avoiding a complete halt in the platform activities [18].
Drawbacks of the Xpert assays included the need for air-conditioning and refrigeration to maintain the platform and samples at correct temperatures [14] and the requirement for a back-up power source for the high-throughput modules when a steady power supply is not readily available [23]. In addition, overloading may occur when a large volume of tests is performed on a single machine and quality control requirements may exceed local capacity [14, 27]. Where implementation is planned for centralized testing, efficient and robust sample transport networks are needed [14]. Cartridge colours for the different Xpert assays are all the same, and therefore separate preparation areas in the laboratory are required to avoid confusion [14, 31]. Samples must also be manually added to the cartridge [23, 32]. Centrifugation is required to obtain plasma samples [14, 31, 38], and the 1000-μl plasma volume required could be a particular challenge for testing infants [38]. Finally, one study reported an instance of contamination of a DBS sample with a plasma VL sample, generating a false positive EID result and highlighting the risk of contamination with PCR assays [14].
Alere q HIV‐1/2 Detect Qual.
Advantages of this assay include: ease of use with a short run time of about 52 minutes [23], limited training required [16, 17, 19, 22], a small sample volume requirement (25μl) [19], the ability to print out test results, and the availability of a dedicated battery pack as an alternative backup power source [23]. Disadvantages include the ability to run only one sample at a time and difficulty using capillaries to load samples [23]. The cartridge ejection mechanism was reported to cause problems and the keyboard design was not considered user-friendly. The lack of information provided when an error message occurred was reported as a further disadvantage in one study; however, valid results were obtained on re-running the specimen [19].
SAMBA HIV Semiquantitative assay.
The short turnaround time of 90 minutes, heat stability, and limited training required have been reported as advantages [30, 40]. The visual detection of nucleic acid on a test strip was an added advantage, enabling staff to show result to the patient [40]. Limitations reported include inability to provide a quantitative result and requirement for a plasma sample, with no potential to use DBS or other sample types [30]. Despite a stronger signal on the test strip distinguishing positive results from negative results than present in current HIV rapid antibody tests, the possibility for misinterpretation of results by users remains [40, 44]. This issue can be overcome by having a paper printout of the test result, or having the results appear on a screen [40].
Liat HIV Quant.
This assay was found easy to use, with no support required for installation and only a brief 2-hour training needed [10]. The closed system requires no special safety precautions. Data can be exported from the device using a USB port or ethernet cable. The Liat platform can test 12–14 specimens in an 8-hour day, using either whole blood or plasma. Disadvantages reported include the short half-life of reagents, the need to maintain a cold chain for the testing cartridges, and the need for centrifugation when plasma is used[10].
Aptima HIV‐1 Quant Dx Assay.
The reported advantage of this assay is its ability to detect low HIV-1 copy numbers [33].
Quality assessment
Quality was assessed based on the 18 modified STARD criteria for the 30 included studies that reported on diagnostic accuracy (Table 3). Most (86.7%, n = 26) were easily classified as studies of diagnostic accuracy, with a clear objective of comparing the diagnostic accuracy of a POC assay to one or more laboratory-based reference assays. All studies stated whether or not their results were applied clinically. Few studies, however, specified inclusion or exclusion criteria (33.3%, n = 10), clinical and demographic data for the population from which samples were obtained (36.6%, n = 11), data collection procedures (16.7%, n = 5), or training requirements for the POC assay evaluated (26.7%, n = 8).
Discussion
This systematic review of the literature aimed to synthesize evidence on the performance and clinical utility of POC quantitative or qualitative HIV-1 RNA testing assays for EID, AHI diagnosis and VL monitoring, and to identify barriers and facilitators to their scale-up in resource-limited settings. We found 32 studies that met inclusion criteria, of which 30 focused on diagnostic accuracy and 4 included results on clinical utility. Overall, the studies of diagnostic accuracy showed excellent performance. Where correlation between quantitative results was reported, it was high across all assays assessed; in addition, inter-assay differences reported were <0.5 log copies/ml, which is considered for clinical practice [45, 46]. Although POC VL assays tended to overestimate virologic failure compared to reference tests, which could lead to early switching, the use of central laboratory testing to confirm treatment failure could overcome this disadvantage [48]. In addition, while sensitivity for birth testing in infants <7 days of age was lower than for routine EID testing at 6–14 weeks, earlier diagnosis for those infants with positive POC results provides important advantages [14, 16]. Given the increasing body of evidence on the diagnostic accuracy of POC RNA testing, studies on clinical utility, implementation barriers and facilitators, and cost-effectiveness should be the focus of future research.
