https://orcid.org/0000-0002-3289-6011
Correction
4 Aug 2023: Yufenyuy EL, Detorio M, Dobbs T, Patel HK, Jackson K, et al. (2023) Correction: Performance evaluation of the Asante Rapid Recency Assay for verification of HIV diagnosis and detection of recent HIV-1 infections: Implications for epidemic control. PLOS Global Public Health 3(8): e0002287. https://doi.org/10.1371/journal.pgph.0002287 View correction
Figures
Abstract
We previously described development of a rapid test for recent infection (RTRI) that can diagnose HIV infection and detect HIV-1 recent infections in a single device. This technology was transferred to a commercial partner as Asante Rapid Recency Assay (ARRA). We evaluated performance of the ARRA kits in the laboratory using a well-characterized panel of specimens. The plasma specimen panel (N = 1500) included HIV-1 (N = 570), HIV-2 (N = 10), and HIV-negatives (N = 920) representing multiple subtypes and geographic locations. Reference diagnostic data were generated using the Bio-Rad HIV-1-2-O EIA/Western blot algorithm with further serotyping performed using the Multispot HIV-1/2 assay. The LAg-Avidity EIA was used to generate reference data on recent and long-term infection for HIV-1 positive specimens at a normalized optical density (ODn) cutoff of 2.0 corresponding to a mean duration of about 6 months. All specimens were tested with ARRA according to the manufacturer’s recommendations. Test strips were also read for line intensities using a reader and results were correlated with visual interpretation. ARRA’s positive verification line (PVL) correctly classified 575 of 580 HIV-positive and 910 of 920 negative specimens resulting in a sensitivity of 99.1% (95% CI: 98.0–99.6) and specificity of 98.9% (95% CI: 98.1–99.4), respectively. The reader-based classification was similar for PVL with sensitivity of 99.3% (576/580) and specificity of 98.8% (909/920). ARRA’s long-term line (LTL) classified 109 of 565 HIV-1 specimens as recent and 456 as long-term compared to 98 as recent and 467 as long-term (LT) by LAg-Avidity EIA (cutoff ODn = 2.0), suggesting a mean duration of recent infection (MDRI) close to 6 months. Agreement of ARRA with LAg recent cases was 81.6% (80/98) and LT cases was 93.8% (438/467), with an overall agreement of 91.7% (kappa = 0.72). The reader (cutoff 2.9) classified 109/566 specimens as recent infections compared to 99 by the LAg-Avidity EIA for recency agreement of 81.8% (81/99), LT agreement of 9% (439/467) with overall agreement of 91.9% (kappa = 0.72). The agreement between visual interpretation and strip reader was 99.9% (95% CI: 99.6–99.9) for the PVL and 98.1% (95% CI: 96.6–98.9) for the LTL. ARRA performed well with HIV diagnostic sensitivity >99% and specificity >98%. Its ability to identify recent infections is comparable to the LA-Avidity EIA corresponding to an MDRI of about 6 months. This point-of-care assay has implications for real-time surveillance of new infections among newly diagnosed individuals for targeted prevention and interrupting ongoing transmission thus accelerating epidemic control.
Citation: Yufenyuy EL, Detorio M, Dobbs T, Patel HK, Jackson K, Vedapuri S, et al. (2022) Performance evaluation of the Asante Rapid Recency Assay for verification of HIV diagnosis and detection of recent HIV-1 infections: Implications for epidemic control. PLOS Glob Public Health 2(5): e0000316. https://doi.org/10.1371/journal.pgph.0000316
Editor: Sanghyuk Shin, University of California Irvine, UNITED STATES
Received: October 19, 2021; Accepted: March 9, 2022; Published: May 3, 2022
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All data are available as a supplementary Excel file.
Funding: Funding for this research has been supported by the President’s Emergency Plan for AIDS Relief (PEPFAR) through the Centers for Disease Control and Prevention (CDC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: I have read the journal’s policy and the authors have the following competing interests. As an inventor of rapid test for recent infection, BSP receives a portion of royalties from the sale of Asante Rapid Recency Assay as per policy of the U.S. Government. The assay is manufactured by Sedia BioSciences under a technology transfer licensing agreement with CDC. This does not compromise author’s obligation to ensure accuracy and integrity of the data presented in this manuscript.
Introduction
The HIV pandemic has been with us for more than 40 years and global HIV burden is now estimated at about 38 million people living with HIV [1]. However, with major investments and commitment from national and international partners, including Ministries of Health, most countries have made major strides in changing the trajectory of the epidemic in most populations and countries in the last 15–20 years. Globally, we have scaled up HIV testing, accelerated prevention of mother-to-child transmission, developed new and effective treatments, placed millions of people on antiretroviral treatment (ART) and innovated novel prevention approaches, resulting in significant reduction in estimated new infections from 2.7 million in 2000 to 1.7 million in 2019 [1]. However, the progress is varied with some countries reaching UNAIDS 90-90-90 goals while others are still working to achieve these goals [2].
Identifying ongoing HIV transmission and targeting prevention strategies for populations with high incidence are important for accelerating epidemic control [3, 4]. This can be best achieved by using tests that can detect recently infected individuals [5–8]. Considerable efforts and resources have been devoted to the development of laboratory assays to detect recent HIV-1 infections [9–23]. However, only two of these assays, both developed in our laboratory, have been commercialized for specific application to estimate HIV-1 incidence in cross-sectional population: the BED-Capture Enzyme Immunoassay (BED-CEIA) [24] and the Limiting Antigen (LAg) Avidity EIA [17]. The LAg-Avidity EIA was developed and commercialized in 2012 to address some of the limitations of the BED assay and is widely used in many countries worldwide [17, 25–34]. Most recently, the LAg-Avidity EIA has been used in population-based HIV impact assessment (PHIA) surveys as a key methodology to estimate incidence in several countries [35, 36]. The LAg-Avidity EIA uses a multi-subtype gp41 protein, rIDR-M, covering the immunodominant region of gp41 of HIV-1, which permits equivalent detection of antibody avidity from all major subtypes [20]. The assay is based on a novel approach of using limiting amount of antigen thereby limiting the ability of low avidity antibodies (usually from recent infection) to bind, while allowing high avidity antibodies present during long term infections to bind [17, 20].
