Review and References: Incidence Inference Using Biomarkers for ‘Recent infection’

Some of the key developments in the area of incidence estimation using cross-sectional surveys testing for biomarkers of ‘recent infection’ are listed below.

Review

In 1995, Brookmeyer and Quinn used the prevalence of p24 antigenemia, in the absence of HIV-antibodies, to identify individuals with recent infection. Incidence of HIV was then estimated by the ratio of the ‘prevalence’ of recent infection (defined to be the ratio of ‘recently’ infected to uninfected individuals) to the mean duration of this recent infection.

Since this initial work, there has been much research and development, in both (i) the methodology for estimating incidence, given categorisations of the population at a point in time into the states ‘recently infected’, ‘non-recently infected’ and ‘uninfected’, as well as (ii) the incidence assays used to distinguish ‘recent’ from ‘non-recent’ infection. There are a number of reviews of these developments. For example, see: Le Vu et al, 2008; Murphy and Parry, 2008; Busch et al, 2010; Welte et al, 2010; and Incidence Assay Critical Path Working Group 2011.

Two approaches for identifying recent infection have been widely used in the field, namely those developed by Janssen et al, 1998, and Parekh et al, 2002. Janssen et al introduced the Serological Testing Algorithm for Recent HIV Seroconversion (STARHS), in which less-sensitive or ‘detuned’ versions of existing tests, that measure antibody titre, are used; while Parekh et al developed the BED IgG-Capture Enzyme Immunoassay (referred to as the BED assay), which measures the proportion of HIV-specific IgG in total IgG (see Bärnighausen et al, 2010 for a review of studies using the BED assay). Measurements below chosen thresholds indicate recent infection.

However, the poor performance of currently used recency tests has posed an obstacle to obtaining reliable incidence estimates. For example, test performance has been found to be population-specific (for example, subtype-specific and affected by treatment), and some subjects with long-standing infection are classified as recently infected. The development of improved incidence assays has been an active area of research in recent years, with a number of new incidence assays being developed, as well as a focus on combining multiple tests into recent infection testing algorithms. For example, genome-based tests (with classifications based on viral diversity) and avidity tests (measuring the strength of binding between antibodies and antigens) have been developed. For example, see: the above-listed reviews; Braunstein et al, 2011; Curtis et al, 2011; Park et al, 2011; Ragonnet-Cronin et al, 2012.

Various incidence estimators have been proposed, incorporating parameters, in addition to the mean duration of recent infection appearing in the original one-parameter estimator, to accommodate the problem of ‘false-recent’ test results:

  • McDougal et al, 2006, formulated an incidence estimator which made use of a ‘sensitivity’, ‘short-term specificity’ and ‘long-term specificity’.
  • Hargrove et al, 2008, proposed a simplification of the formula proposed by McDougal et al.
  • McWalter and Welte, 2010 (epub 2009), used mathematical modelling to show that one can obtain a weighted average of recent incidence, and that two parameters naturally emerge out of the analysis, namely a false-recent rate and mean duration of recent infection.
  • Wang and Lagakos, 2009, showed, with the use of additional assumptions, that the estimator of McWalter and Welte is indeed the maximum likelihood estimator.

These estimators are compared in McWalter and Welte, 2009. However, the derivations of the estimators to account for ‘false-recent’ results led to some debate. For example, see: Brookmeyer 2009; and related correspondence by Hargrove; Welte et al, McDougal and Brookmeyer. Furthermore, derivations of the estimators relied on specific assumptions, of epidemiological and demographic equilibrium, about post-infection survival, and  about the dynamics of the test for recent infection. These assumptions are known to be substantially violated, and this led to lack of consensus in the field. SACEMA has recently developed a very general formal methodology that relaxes the assumptions of previous estimators:

  • Kassanjee R, McWalter TA, Bärnighausen T, Welte A. A new general biomarker-based incidence estimator. Epidemiology. 2012; 23(5): 721–728.

The estimator formalises robust definitions of the ‘mean duration of recent infection’ and the ‘false-recent rate’ (defined below), and formally accounts for arbitrary dynamics of the test for recent infection. Any biases arising from deviations from epidemiological or demographic equilibrium, and imperfect post-infection survival, over the relevant time scale, are also formally quantified.

It is increasingly agreed that using two parameters to describe the performance of the test for recent infection is appropriate for incidence estimation, namely a (i) ‘false-recent rate’, describing the propensity for individuals with long-standing infection to return recent results, which needs to be suitably low (for example, not more than 2%), and (ii) ‘mean duration of recent infection’, which needs to be sufficiently large (for example, of order of half a year at least). For example, see: Incidence Assay Critical Path Working Group, 2011. The accurate and precise estimation of these parameters, which are now formally and generally defined in Kassanjee et al 2012, is essential to obtain reliable incidence estimates, and therefore characterisation of incidence assays has in parallel also been an area of active research. For example, see: Claggett et al, 2011; Kassanjee et al, 2011; Parekh et al, 2011. The CEPHIA project, initiated in 2011, aims to develop a repository of specimens specifically for the characterisation of existing and new incidence assays.