The WHO recommends HIV-1 testing of exposed infants at the earliest opportunity with an assay that detects HIV-1 DNA or RNA, by 4–6 weeks of age at the latest [49]. Without ART, about 50% of perinatally-infected infants progress to advanced disease by 8–12 weeks of age or die [50, 51]. Unfortunately, only half of all HIV-exposed infants in RLS receive an EID test within 2 months of age [49]. Challenges with current EID programs include loss to follow-up of 30%-80% of mother-infant pairs and late presentation of many infants, who miss out on the benefits of early ART initiation as well as lack of diversified testing and sample collection sites [49, 52]. In the two studies that used POC EID results, turnaround time was reduced to one day, leading to rapid ART initiation and reduced loss to follow-up [14, 15]. With the scale-up of prevention of mother to child transmission (PMTCT) programmes in sub-Saharan Africa, where all EID studies in this review were conducted, POC testing by non-specialized personnel in field settings could help decentralize services and improve infant outcomes [19]. Integration of POC HIV testing services for women and infants in other high-yield settings for paediatric HIV case-finding, including TB clinics, malnutrition clinics and inpatient wards, could lead to the identification of HIV-exposed or -infected infants missed by routine PMTCT programmes [52]. While effects of infant NVP prophylaxis on birth testing outcomes and algorithms for confirmatory testing before ART initiation still need to be addressed [17], POC testing for EID holds promise.
The WHO estimates that by mid-2016, more than 18 million HIV-infected individuals were receiving ART, with access increasing due to “test and treat” approaches [53, 54]. With continued scale-up, virologic monitoring to ensure treatment efficacy and combat HIV drug resistance is necessary [53, 54], particularly in low- and middle-income countries, where delayed treatment and poor care engagement can result from financial, human resource, and infrastructural barriers [55, 56]. Currently, only ≈20% of ART patients in low-middle income countries receive VL testing [54, 57, 58]. This low coverage has been attributed to a number of challenges, including poor sample referral systems, a lack of electronic data systems for results, and long turnaround times resulting in patient loss to follow-up [56].Improved health data systems are needed to flag those in need of VL testing and ensure fast turnaround times for prompt clinical decision-making [56, 58]. POC VL testing could meet a critical need, by ensuring same-day results for providers in rural or hard-to-reach areas where VL test access is currently limited [57, 58]. The advantages of POC VL assays may outweigh concerns about upward misclassification, especially as the current tendency is to switch too late, rather than too early [56]. Centralized confirmatory testing with standard-of-care assays for patients with suspected virological failure could address concerns about false positive results.
AHI diagnosis is a concern that has often been overlooked in RLS [59]. An estimated 10%–50% of all HIV transmission events may be attributable to AHI, a period associated with high transmission risk due to extremely high viral load and high infectivity of founder viruses [60]. Despite the importance of diagnosing AHI, there is currently no WHO recommendation on AHI diagnosis, and very few studies have considered this application of POC HIV-1 RNA assays [48]. New WHO recommendations to exclude acute or early HIV infection prior to initiating PrEP (pre-exposure prophylaxis) or PEP (post-exposure prophylaxis)[48] may provide the impetus needed to investigate the utility of POC HIV-1 RNA assays for AHI diagnosis among patients with symptoms of acute retroviral syndrome. In our review, we found two studies investigating performance of the Cepheid GeneXpert HIV-1 Qual for AHI diagnosis. Additional studies are on the horizon, including the Tambua Mapema Plus study, a proof-of-concept study evaluating the impact of an HIV-1 RNA testing intervention (Cepheid GeneXpert HIV-1 Qual) targeting young adult Kenyan patients aged 18–39 years who seek urgent care for symptoms associated with AHI (ClinicalTrials.gov Identifier: NCT03508908) [61]. Prompt diagnosis of AHI is needed to maximize the benefits of a test and treat approach [59]. If AHI diagnosis is augmented by assisted partner notification with HIV-1 RNA testing, partners with acute or prevalent HIV infection could also be identified and linked to ART if infected or PrEP if uninfected, maximizing impact.