Until recently, the utility of incidence assays was limited to the estimation of HIV incidence from cross-sectional population surveys, which may also permit association of risk factors and identification of populations with high incidence. The utility of incidence assays has evolved to include the identification of recently infected individuals in some European countries for contact tracing and index testing [37, 38]. However, such application of laboratory-based tests for recent infection have been limited due to the delay in results since these assays are conducted in a central laboratory and results may not be available for several days or weeks. Development of a rapid test for recent infection allows one to perform HIV recency testing in the routine HIV program in real time and utilize data for targeted prevention [18]. Our laboratory developed a rapid test for recent infection [18] which now has been commercialized by Sedia Biosciences (Portland, OR) as Asante Rapid Recency Assay (ARRA).
The ARRA is a strip-test that combines HIV diagnosis with marker of time since infection as previously described [18]. This is achieved by including an additional line where antigen is striped at limiting antigen concentration to distinguish recent from long term infection. The principle of the assay is similar to the laboratory-based LAg-Avidity EIA and uses the same rIDR-M antigen but in a rapid test format. The lines can be read visually as well as with a strip reader, which produces quantitative results in intensity units (IU). However, ARRA has been not evaluated using a large panel of specimens that mimic cross-sectional specimens. We describe here the performance evaluation of this test kit using a large well-characterized global panel of specimens. The purpose of this evaluation was to compare the performance of both the PVL and the LTL on the Asante strips to the standard HIV diagnostic algorithm and the LAg-Avidity EIA for recency classification, respectively.
Methods
Specimens
Our world-wide panel of plasma specimens (N = 1500; HIV-1 = 570, HIV-2 = 10, Neg = 920) was derived from rejected blood units from adult donors in blood banks from Kenya, Uganda, Cameroon, Cote d’Ivoire, Sierra Leone, South Africa, Thailand, and the United States, collected in 2005–2006 for validation of HIV rapid tests in support of President’s Emergency Plan for AIDS Relief (PEPFAR) under a CDC approved protocol [39]. Blood units were rejected due their reactivity to one or more pathogens (HIV, Syphilis, Hepatitis, etc.) that could be potentially transmitted by blood transfusion. The protocol was approved by the CDC as non-research activity and did not require full IRB review due to use of unlinked blood donor specimens for test evaluations. Informed consent was obtained from the donors for use of their specimens for quality assurance and test evaluations. De-identified blood units were separated into plasma at site, frozen at -20± C, and transported by air to our laboratory on dry ice. Upon receipt, unlinked specimen IDs were matched with specimen log, given a new CDC ID, and aliquoted for long-term storage at -70± C. The specimens were characterized for the presence or absence of HIV antibodies using the 3rd generation Genetic Systems Enzyme Immunoassay (EIA) (Bio-Rad Laboratory, Hercules, CA) followed by Cambridge Biotech HIV-1 Western blot assay (Maxim, Rockville, MD) for confirmation. HIV-1 and HIV-2 serotyping was done using the Multispot HIV-1/2 (Bio-Rad) assay. The HIV-positive specimens were inherently made up of different subtypes and recombinants due to diverse geographic origins, although sequencing was not performed. Additional characterization of HIV-1 positive specimens was done using the LAg-Avidity EIA (Sedia Biosciences, Portland, OR) to classify specimens as either recent or long-term infections. The specimens were not tested for anti-retroviral (ARV) drugs but were likely from ARV naïve individuals because they were collected from blood donors in 2005–2006 when treatment coverage was low in most countries in sub-Saharan Africa. The panel consisted of HIV positive (n = 580; of which 570 were HIV-1 and 10 were HIV-2), and HIV negative specimens (n = 920). The HIV-1 specimens (n = 570) included 103 (18.1%) recent and 467 (81.9%) long-term infections as classified by the LAg-Avidity EIA at the threshold cutoff of 2.0 ODn, matching the MDRI of about 6 months [25]. No other clinical information was available for individual donors.
Assay procedure
The ARRA was performed according to the manufacturer’s instructions [40]. Briefly, serum or plasma specimens were brought to room temperature and approximately 5μl of sample was collected using the specimen collection loop provided in the kit and transferred into a buffer tube containing 0.5mL of buffer and mixed. After discarding the specimen collection loop, a test strip was dropped into the tube and incubated for 20 minutes. Developed test strips were removed after 20 min, dabbed onto filter paper to remove excess buffer and then read visually as recent, long-term, or HIV negative followed by reading with Asante strip reader by the same technician (see below). Visual interpretation was done first to avoid any bias that may result from reader-based interpretation. The results were recorded in a worksheet, transferred to a central database and analyzed (see Data and Statistical Analysis sections below).