References

  • Brookmeyer R, Quinn TC. Estimation of current human immunodeficiency virus incidence rates from a cross-sectional survey using early diagnostic tests. Am J Epidemiol 1995; 141(2):166-72.
  • Le Vu S, Pillonel J, Semaille C, Bernillon P, Le Strat Y, Meyer L, Desenclos JC. Principles and uses of HIV incidence estimation from recent infection testing--a review. Euro Surveill 2008; 13(36):11-16.
  • Murphy G, Parry JV. Assays for the detection of recent infections with human immunodeficiency virus type 1. Euro Surveill 2008; 13(36):4-10.
  • Busch MP, Pilcher CD, Mastro TD, Kaldor J, Vercauteren G, Rodriguez W, Rousseau C, Rehle TM, Welte A, Averill MD, Garcia Calleja JM. Beyond detuning: 10 years of progress and new challenges in the development and application of assays for HIV incidence estimation. AIDS 2010; 24(18):2763-2771.
  • Welte A, McWalter TA, Laeyendecker O, Hallett TB. Using tests for recent infection to estimate incidence: problems and prospects for HIV. Euro Surveill 2010; 15(24):pii=19589.
  • Incidence Assay Critical Path Working Group. More and better information to tackle HIV epidemics: Towards improved HIV incidence assays. PLoS Med 2011; 8(6): e1001045.
  • Janssen RS, Satten GA, Stramer SL, Rawal BD, O'Brien TR, Weiblen BJ, Hecht FM, Jack N, Cleghorn FR, Kahn JO, Chesney MA, Busch MP. 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
  • Parekh BS, Kennedy MS, Dobbs T, Pau CP, Byers R, Green T, Hu DJ, Vanichseni S, Young NL, Choopanya K, Mastro TD, McDougal JS. 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.
  • Bärnighausen T, McWalter TA, Rosner Z, Newell ML, Welte A. HIV incidence estimation using the BED capture enzyme immunoassay: systematic review and sensitivity analysis. Epidemiology 2010; 21(5):685-97.
  • Braunstein SL, Nash D, Kim AA, Ford K, Mwambarangwe L, Ingabire CM, Vyankandondera J, van de Wijgert JH. Dual testing algorithm of BED-CEIA and AxSYM avidity index assays performs best in identifying recent HIV infection in a sample of Rwandan sex workers. PLoS One 2011; 6(4):e18402.
  • Curtis K, Kennedy S, Charurat ME, Nasidi A, Delaney K, Spira T, Owen M. Development and characterization of a bead-based, multiplex assay for estimation of recent HIV-1 infection. AIDS Res Hum Retroviruses 2012 (epub 2011); 28(2):188-197.
  • Park SY, Love TM, Nelson J, Thurston SW, Perelson AS, Lee HY. Designing a genome-based HIV incidence assay with high sensitivity and specificity. AIDS 2011, 25(16):F13-9.
  • McDougal JS, Parekh BS, Peterson ML, Branson BM, Dobbs T, Ackers M, Gurwith M. Comparison of HIV type 1 incidence observed during longitudinal follow-up with incidence estimated by cross-sectional analysis using the BED capture enzyme immunoassay. AIDS Res Hum Retroviruses 2006; 22(10):945-52.
  • Hargrove JW, Humphrey JH, Mutasa K, Parekh BS, McDougal JS, Ntozini R, Chidawanyika H, Moulton LH, Ward B, Nathoo K, Iliff PJ, Kopp E. Improved HIV-1 incidence estimates using the BED capture enzyme immunoassay. AIDS 2008; 22(4):511-8.
  • McWalter TA, Welte A. Relating recent infection prevalence to incidence with a sub-population of assay non-progressors. J Math Biol 2010 (epub 2009); 60(5):687-710.
  • Wang R, Lagakos SW. On the use of adjusted cross-sectional estimators of HIV incidence. J Acquir Immune Defic Syndr 2009; 52(5):538-47.
  • McWalter TA, Welte A. A comparison of biomarker based incidence estimators. PLoS One 2009; 4(10):e7368.
  • Brookmeyer R. Should biomarker estimates of HIV incidence be adjusted? AIDS 2009; 23(4):485-91
  • Hargrove JW. BED estimates of HIV incidence must be adjusted. AIDS 2009; 23(15):2061-2.
  • Welte A, McWalter TA, Bärnighausen T. Reply to 'Should biomarker estimates of HIV incidence be adjusted?' AIDS 2009; 23(15):2062-3.
  • McDougal JS. BED estimates of HIV incidence must be adjusted. AIDS 2009; 23(15):2064-5
  • Brookmeyer R. Response to correspondence on 'Should Biomarker Estimates of HIV Incidence be Adjusted?' AIDS 2009; 23(5):2066-8.
  • Claggett B, Lagakos SW, Wang R. Augmented cross-sectional studies with abbreviated follow-up for estimating HIV incidence. Biometrics 2012 (epub 2011); 68(1):62-74.
  • Kassanjee R, Welte A, McWalter TA, Keating SM, Vermeulen M, Stramer SL, Busch MP. Seroconverting blood donors as a resource for characterising and optimising recent infection testing algorithms for incidence estimation. PLoS One 2011; 6(6):e20027.
  • Parekh BS, Hanson DL, Hargrove J, Branson B, Green T, Dobbs T, Constantine N, Overbaugh J, McDougal JS. Determination of mean recency period for estimation of HIV type 1 Incidence with the BED-capture EIA in persons infected with diverse subtypes. AIDS Res Hum Retroviruses 2011; 27(3):265-73.
  • Kassanjee R, McWalter TA, Bärnighausen T, Welte A. A new general biomarker-based incidence estimator. Epidemiology 2012; 23(5): 721–28.
  • Ragonnet-Cronin M, Aris-Brosou S, Joanisse I, Merks H, Vallée D, Caminiti K, Rekart M, Krajden M, Cook D, Kim J, Malloch L, Sandstrom P, Brooks J. Genetic diversity as a marker for timing infection in HIV-infected patients: evaluation of a 6-month window and comparison with BED. J Infect Dis 2012; 206(5):756-64.