Various factors should be considered prior to implementation and scale-up of POC HIV-1 RNA testing. Staff will require training and rigorous quality control measures should be put in place. Costs for reagents and consumables, shipment, customs charges, tax, service, and maintenance remain important considerations, which we were unable to address in the current review given the paucity of published data on this aspect of POC HIV-1 RNA tests. However, 2017 data from the Global Fund suggests the maximum price per test including consumables is $10.60 per test for the Aptima Quant Dx assay for VL, $16.80 per test for Xpert Quant for VL, $17.95 per tests for Xpert Qual for EID, up to $25 per test for Alere Q for EID, and $37.40 each for SAMBA I and II for VL and EID [62]. The cost of the POC-of-care equipment excluding service and maintenance varies greatly; from $12,280 for the GeneXpert IV-2 platform, $24,800 for SAMBA II, and $25,000 for Alere Q, up to $71,500 for the GeneXpert XVI and $72,000 for SAMBA I, and as high as $150,000 for the Hologic panther system on which the Aptima Quant Dx assay operates [62]. The primary strategy proposed to mitigate cost constraints is pooled procurement through PEPFAR or the Global Fund, with the hope that eventually higher demand will bring about competition and drive down costs [57].
Our review has a number of limitations. First, only published studies, of which the majority were field reviews, were included. Very few studies reported on clinical utility of the POC devices. Second, conference abstracts were excluded due to insufficient detail and inability to assess the quality of the study. This resulted in the exclusion of some relevant studies. Third, pricing data are often difficult to find and can change rapidly. Lastly, there was variability in the measures of accuracy, precision and agreement reported across the studies, which did not allow for pooling of data and limits its generalisability.
Conclusion
This systematic review has identified a number of studies investigating POC HIV-1 RNA assays in RLS, of which most demonstrate acceptable clinical accuracy. Very few studies have investigated clinical utility and strategies for scale-up. As POC HIV-1 RNA assays are more widely evaluated for the uses discussed in this review, the requirement for plasma samples and thus continued need for trained phlebotomists and centrifugation of samples may remain a barrier in some RLS [57]. In addition, the need for additional resources such as air-conditioning, cold chain for reagents and a back-up power source may prove an additional challenge. If POC HIV testing is considered as part of an integrated laboratory network, strong tracking systems, good documentation, and robust sample transport and supply chain systems are needed. Moving forward, further research is needed on clinical utility, quality assurance, algorithms for confirming positive results, reliability of results for clinical decision-making, and cost-effectiveness [63]. In general, however, it is clear that POC HIV-1 RNA assays are here to stay and offer clear advantages that will help advance HIV prevention and care globally.
Supporting information
S1 File. Search strategy for systematic review.
https://doi.org/10.1371/journal.pone.0218369.s001
(PDF)
References
- 1. Niemz A, Ferguson TM, Boyle DS. Point-of-care nucleic acid testing for infectious diseases. Trends in Biotechnology. 2011;29(5):240–50. pmid:21377748
- 2. Usdin M, Guillerm M, Calmy A. Patient needs and point-of-care requirements for HIV load testing in resource-limited settings. Journal of infectious diseases. 2010;201(Supplement 1):S73–S7.
- 3.
UNITAID WHO. HIV/AIDS Diagnostics Technology Landscape 5th edition October 2015 [cited 2017 06 Feb 2017 ]. Available from: https://aidsfree.usaid.gov/sites/default/files/hivaids_diag_tech.pdf.
- 4. Shafiee H, Wang S, Inci F, Toy M, Henrich TJ, Kuritzkes DR, et al. Emerging technologies for point-of-care management of HIV infection. Annual review of medicine. 2015;66:387–405. pmid:25423597
- 5. Rosenberg NE, Kamanga G, Phiri S, Nsona D, Pettifor A, Rutstein SE, et al. Detection of Acute HIV Infection: A Field Evaluation of the Determine® HIV-1/2 Ag/Ab Combo Test. The Journal of Infectious Diseases. 2012;205(4):528–34. pmid:22207651
- 6. Lewis JM, Macpherson P, Adams ER, Ochodo E, Sands A, Taegtmeyer M. Field accuracy of 4th generation rapid diagnostic tests for acute HIV-1: a systematic review. AIDS (London, England). 2015;29(18):2465–71. PMC4645957. pmid:26558545
- 7.