Interpretation of results
Visually, the presence of all three lines [control line (CL), positive verification line (PVL), and long-term line (LTL)] indicated HIV-positive specimen with long term infection; presence of only two lines (CL and PVL) indicated HIV-positive specimens with recent infection; and presence of only CL indicated a seronegative specimen (Fig 1). The Asante strip reader measures the intensity of the three lines whose values are read in intensity units (IU). The reader cutoff for the PVL was 2.8 IU, where IU≥2.8 were HIV-reactive/positive and IU<2.8 were HIV negative. For the separation of recent from long term infections we used a cutoff of 2.9 IU for the LTL because it matched well with visual results (see Results section). IU of <2.9 on HIV-1 positive specimen were classified as recent infection, while IU≥2.9 were long term.
Arrows 1, 2 and 3 represent control line (CL), positive verification line (PVL) and long-term line (LTL), respectively. Presence of all three lines CL, PVL and LTL indicate HIV-positive diagnosis with long-term infection. The presence of CL and PVL indicate HIV-positive diagnosis with recent infection while the presence of only the control line (CL) indicates an HIV-seronegative sample. Quantitative results (in intensity units) can also be obtained with a strip reader.
Data analysis
The performance of the PVL was assessed by calculating the sensitivity and specificity of the PVL with reference to the standard algorithm: EIA + Western Blot, and Multispot or Geenius for serotyping. The performance of the LTL was calculated by correlating results from the LAg-Avidity EIA and Asante LTL for specimens with HIV-1 infections. Asante visual and Asante strip reader results were also compared.
Statistical analysis
This was performed using both Microsoft Excel (See S1 Data) and GraphPad Prism. The sensitivity and specificity of the assay for PVL were calculated using numerical values of true positives (TP), false positives (FP), true negatives (TN) and false negatives (FN). Agreement between the standard diagnostic algorithm and the ARRA were assessed using Kappa values where Kappa values were categorized as good (0.61 to 0.8), very good (0.81 to 0.99) and perfect (0.99 to 1.0). Two-by-two table and Kappa values were calculated for LTL comparing rapid test for recent infection (RTRI) classification of recent or LT to classification by LAg-Avidity EIA. The Spearman ranked correlation coefficient was calculated for the ODn of the LAg-Avidity EIA and the reader values of ARRA LTL. Percent agreement was calculated between Asante visual and Asante strip reader results, between visual and standard assays, and between strip reader and standard assays; all with two-sided 95% Wilson score confidence intervals.
Results
Sensitivity and specificity of the positive verification line (PVL)
Of the 580 HIV positive specimens in the panel, the ARRA correctly identified 575 as HIV positive with 5 false negatives, with visual interpretation giving a sensitivity of 99.1% (95% CI: 98.0%-99.6%) (Table 1A). The assay also correctly identified 910 of 920 negative specimens; 10 specimens were false-positive resulting in a specificity of 98.9% (98.0%-99.4%). The reader-based interpretation using a manufacturer recommended cutoff of 2.8 was similar to visual interpretation, yielding a sensitivity and specificity of 99.3% (576/580) and 98.8% (909/920), respectively (Table 1B). In both cases, kappa statistics comparing reference results with visual or reader-based interpretation were almost perfect (0.979) [41].
Sensitivity, specificity, % accuracy and kappa values are shown (with 95% confidence intervals).
Detection of recent or Long-term (LT) infections
The performance of the ARRA LTL was assessed by comparing the LTL visual results with the LAg-Avidity EIA at an ODn cutoff of 2.0, matching with the mean duration of recent infection (MDRI) of about 6 months [25]. The LAg-Avidity-EIA classified 98 of the qualified 566 HIV-1 specimens as recent and the ARRA classified 109 as recent (Table 2A); 4 ARRA false-negatives were excluded from the analysis. Eighty (81.6%) of 98 LAg recent cases were recent and 438 (93.8%) of 467 LAg LT cases were LT by the ARRA, yielding an overall agreement of 91.7% (518/565) and kappa of 0.722. For reader-based interpretation, a cutoff of 3.0 IU recommended by the manufacturer was initially also used for analysis, but the LAg-Avidity EIA better correlated to the Asante strip reader at a cutoff of 2.9 IU which also correlated best with visual results (see below, Fig 3). Using this cutoff for LTL, results matched well with LAg-EIA (Table 2B) and were similar to visual interpretation, with an overall agreement of 91.9% and kappa of 0.729. The agreement between visual and reader-based interpretation for classification of recent or LT infection is shown in Table 2C demonstrating high level of overall agreement (>98%) and almost perfect kappa (0.937).
Two-by- two tables comparing Asante Rapid Recency Assay results with LAg-Avidity EIA results for classification of recent or LT infections using (A) visual interpretation and B) strip reader @ cutoff of 2.9. C) Comparison of visual results with those generated with the strip reader for recency classification. Percent agreement and kappa values are shown for each comparison (with 95% CI).
We further examined quantitative reader values for HIV diagnosis (PVL) or recency status (LTL). Distribution of the Asante reader values for the PVL for HIV-positive and negative specimens is depicted in Fig 2A showing overall concordance for HIV diagnosis and reader values for a few false positives (N = 11) and false negatives (N = 4). Similarly, distribution of the Asante reader values for the LTL for HIV-1 recent and LT specimens, as classified by LAg-Avidity EIA, is shown in Fig 2B demonstrating concordance and switches among recent and LT infections when classified by ARRA compared to LAg EIA.
Horizontal lines represent threshold cutoff values of 2.8 (PVL) and 2.9 (LTL).