WHO. WHO list of prequalified in vitro diagnostic products 2019 [updated 04 April 2019; cited 2019 06 May]. Available from: https://www.who.int/diagnostics_laboratory/evaluations/190404_prequalified_product_list.pdf.
- 8.
WHO. WHO Prequalification of In Vitro Diagnostics Public Report: m-PIMA HIV-1/2 VL 2019 [updated April 2019; cited 2019 06 May]. Available from: https://www.who.int/diagnostics_laboratory/evaluations/pq-list/190408_pqdx_0359_032_00_pqpr_mpima.pdf.
- 9.
WHO. Virological Technologies: progress of the active applications in the prequalification of IVDs assessment pipeline 2019 [updated 23 April 2019; cited 2019 06 May]. Available from: https://www.who.int/diagnostics_laboratory/190423_vl.pdf?ua=1.
- 10. Scott L, Gous N, Carmona S, Stevens W. Laboratory evaluation of the Liat HIV Quant (IQuum) whole-blood and plasma HIV-1 viral load assays for point-of-care testing in South Africa. Journal of clinical microbiology. 2015;53(5):1616–21. pmid:25740777
- 11. Nash M, Huddart S, Badar S, Baliga S, Saravu K, Pai M. Performance of the Xpert® HIV-1 Viral Load assay: A systematic review and meta-analysis. Journal of clinical microbiology. 2018:JCM. 01673–17.
- 12. Ritchie AV, Goel N, Sembongi H, Lehga J, Farleigh LE, Edemaga D, et al. Performance evaluation of the point-of-care SAMBA I and II HIV-1 Qual whole blood tests. Journal of Virological Methods. 2016;237:143–9. pmid:27568275
- 13. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. BMJ: British Medical Journal. 2003;326(7379):41–4. PMC1124931. pmid:12511463
- 14. Ndlovu Z, Fajardo E, Mbofana E, Maparo T, Garone D, Metcalf C, et al. Multidisease testing for HIV and TB using the GeneXpert platform: A feasibility study in rural Zimbabwe. PloS one. 2018;13(3):e0193577. pmid:29499042
- 15. Technau K-G, Kuhn L, Coovadia A, Murnane PM, Sherman G. Xpert HIV-1 point-of-care test for neonatal diagnosis of HIV in the birth testing programme of a maternity hospital: a field evaluation study. The Lancet HIV. 2017;4(10):e442–e8. pmid:28711526
- 16. Jani IV, Meggi B, Loquiha O, Tobaiwa O, Mudenyanga C, Zitha A, et al. Effect of point-of-care early infant diagnosis on antiretroviral therapy initiation and retention of patients. AIDS. 2018;32(11):1453–63. 00002030-201807170-00008. pmid:29746301
- 17. Meggi B, Vojnov L, Mabunda N, Vubil A, Zitha A, Tobaiwa O, et al. Performance of point-of-care birth HIV testing in primary health care clinics: An observational cohort study. PLOS ONE. 2018;13(6):e0198344. pmid:29912987
- 18. Ceffa S, Luhanga R, Andreotti M, Brambilla D, Erba F, Jere H, et al. Comparison of the Cepheid GeneXpert and Abbott M2000 HIV-1 real time molecular assays for monitoring HIV-1 viral load and detecting HIV-1 infection. Journal of virological methods. 2016;229:35–9. pmid:26709099
- 19. Dunning L, Kroon M, Hsiao N-y, Myer L. Field evaluation of HIV point-of-care testing for early infant diagnosis in Cape Town, South Africa. PloS one. 2017;12(12):e0189226. pmid:29261707
- 20. Hsiao N-y, Dunning L, Kroon M, Myer L. Laboratory evaluation of the Alere q point-of-care system for early infant HIV diagnosis. PloS one. 2016;11(3):e0152672. pmid:27032094
- 21. Ibrahim M, Moyo S, Mohammed T, Mupfumi L, Gaseitsiwe S, Maswabi K, et al. Brief Report: High Sensitivity and Specificity of the Cepheid Xpert HIV-1 Qualitative Point-of-Care Test Among Newborns in Botswana. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2017;75(5):e128–e31. pmid:28350554
- 22. Jani IV, Meggi B, Mabunda N, Vubil A, Sitoe NE, Tobaiwa O, et al. Accurate early infant HIV diagnosis in primary health clinics using a point-of-care nucleic acid test. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2014;67(1):e1–e4. pmid:24933096
- 23. Murray TY, Sherman GG, Nakwa F, MacLeod WB, Sipambo N, Velaphi S, et al. Field Evaluation of Performance of Alere and Cepheid Qualitative HIV Assays for Pediatric Point-of-Care Testing in an Academic Hospital in Soweto, South Africa. Journal of clinical microbiology. 2017;55(11):3227–35. pmid:28855305
- 24. Ondiek J, Namukaya Z, Mtapuri-Zinyowera S, Balkan S, Elbireer A, Lumb IU, et al. Multicountry Validation of SAMBA-A Novel Molecular Point-of-Care Test for HIV-1 Detection in Resource-Limited Setting. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2017;76(2):e52–e7. pmid:28902680
- 25. Michaeli M, Wax M, Gozlan Y, Rakovsky A, Mendelson E, Mor O. Evaluation of xpert HIV-1 qual assay for resolution of HIV-1 infection in samples with negative or indeterminate geenius HIV-1/2 results. Journal of Clinical Virology. 2016;76:1–3. pmid:26761545
- 26. Rakovsky A, Gozlan Y, Bassal R, Wax M, Shirazi R, Bakhanashvili M, et al. Diagnosis of HIV-1 infection: Performance of Xpert Qual and Geenius supplemental assays in fourth generation ELISA-reactive samples. Journal of Clinical Virology. 2018;101:7–10. pmid:29414189
- 27. Garrett NJ, Drain P, Werner L, Samsunder N, Karim SSA. Diagnostic Accuracy of the Point-of-care Xpert® HIV-1 Viral Load Assay in a South African HIV clinic. Journal of acquired immune deficiency syndromes (1999). 2016;72(2):e45.
- 28. Avidor B, Matus N, Girshengorn S, Achsanov S, Gielman S, Zeldis I, et al. Comparison between Roche and Xpert in HIV-1 RNA quantitation: A high concordance between the two techniques except for a CRF02_AG subtype variant with high viral load titters detected by Roche but undetected by Xpert. Journal of Clinical Virology. 2017;93:15–9. pmid:28564629
- 29. Bruzzone B, Caligiuri P, Nigro N, Arcuri C, Delucis S, Di Biagio A, et al. Xpert HIV-1 Viral Load Assay and VERSANT HIV-1 RNA 1.5 Assay: A Performance Comparison. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2017;74(3):e86–e8. pmid:27828879
- 30. Goel N, Ritchie AV, Mtapuri-Zinyowera S, Zeh C, Stepchenkova T, Lehga J, et al. Performance of the SAMBA I and II HIV-1 Semi-Q Tests for viral load monitoring at the point-of-care. Journal of virological methods. 2017;244:39–45. pmid:28274744
- 31. Gous N, Scott L, Berrie L, Stevens W. Options to expand HIV viral load testing in South Africa: evaluation of the GeneXpert® HIV-1 viral load assay. PloS one. 2016;11(12):e0168244. pmid:27992495
- 32. Gueudin M, Baron A, Alessandri-Gradt E, Lemée V, Mourez T, Etienne M, et al. Performance evaluation of the new HIV-1 quantification assay, Xpert HIV-1 viral load, on a wide panel of HIV-1 variants. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2016;72(5):521–6. pmid:27007866
- 33. Hopkins M, Hau S, Tiernan C, Papadimitropoulos A, Chawla A, Beloukas A, et al. Comparative performance of the new Aptima HIV-1 Quant Dx assay with three commercial PCR-based HIV-1 RNA quantitation assays. Journal of Clinical Virology. 2015;69:56–62. pmid:26209380
- 34. Jani IV, Meggi B, Vubil A, Sitoe NE, Bhatt N, Tobaiwa O, et al. Evaluation of the whole-blood Alere Q NAT point-of-care RNA assay for HIV-1 viral load monitoring in a primary health care setting in Mozambique. Journal of clinical microbiology. 2016;54(8):2104–8. pmid:27252459
- 35. Jordan JA, Plantier J, Templeton K, Wu A. Multi-site clinical evaluation of the Xpert® HIV-1 viral load assay. Journal of Clinical Virology. 2016;80:27–32. pmid:27135387
- 36. Kulkarni S, Jadhav S, Khopkar P, Sane S, Chimanpure V, Dhilpe V, et al., editors. GeneXpert HIV-1 Quant: a tool for monitoring the success of ART programme in developing countries. JOURNAL OF THE INTERNATIONAL AIDS SOCIETY; 2016: INT AIDS SOCIETY AVENUE DE FRANCE 23, GENEVA, 1202, SWITZERLAND.