Comparison of visual reading with Asante strip reader
The percentage agreement between Asante visual and Asante strip reader in the interpretation of the PVL was 99.87% (95% CI: 99.52–99.98) with a kappa value of 0.997 (95% CI: 99.3–100). The percent agreement was 98.05% (95% CI: 96.54–99.02) for the LTL with a kappa value of 0.937 (0.901–0.974) (Table 2C). Specimens that had discrepant classification on the LTL were primarily located close to the cutoff of 2.9 and ranged from 2.7< to <3.2. Specimens in this zone were referred to as grey zone specimens and could occasionally switch between recent and long term. Fig 3 represents the relationship between visual interpretation and reader values with specimens arranged in ascending order of LT IU values. Visual results matched better with reader interpretation when cutoff of 2.9 was used for the LTL instead of cutoff of 3.0 as recommended by the manufacturer.
The results show strip reader values arranged in ascending order and the corresponding visual result. Shown is a transition zone between two vertical arrows where visual representation and strip reader results may not agree. As shown, all values are close to the cutoff and range between >2.7 IU to <3.2 IU. Horizontal line at 3.0 IU is a manufacturer-suggested cutoff, while line at 2.9 is the cutoff that matches well with visual interpretation.
Discussion
Access to HIV diagnostic testing has greatly expanded globally due to availability of point-of-care (POC) rapid diagnostic tests that are simple, inexpensive, and easy to perform. However, use of assays to detect recent HIV-1 infections has been limited mainly to cross-sectional surveys for estimation of incidence. Identification of recent infection is critical for epidemic control since evidence suggest that a high proportion of new infections originate from recently infected individuals [42–45]. However, identifying recent infections at POC has been difficult because of the complexities associated with current laboratory-based incidence assays that require collection of the specimens, transportation, processing, and storage before testing can be conducted at a centralized facility. Moreover, most HIV-1 incidence tests are technically complex, take several hours to perform, may require sophisticated equipment and specimen transport if testing will be done in the central laboratory [17, 46].
Our laboratory validation of the ARRA, a POC test that can simultaneously confirm HIV diagnosis and identify recent from long-term infections, shows great promise. This is the first evaluation of Asante Rapid Recency Assay to assess performance of both PV line and LT line. The assay showed great performance for HIV diagnosis with high sensitivity and specificity (99.1% and 98.9%, respectively; Table 1A) meeting and surpassing WHO-PQ/FDA requirements of minimum sensitivity of 99% or greater and specificity of 98% or greater [47]. In addition, ARRA simultaneously identified recent and LT infections in our cross-sectional specimen set with results similar to the LAg-Avidity EIA (>91% concordance and kappa >0.72) (Table 2A). The discordant rate of about 8% for detection of recent/LT cases between the two methodologies is not unexpected considering several cases are close to the cutoff (Fig 3B) and can switch between recent and LT status. This should have minimal impact on intended use of the assay for recent infection surveillance in programmatic settings and identifying risk factors of acquiring new infections for targeted prevention. Performance among subtype B specimens (n = 35) from the U.S. show 94.3% agreement with the LAg-EIA and non-B subtype specimens (n = 535) from other countries show 90.7% agreement with the LAg-EIA. However, only 35/570 (6.1%) HIV-1 specimens were from the U.S, making this difference is not statistically meaningful.
Combining HIV diagnostic testing with recent infection detection in a single test is important because often HIV false-positive results have been reported in the surveys and in the HIV program [48–50]. The testing of false-positive specimens with recent infection assays, without the diagnosis component, can result in misinterpretation since they will be misclassified as recent. Therefore, the laboratory-based LAg-Avidity EIA which only includes recency testing requires further confirmation of HIV diagnosis of individuals when ODn<0.4 to rule out false positives [40]. The ARRA that includes verification of HIV diagnosis serves to confirm HIV-positive status at the same time while classifying individuals as recent or LT.
Validation of antibody-based recency assays can be conducted by 1) use of longitudinal specimens (seroconversion panels) from recently infected persons to confirm expected antibody maturation profile and to determine the MDRI, 2) using specimens from persons with long-term infections (>1 year) to determine the proportion likely to be misclassified as false-recent and understand the reasons (e.g. elite controllers, ART) for misclassification for potential mitigation (e.g. viral load), and 3) comparison with a previously validated test (such as LAg-Avidity EIA) using a cross-sectional specimen set. We describe here the latter approach that allowed not only comparative evaluation of ARRA but also helped infer the MDRI without the use of seroconversion panels that are increasingly difficult to get.
The MDRI of the ARRA was inferred from correlation with the LAg-Avidity EIA data. An ODn cutoff of 2.0 correlated well with the long-term line on the ARRA (Table 2 and Fig 2), from which we inferred a tentative MDRI for the ARRA of approximately 6 months [25]. Additional work is underway to confirm the MDRI of the ARRA using a series of seroconversion panels. However, the MDRI is important for estimating HIV-1 incidence in a population, which is not the intended use of the assay at POC. ARRA is designed to identify recently infected individuals in routine HIV testing services, as part of surveillance of recent infections in HIV program for targeted prevention. Moreover, subtype differences will have minimal influence on the classification of recent and long-term infections which is interpreted as less than or greater than 1 year respectively, when interpreted at the individual level. There is a distribution around MDRI indicating individual may transition from recent to LT status earlier or later than 6 months, but almost all will transition within 1 year. Identifying individuals during this early phase can help interrupt ongoing transmission during their lifetime.