- 37. Mor O, Gozlan Y, Wax M, Mileguir F, Rakovsky A, Noy B, et al. Evaluation of the RealTime HIV-1, Xpert HIV-1, and Aptima HIV-1 quant Dx assays in comparison to the NucliSens EasyQ HIV-1 v2. 0 assay for quantification of HIV-1 viral load. Journal of clinical microbiology. 2015;53(11):3458–65. pmid:26292298
- 38. Moyo S, Mohammed T, Wirth KE, Prague M, Bennett K, Holme MP, et al. Point-of-care Cepheid Xpert HIV-1 viral load test in rural African communities is feasible and reliable. Journal of clinical microbiology. 2016;54(12):3050–5. pmid:27733636
- 39. Nash M, Ramapuram J, Kaiya R, Huddart S, Pai M, Baliga S. Use of the GeneXpert tuberculosis system for HIV viral load testing in India. The Lancet Global Health. 2017;5(8):e754–e5. pmid:28716346
- 40. Ritchie AV, Ushiro-Lumb I, Edemaga D, Joshi HA, De Ruiter A, Szumilin E, et al. SAMBA HIV semiquantitative test, a new point-of-care viral-load-monitoring assay for resource-limited settings. Journal of clinical microbiology. 2014;52(9):3377–83. pmid:25031444
- 41. Schalasta G, Börner A, Speicher A, Enders M. Comparative evaluation of the Aptima HIV-1 Quant Dx assay and COBAS TaqMan HIV-1 v2. 0 assay using the Roche High Pure System for the quantification of HIV-1 RNA in plasma. Clinical Chemistry and Laboratory Medicine (CCLM). 2016;54(3):493–9. pmid:26351942
- 42. Schønning K, Johansen K, Landt B, Benfield T, Westh H. Comparison of the Hologic Aptima HIV-1 Quant Dx Assay to the Roche COBAS Ampliprep/COBAS TaqMan HIV-1 Test v2. 0 for the quantification of HIV-1 RNA in plasma samples. Journal of Clinical Virology. 2017;92:14–9. pmid:28505569
- 43. Swathirajan CR, Vignesh R, Boobalan J, Solomon SS, Saravanan S, Balakrishnan P. Performance of point-of-care Xpert HIV-1 plasma viral load assay at a tertiary HIV care centre in Southern India. Journal of medical microbiology. 2017;66(10):1379–82. pmid:28901908
- 44. Titchmarsh L, Zeh C, Verpoort T, Allain J-P, Lee H. Leukodepletion as a point-of-care method for monitoring HIV-1 viral load in whole blood. Journal of clinical microbiology. 2015;53(4):1080–6. pmid:25428162
- 45. Saag MS, Holodniy M, Kuritzkes D, O'Brien W, Coombs R, Poscher M, et al. HIV viral load markers in clinical practice. Nature medicine. 1996;2(6):625–9. pmid:8640545
- 46. Sollis KA, Smit PW, Fiscus S, Ford N, Vitoria M, Essajee S, et al. Systematic Review of the Performance of HIV Viral Load Technologies on Plasma Samples. PLOS ONE. 2014;9(2):e85869. pmid:24558359
- 47.
WHO. WHO Prequalification of In Vitro Diagnostics PUBLIC REPORT Product: Xpert® HIV-1 Viral Load 2017 [cited 2018 17 April ]. Available from: http://www.who.int/diagnostics_laboratory/evaluations/pq-list/hiv-vrl/170720_final_pq_report_pqdx_0192_0193_0194_0195_070-00.pdf?ua=1.
- 48.
WHO. CONSOLIDATED GUIDELINES ON THE USE OF ANTIRETROVIRAL DRUGS FOR TREATING AND PREVENTING HIV INFECTION 2016 [cited 2018 16 April]. Second:[Available from: http://apps.who.int/iris/bitstream/handle/10665/208825/9789241549684_eng.pdf?sequence=1.