Until recently the utility of incidence testing was mostly limited to the estimation of HIV incidence at the population level, except for a few cases in the developed world where it has been used for case-based surveillance [38]. At the population level, cross-sectional national surveys have been conducted periodically (3–5 years) and incidence estimates from these surveys in sub-Saharan Africa have helped in the allocation of resources, defining prevention programs, and in assessing the impact of HIV programs [27, 35, 36]. However, these national surveys take significant amount of time (8–12 months for completion) and resources but detect only few recent infections [51]. Detection of only 10 to 50 recent infections out of 20,000 to 30,000 participants per survey limited our ability to use this information for further sub-group analysis or real-time intervention to prevent ongoing transmission in at risk groups. Estimation of new infections in the U.S. was mainly a retrospective activity and academic in nature [52, 53]. This did not allow timely public health response making a quick impact on ongoing transmission. This may partially explain why we have not seen a significant decline of new infections over the last 15 years in the U.S. Renewed focus on use of the algorithm to facilitate detection of acute infections, contact tracing, and targeted prevention should help prevent ongoing transmission and help reduce new infections [54].
The development of a rapid test for recent infection can quickly identify individuals who are recently infected, as this is the target group that is known to contribute significant transmission of HIV due to their immature immune response and high viral load [17, 24]. Some estimates suggest that 40%-60% of ongoing HIV transmission may be contributed by recently infected persons [44, 45, 55, 56]. Thus, early identification of recent infections, concomitant with HIV treatment and viral suppression, would interrupt HIV transmission at an early stage accelerating epidemic control. Contact tracing and partner testing will inherently increase the yield of HIV-positive cases as well as prevent further transmission through counseling and the provision of PrEP to those at risk. This shift from monitoring only cross-sectional populations to identifying new infections in the routine HIV testing services (HTS) and most importantly at the POC, will serve as an important epidemic control tool to reach zero new infections.
We used the Asante strip reader in conjunction with visual reading to evaluate its performance and establish reader cutoffs because it is useful in order to ensure kit lot consistency. The Asante strip reader is relatively expensive and may not be practical to use at the testing sites. Our results show that there is a high level of agreement between visual and strip reader results for both the PVL and LT lines (99.87% and 98.05% respectively). A few specimens changed classification on both reads, but these were limited to samples that were borderline which is expected. Based on the comparative results, a strip reader is not needed for the identification of recent infection for the intended use in public health surveillance as visual results are substantially similar to strip reader results. Because of this and other issues such as annual maintenance, calibration, fragility, and difficulty of using the reader at POC, we do not currently recommend the use of a strip reader at site level where rapid tests are routinely conducted.
As part of limitations, we do recognize that the ARRA is an antibody-based test similar to LAg-Avidity EIA and, therefore, will require use of the appropriate recent infection testing algorithm (RITA) to include viral load to correctly classify recent infection cases [27, 57]. If implemented in conjunction with case-based surveillance, it is possible that unique ID and clinical history can identify retesting individuals and individuals on ART, thus minimizing the need for VL. We also acknowledge the need for additional studies to determine and confirm MDRI using seroconversion panels and using specimens with varying duration of treatment, including breakthrough infections among those on pre-exposure prophylaxis (PrEP). Future studies using specimens with additional demographic and HIV-1 subtype information can be helpful to better understand performance and limitations of the assay. Additional limitation is HIV-2 positive specimens which, in the absence of type-specific diagnosis, will be classified as recent infections since rIDR-M antigen used for recency classification is derived from HIV-1 gp41 immunodominant region [20] and will not react with HIV-2 specimens. If HIV-2 is suspected, type-specific diagnosis should be conducted using peptide-based assays such as Multispot, Geenius or InnoLIA.
RTRI is now being implemented in several PEPFAR-supported countries as a key tool to accelerate epidemic control [58–60]. Although our study and additional data from other laboratories demonstrate high sensitivity and specificity of ARRA for HIV diagnosis, the test is not currently used as part of the national testing algorithm for HIV diagnosis because of lack of regulatory approval. Accordingly, the test is used as an addition to the national testing algorithm which increases the cost of testing and labor. Historically, HIV-1 incidence assays have been used mainly for incidence surveillance and did not require regulatory approval. However, future regulatory approval of a RTRI that combines HIV diagnosis and recency status in the same device will be beneficial to further simplify HIV diagnosis as well as recent infection surveillance.
Supporting information
S1 Data. Excel file with raw data that was used for all analysis shown in the paper.
https://doi.org/10.1371/journal.pgph.0000316.s001
(XLSX)
Acknowledgments
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the funding agencies.
References
- 1.
UNAIDS. 2020. http://aidsinfo.unaids.org/.
- 2. De Cock KM, J HW., J W. C. Reflections on 40 Years of AIDS. Emerging Infectious Diseases. 2021;27(6):1553–60. pmid:34013858
- 3. Oster AM, France AM, Panneer N, Banez Ocfemia MC, Campbell E, Dasgupta S, et al. Identifying Clusters of Recent and Rapid HIV Transmission Through Analysis of Molecular Surveillance Data. J Acquir Immune Defic Syndr. 2018;79(5):543–50. Epub 2018/09/18. pmid:30222659
- 4. Campbell EM, Jia H, Shankar A, Hanson D, Luo W, Masciotra S, et al. Detailed Transmission Network Analysis of a Large Opiate-Driven Outbreak of HIV Infection in the United States. J Infect Dis. 2017;216(9):1053–62. Epub 2017/10/14. pmid:29029156
- 5. Bicego GT, Nkambule R, Peterson I, Reed J, Donnell D, Ginindza H, et al. Recent patterns in population-based HIV prevalence in Swaziland. PLoS One. 2013;8(10):e77101. pmid:24143205
- 6. Justman J, Reed JB, Bicego G, Donnell D, Li K, Bock N, et al. Swaziland HIV Incidence Measurement Survey (SHIMS): a prospective national cohort study. Lancet HIV. 2017;4(2):e83–e92. Epub 2016/11/20. pmid:27863998
- 7. Kim AA, Parekh BS, Umuro M, Galgalo T, Bunnell R, Makokha E, et al. Identifying Risk Factors for Recent HIV Infection in Kenya Using a Recent Infection Testing Algorithm: Results from a Nationally Representative Population-Based Survey. PLoS One. 2016;11(5):e0155498. Epub 2016/05/20. pmid:27195800
- 8. Kimanga DO, Ogola S, Umuro M, Ng’ang’a A, Kimondo L, Murithi P, et al. Prevalence and incidence of HIV infection, trends, and risk factors among persons aged 15–64 years in Kenya: results from a nationally representative study. J Acquir Immune Defic Syndr. 2014;66 Suppl 1:S13–26. Epub 2014/01/22. pmid:24445338
- 9. Barin F, Meyer L, Lancar R, Deveau C, Gharib M, Laporte A, et al. Development and validation of an immunoassay for identification of recent human immunodeficiency virus type 1 infections and its use on dried serum spots. J Clin Microbiol. 2005;43(9):4441–7. pmid:16145089
- 10. Janssen RS, Satten GA, Stramer SL, Rawal BD, O’Brien TR, Weiblen BJ, et al. New testing strategy to detect early HIV-1 infection for use in incidence estimates and for clinical and prevention purposes. JAMA. 1998;280(1):42–8. Epub 1998/07/11. pmid:9660362.