- 49. Essajee S, Bhairavabhotla R, Penazzato M, Kiragu K, Jani I, Carmona S, et al. Scale-up of Early Infant HIV Diagnosis and Improving Access to Pediatric HIV Care in Global Plan Countries: Past and Future Perspectives. JAIDS Journal of Acquired Immune Deficiency Syndromes. 2017;75:S51–S8. 00126334-201705011-00008. pmid:28398997
- 50. Marston M, Becquet R, Zaba B, Moulton LH, Gray G, Coovadia H, et al. Net survival of perinatally and postnatally HIV-infected children: a pooled analysis of individual data from sub-Saharan Africa. International Journal of Epidemiology. 2011;40(2):385–96. pmid:21247884
- 51. Innes S, Lazarus E, Otwombe K, Liberty A, Germanus R, Van Rensburg AJ, et al. Early severe HIV disease precedes early antiretroviral therapy in infants: Are we too late? Journal of the International AIDS Society. 2014;17(1).
- 52. Karidia D, Surbhi M, Mackenzie H, Suzanne BR, NJ N. A Proposed Framework for the Implementation of Early Infant Diagnosis Point-of-Care. AIDS Research and Human Retroviruses. 2017;33(3):203–10. pmid:27758117.
- 53.
WHO. PROGRESS REPORT 2016: PREVENT HIV, TEST AND TREAT ALL 2016 [cited 2018 10 April]. Available from: http://apps.who.int/iris/bitstream/handle/10665/251713/WHO-HIV-2016.24-eng.pdf?sequence=1.
- 54.
WHO. Global Action Plan on HIV Drug Resistance 2017–2021: 2018 Progress Report: WHO; 2018 [updated July cited 2018 09 October ]. Available from: http://apps.who.int/iris/bitstream/handle/10665/273049/WHO-CDS-HIV-18.12-eng.pdf?ua=1.
- 55.
WHO. Low-cost tools for diagnosing and monitoring HIV infection in low-resource settings 2012 [cited 2018 14 April]. Available from: http://www.who.int/bulletin/volumes/90/12/12-102780/en/.
- 56. Peter T, Zeh C, Katz Z, Elbireer A, Alemayehu B, Vojnov L, et al. Scaling up HIV viral load–lessons from the large-scale implementation of HIV early infant diagnosis and CD4 testing. Journal of the International AIDS Society. 2017;20(S7):e25008. pmid:29130601
- 57. Roberts T, Cohn J, Bonner K, Hargreaves S. Scale-up of Routine Viral Load Testing in Resource-Poor Settings: Current and Future Implementation Challenges. Clinical Infectious Diseases. 2016;62(8):1043–8. pmid:26743094
- 58.
MSF. Making Viral Load Routine 2016 [cited 2018 09 October ]. Available from: https://www.msf.org/sites/msf.org/files/making_viral_load_routine_part_1_programmatic_strategies.pdf.
- 59. Powers KA, Cohen MS. Acute HIV-1 infection in sub-Saharan Africa: a common occurrence overlooked. AIDS. 2014;28(9):1365–7. 00002030-201406010-00015. pmid:24959964
- 60. Rutstein SE, Ananworanich J, Fidler S, Johnson C, Sanders EJ, Sued O, et al. Clinical and public health implications of acute and early HIV detection and treatment: a scoping review. Journal of the International AIDS Society. 2017;20(1):21579. PMC5515019. pmid:28691435
- 61. Sanders EJ, Wahome E, Powers KA, Werner L, Fegan G, Lavreys L, et al. Targeted screening of at-risk adults for acute HIV-1 infection in sub-Saharan Africa. AIDS (London, England). 2015;29(0 3):S221.
- 62.
Fund TG. HIV Viral Load and Early Infant Diagnosis Selection and Procurement Information Tool 2017 [updated April 2017; cited 2018 18 October ]. Available from: https://www.theglobalfund.org/media/5765/psm_viralloadearlyinfantdiagnosis_content_en.pdf?u=636737337280000000.
- 63. Stevens W, Gous N, Ford N, Scott LE. Feasibility of HIV point-of-care tests for resource-limited settings: challenges and solutions. BMC Medicine. 2014;12(1):173. pmid:25197773