- 11. Kothe D, Byers RH, Caudill SP, Satten GA, Janssen RS, Hannon WH, et al. Performance characteristics of a new less sensitive HIV-1 enzyme immunoassay for use in estimating HIV seroincidence. J Acquir Immune Defic Syndr. 2003;33(5):625–34. pmid:12902808.
- 12. Parekh BS, Pau CP, Kennedy MS, Dobbs TL, McDougal JS. Assessment of antibody assays for identifying and distinguishing recent from long-term HIV type 1 infection. AIDS Res Hum Retroviruses. 2001;17(2):137–46. Epub 2001/02/15. pmid:11177393.
- 13. Rawal BD, Degula A, Lebedeva L, Janssen RS, Hecht FM, Sheppard HW, et al. Development of a new less-sensitive enzyme immunoassay for detection of early HIV-1 infection. J Acquir Immune Defic Syndr. 2003;33(3):349–55. pmid:12843746.
- 14. Suligoi B, Galli C, Massi M, Di Sora F, Sciandra M, Pezzotti P, et al. Precision and accuracy of a procedure for detecting recent human immunodeficiency virus infections by calculating the antibody avidity index by an automated immunoassay-based method. J Clin Microbiol. 2002;40(11):4015–20. pmid:12409368
- 15. Andersson E, Shao W, Bontell I, Cham F, Cuong do D, Wondwossen A, et al. Evaluation of sequence ambiguities of the HIV-1 pol gene as a method to identify recent HIV-1 infection in transmitted drug resistance surveys. Infect Genet Evol. 2013;18:125–31. Epub 2013/04/16. pmid:23583545
- 16. Curtis KA, Kennedy MS, Charurat M, Nasidi A, Delaney K, Spira TJ, et al. Development and characterization of a bead-based, multiplex assay for estimation of recent HIV type 1 infection. AIDS Res Hum Retroviruses. 2012;28(2):188–97. Epub 2011/05/19. pmid:21585287.
- 17. Duong YT, Qiu M, De AK, Jackson K, Dobbs T, Kim AA, et al. Detection of recent HIV-1 infection using a new limiting-antigen avidity assay: potential for HIV-1 incidence estimates and avidity maturation studies. PLoS One. 2012;7(3):e33328. Epub 2012/04/06. pmid:22479384
- 18. Granade TC, Nguyen S, Kuehl DS, Parekh BS. Development of a novel rapid HIV test for simultaneous detection of recent or long-term HIV type 1 infection using a single testing device. AIDS Res Hum Retroviruses. 2013;29(1):61–7. Epub 2013/01/04. pmid:23281586.
- 19. Hu DJ, Vanichseni S, Mock PA, Young NL, Dobbs T, Byers RH Jr., et al. HIV type 1 incidence estimates by detection of recent infection from a cross-sectional sampling of injection drug users in Bangkok: use of the IgG capture BED enzyme immunoassay. AIDS Res Hum Retroviruses. 2003;19(9):727–30. Epub 2003/10/31. pmid:14585202.
- 20. Wei X, Liu X, Dobbs T, Kuehl D, Nkengasong JN, Hu DJ, et al. Development of two avidity-based assays to detect recent HIV type 1 seroconversion using a multisubtype gp41 recombinant protein. AIDS Res Hum Retroviruses. 2010;26(1):61–71. Epub 2010/01/13. pmid:20063992.
- 21. Rikhtegaran Tehrani Z, Azadmanesh K, Mostafavi E, Gharibzadeh S, Soori S, Azizi M, et al. High avidity anti-integrase antibodies discriminate recent and non-recent HIV infection: Implications for HIV incidence assay. Journal of Virological Methods. 2018;253:5–10. https://doi.org/10.1016/j.jviromet.2017.12.003 pmid:29248529
- 22. Seaton KE, Vandergrift NA, Deal AW, Rountree W, Bainbridge J, Grebe E, et al. Computational analysis of antibody dynamics identifies recent HIV-1 infection. JCI Insight. 2017;2(24). Epub 2017/12/22. pmid:29263306
- 23. Yufenyuy EL, Parekh B. Development of a Multiplex Assay for Concurrent Diagnoses and Detection of Hiv-1, Hiv-2, and Recent Hiv-1 Infection in a Single Test. AIDS Res Hum Retroviruses. 2018. Epub 2018/07/31. pmid:30056751.
- 24. Parekh BS, Kennedy MS, Dobbs T, Pau CP, Byers R, Green T, et al. Quantitative detection of increasing HIV type 1 antibodies after seroconversion: a simple assay for detecting recent HIV infection and estimating incidence. AIDS Res Hum Retroviruses. 2002;18(4):295–307. Epub 2002/02/28. pmid:11860677.
- 25. Duong YT, Kassanjee R, Welte A, Morgan M, De A, Dobbs T, et al. Recalibration of the limiting antigen avidity EIA to determine mean duration of recent infection in divergent HIV-1 subtypes. PLoS One. 2015;10(2):e0114947. Epub 2015/02/25. pmid:25710171
- 26. Hargrove JW, van Schalkwyk C, Humphrey JH, Mutasa K, Ntozini R, Owen SM, et al. Short Communication: Heightened HIV Antibody Responses in Postpartum Women as Exemplified by Recent Infection Assays: Implications for Incidence Estimates. AIDS Res Hum Retroviruses. 2017;33(9):902–4. Epub 2017/04/27. pmid:28443672
- 27. Duong YT, Dobbs T, Mavengere Y, Manjengwa J, Rottinghaus E, Saito S, et al. Field Validation of Limiting-Antigen Avidity Enzyme Immunoassay to Estimate HIV-1 Incidence in Cross-Sectional Survey in Swaziland. AIDS Res Hum Retroviruses. 2019;35(10):896–905. Epub 2019/06/18. pmid:31204867.
- 28. Gonese E, Kilmarx PH, van Schalkwyk C, Grebe E, Mutasa K, Ntozini R, et al. Evaluation of the Performance of Three Biomarker Assays for Recent HIV Infection Using a Well-Characterized HIV-1 Subtype C Incidence Cohort. AIDS Res Hum Retroviruses. 2019;35(7):615–27. Epub 2019/04/03. pmid:30938164.
- 29. Vermeulen M, Chowdhury D, Swanevelder R, Grebe E, Brambilla D, Jentsch U, et al. HIV incidence in South African blood donors from 2012 to 2016: a comparison of estimation methods. Vox Sang. 2020. Epub 2020/08/08. pmid:32762088.
- 30. Karatzas-Delgado EF, Ruiz-Gonzalez V, Garcia-Cisneros S, Olamendi-Portugal ML, Herrera-Ortiz A, Lopez-Gatell H, et al. Evaluation of an HIV recent infection testing algorithm with serological assays among men who have sex with men in Mexico. J Infect Public Health. 2020;13(4):509–13. Epub 2019/12/10. pmid:31813835.
- 31. Kim AA, Rehle T. Short Communication: Assessing Estimates of HIV Incidence with a Recent Infection Testing Algorithm That Includes Viral Load Testing and Exposure to Antiretroviral Therapy. AIDS Res Hum Retroviruses. 2018;34(10):863–6. Epub 2018/06/22. pmid:29926735.
- 32. Shah NS, Duong YT, Le LV, Tuan NA, Parekh BS, Ha HTT, et al. Estimating False-Recent Classification for the Limiting-Antigen Avidity EIA and BED-Capture Enzyme Immunoassay in Vietnam: Implications for HIV-1 Incidence Estimates. AIDS Res Hum Retroviruses. 2017;33(6):546–54. Epub 2017/02/15. pmid:28193090
- 33. Laeyendecker O, Gray RH, Grabowski MK, Reynolds SJ, Ndyanabo A, Ssekasanvu J, et al. Validation of the Limiting Antigen Avidity Assay to Estimate Level and Trends in HIV Incidence in an A/D Epidemic in Rakai, Uganda. AIDS Res Hum Retroviruses. 2019;35(4):364–7. Epub 2018/12/19. pmid:30560723
- 34. Huik K, Soodla P, Pauskar M, Owen SM, Luo W, Murphy G, et al. The concordance of the limiting antigen and the Bio-Rad avidity assays in persons from Estonia infected mainly with HIV-1 CRF06_cpx. PLoS One. 2019;14(5):e0217048. Epub 2019/05/28. pmid:31125356
- 35. Gonese E, Musuka G, Ruangtragool L, Hakim A, Parekh B, Dobbs T, et al. Comparison of HIV Incidence in the Zimbabwe Population-Based HIV Impact Assessment Survey (2015–2016) with Modeled Estimates: Progress Toward Epidemic Control. AIDS Res Hum Retroviruses. 2020;36(8):656–62. Epub 2020/06/06. pmid:32498542.
- 36.
ICAP. PHIA Project: Key data to guide the global response to the HIV epidemic: ICAP, Columnia University; 2021 [cited 2021 01/04/2021]. https://phia.icap.columbia.edu/.
- 37. Le Vu S, Velter A, Meyer L, Peytavin G, Guinard J, Pillonel J, et al. Biomarker-based HIV incidence in a community sample of men who have sex with men in Paris, France. PLoS One. 2012;7(6):e39872. pmid:22768150
- 38. Aghaizu A, Murphy G, Tosswill J, DeAngelis D, Charlett A, Gill ON, et al. Recent infection testing algorithm (RITA) applied to new HIV diagnoses in England, Wales and Northern Ireland, 2009 to 2011. Euro Surveill. 2014;19(2). Epub 2014/01/25. pmid:24457006.
- 39. Parekh BS, Kalou MB, Alemnji G, Ou CY, Gershy-Damet GM, Nkengasong JN. Scaling up HIV rapid testing in developing countries: comprehensive approach for implementing quality assurance. Am J Clin Pathol. 2010;134(4):573–84. Epub 2010/09/22. pmid:20855638.
- 40.
Sedia Biosciences Corporation. Asante™ HIV-1 Rapid Recency™ Assay. 2018 [cited 2021 June 7]. Kit insert]. http://www.sediabio.com/products/lag-avidity-eia.
- 41. McHugh ML. Interrater reliability: the kappa statistic. Biochemia Medica. 2012;22(3):276–82. pmid:23092060
- 42. Brenner BG, Roger M, Routy JP, Moisi D, Ntemgwa M, Matte C, et al. High rates of forward transmission events after acute/early HIV-1 infection. J Infect Dis. 2007;195(7):951–9. pmid:17330784.
- 43. Miller WC, Rosenberg NE, Rutstein SE, Powers KA. Role of acute and early HIV infection in the sexual transmission of HIV. Curr Opin HIV AIDS. 2010;5(4):277–82. pmid:20543601
- 44. Chibo D, Kaye M, Birch C. HIV transmissions during seroconversion contribute significantly to new infections in men who have sex with men in Australia. AIDS Res Hum Retroviruses. 2012;28(5):460–4. pmid:21806486.
- 45. Vasylyeva TI, Friedman SR, Smyrnov P, Bondarenko K. A new approach to prevent HIV transmission: Project Protect intervention for recently infected individuals. AIDS Care. 2015;27(2):223–8. pmid:25244688
- 46. Yufenyuy EL, Parekh BS. Development of a Multiplex Assay for Concurrent Diagnoses and Detection of HIV-1, HIV-2, and Recent HIV-1 Infection in a Single Test. AIDS Res Hum Retroviruses. 2018;34(12):1017–27. Epub 2018/07/31. pmid:30056751.
- 47.
WHO/CDC. Improving the Quality of HIV-Related Point-of-Care Testing: Ensuring the Reliability and Accuracy of Test Results. 2015.
- 48. Augusto ADR, Iriemenam NC, Kohatsu L, de Sousa L, Maueia C, Hara C, et al. High level of HIV false positives using EIA-based algorithm in survey: Importance of confirmatory testing. PLoS One. 2020;15(10):e0239782. Epub 2020/10/23. pmid:33091019
- 49. Kosack CS, Shanks L, Beelaert G, Benson T, Savane A, Ng’ang’a A, et al. HIV misdiagnosis in sub-Saharan Africa: performance of diagnostic algorithms at six testing sites. J Int AIDS Soc. 2017;20(1):21419. Epub 2017/07/12. pmid:28691437
- 50. Shanks L, Siddiqui MR, Abebe A, Piriou E, Pearce N, Ariti C, et al. Dilution testing using rapid diagnostic tests in a HIV diagnostic algorithm: a novel alternative for confirmation testing in resource limited settings. Virol J. 2015;12:75. Epub 2015/05/15. pmid:25972188
- 51. Voetsch AC, Duong YT, Stupp P, Saito S, McCracken S, Dobbs T, et al. HIV-1 Recent Infection Testing Algorithm With Antiretroviral Drug Detection to Improve Accuracy of Incidence Estimates. J Acquir Immune Defic Syndr. 2021;87(Suppl 1):S73–S80. Epub 2021/06/25. pmid:34166315.
- 52. Bosh KA, Hall HI, Eastham L, Daskalakis DC, Mermin JH. Estimated Annual Number of HIV Infections horizontal line United States, 1981–2019. MMWR Morb Mortal Wkly Rep. 2021;70(22):801–6. Epub 2021/06/04. pmid:34081686 Journal Editors form for disclosure of potential conflicts of interest. No potential conflicts of interest were disclosed.
- 53. Song R, Hall HI, Green TA, Szwarcwald CL, Pantazis N. Using CD4 Data to Estimate HIV Incidence, Prevalence, and Percent of Undiagnosed Infections in the United States. J Acquir Immune Defic Syndr. 2017;74(1):3–9. Epub 2016/08/11. pmid:27509244.
- 54. Selik RM, Mokotoff ED, Branson B, Owen SM, Whitmore S, Hall HI. Revised Surveillance Case Definition for HIV Infection—United States, 2014. Morb Mortal Wkly Rep. 2014;63(3):1–10.
- 55. Hollingsworth TD, Anderson RM, Fraser C. HIV-1 transmission, by stage of infection. J Infect Dis. 2008;198(5):687–93. pmid:18662132.
- 56. Marks G, Crepaz N, Senterfitt JW, Janssen RS. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: implications for HIV prevention programs. J Acquir Immune Defic Syndr. 2005;39(4):446–53. pmid:16010168.
- 57.
UNAIDS/WHO. WHO Working Group on HIV Incidence Assays Meeting Report: Estimating HIV Incidence Using Case Surveillance. https://www.who.int/hiv/pub/meetingreports/estimating-hiv-incidence-using-case-surveillance/en/2017.
- 58. Kim AA, Behel S, Northbrook S, Parekh BS. Tracking with recency assays to control the epidemic: real-time HIV surveillance and public health response. AIDS. 2019;33(9):1527–9. Epub 2019/04/26. pmid:31021850
- 59.
PEPFAR Solutions. Surveillance of Recent HIV Infections: Using Point-of-Care Recency Tests to Rapidly Detect and Respond to Recent Infections 2018. https://www.pepfarsolutions.org/emerging-technologies-innovations/2018/7/11/surveillance-of-recent-hiv-infections-using-point-of-care-recency-tests-to-rapidly-detect-and-respond-to-recent-infections.
- 60.
UCSF & ICAP. TRACE: Tracking with Recency Assays to Control the Epidemic 2021 [cited 2021 June 20]. https://trace-recency.org/.