Ultrasensitive quantitative HIV-1 p24 antigen assay adapted to dried plasma spots to improve treatment monitoring in low-resource settings

Results A specificity of 100% was observed when p24 antigen was measured using DPS, and no differences of p24 concentration could be seen between DPS and venous plasma. The correlation between DPS and venous plasma p24 was excellent ( R = 0.93, CI 95% = 0.88–0.96, p < 0.0001). Similarly, p24 antigen concentrations using DPS were well correlated with RNA viral load ( R = 0.53, CI 95% = 0.27–0.72, p = 0.0002). Conclusions This quantitative p24 antigen test has similar sensitivity and specificity using DPS and venous plasma, and has the potential to improve health care delivery to HIV-affected individuals in resource-constrained countries. Keywords HIV-1 Treatment monitoring Viral load p24 antigen Dried plasma spots Low-resource setting 1 Introduction Today's standard of care for the management of HIV infection includes the quantitative assessment of HIV-1 viremia in order to monitor the response to antiretroviral therapy (ART) or to detect viral relapses. This remains a challenge in resource-constrained countries, mainly because a quantitative assay for HIV-1 remains inaccessible for most patients. Quantitative tests such as reverse transcriptase polymerase chain reaction (RT-PCR), nucleic acid sequence-based amplification (NASBA), or branched chain DNA assay (bDNA), are technically demanding and expensive and, thus, rarely available at the district hospital level. Consequently, ART initiation and treatment monitoring are mostly based on immunological or clinical markers, such as CD4 counts or diagnosis of opportunistic infections. Introduction of new tests often requires adjustments in the local setting. Inexpensive and easy-to-perform tests are required and simplified sampling and transport procedures may also be necessary. In this respect, the use of dried plasma spots (DPS) on filter papers is a convenient and suitable technique that can be used even in remote district hospitals and rural settings ( Behets et al., 1992; Sherman et al., 2005; Stevens et al., 1992 ). Our group has previously developed an ultrasensitive p24 antigen assay that is inexpensive, easy-to-perform, and based on use of venous plasma ( Schupbach, 2003 ). This test does not require large and expensive equipment, which makes it suitable for smaller laboratories. Many clinical studies have been done with this assay using samples from patients with subtype B infections ( Mascolini, 2002; Respess et al., 2003, 2005; Sterling et al., 2002 ). The assay has also been evaluated on non-B subtypes ( Burgisser et al., 2000; Ribas et al., 2003 ) and has been introduced successfully in different African and Asian countries ( de Baets et al., 2005; Sutthent et al., 2003; Tehe et al., submitted for publication ). Recently, de Baets et al. (2005) described the use of DPS for diagnosis of pediatric infection in Central Africa and reported a 100% specificity and sensitivity for the p24 test. Our goal was to develop a quantitative test based on DPS to monitor HIV patients including children under ART. We are herewith reporting the adaptation of the p24 antigen ELISA to DPS in an effort to develop a simple, low-tech, quality-assured method that allows for standard-of-care management of HIV-1 infected patients in remote areas with no direct access to laboratories. 2 Methods Forty-seven EDTA-anticoagulated venous blood samples were obtained from 42 HIV-1-seropositive, treated or untreated adult individuals. Aliquots of 80 μl plasma were applied to specimen collection papers (Single-part Generic Card, S&S #903, Schleicher & Schuell, Basel, Switzerland) and left to dry overnight at room temperature. These filters were stored until use at 4 °C or −20 °C in small plastic bags containing a desiccant pack (Schleicher & Schuell). Round disks of 16 mm of diameter were cut out, transferred into the wells of 24-well tissue culture plates (NUNC, Wiesbaden, Germany), sealed with a plastic cover sheet, and eluted for 8 h at room temperature into 0.8 ml of an elution buffer containing 1 part H 2 O, 1 part SNCR virus lysis buffer ( Schupbach et al., 2003 ), and 9 parts 0.5% Triton X-100. The eluate was denatured for 5 min on a dry-heat block preheated to 100 °C (Techne, Cambridge, UK) and 250 μl/well were then applied to the Perkin-Elmer HIV-1 Core Profile ELISA plate and processed as previously described ( Schupbach, 2003; Schupbach et al., 2003 ) with three modifications to maximize the reaction: (i) the eluated DPS samples were incubated overnight at 37 °C on the ELISA plate, (ii) the detector antibody was incubated for 75 min at 37 °C, and (iii) the horseradish peroxidase was diluted 1:25 in its diluent (for details of procedure see also http://www.nzr.unizh.ch/hivproteinload/index.html ). For a few experiments, a different substrate, TMB SureBlue Reserve (KPL, Gaithersburg) was used to improve sensitivity. The p24 concentrations were determined retrospectively and in batches, whereas the viral load and CD4 counts were determined prospectively by routine procedures ( Friedl et al., 2001 ). 3 Results Thirty HIV-negative individuals as well as 42 known HIV-infected patients were recruited for this evaluation. These patients were infected with HIV-1 subtype B, with the exception of one HIV-1 subtype C and one HIV-1 subtype CRF01_AE infection. Viral loads and CD4 counts were assessed prospectively and are shown in Table 1 , together with the characteristics of the individuals. None of the HIV-negative individuals yielded a result above the cut-off of the assay, revealing a specificity of 100% for the dried plasma spots (DPS) p24 assay. For the HIV-1 infected patients, concentrations of p24 in DPS were slightly higher than in plasma (Wilcoxon rank test, p = 0.027, N = 45). An excellent correlation was obtained for the comparison of DPS p24 and plasma p24 ( R = 0.93, CI 95% = 0.88–0.96, p < 0.0001; Fig. 1 A ). We further compared the DPS p24 results with those obtained for HIV-1 RNA ( Fig. 1 B). A significant positive correlation was seen ( R = 0.53, CI 95% = 0.27–0.72, p = 0.0002), and similar results were obtained when plasma p24 was correlated with HIV-1 RNA ( R = 0.56, CI 95% = 0.32–0.74, p < 0.0001, data not shown). The sensitivity of the p24 test with plasma or DPS was similar with 42/47 (92%) and 41/47 (90%), respectively. The few samples that were undetectable using plasma or DPS were re-tested using TMB SureBlue Reserve (KPL) as substrate, and were then mostly measurable (45/47), conveying a sensitivity of 96% for plasma as well as for DPS p24 antigen. The same two patients were undetectable in both p24 procedures ( Fig. 1 B, open diamonds), notably, one of them had a sustained viral suppression with occasional blips and an actual viral load of 382 copies/ml. The other patient started therapy during primary infection and experienced a very rapid decrease of HIV RNA load. 4 Discussion Our previously developed p24 assay based on the use of heat-denatured venous plasma was adapted and validated for quantitative assessment of p24 antigen from DPS. Importantly, DPS are very stable. We submitted DPS for 10 days to daily temperature cycling of 37 °C and 25 °C, respectively, and could not detect any loss in p24 antigen (data not shown). This corroborates the observation of Li et al. (2005) who stored DPS at 0 °C, 24 °C, or 37 °C for 2 weeks without any loss of p24 antigen. Thus, DPS are easy to store and transport, which renders standard of care management of HIV-infected patients more accessible in resource-constrained countries. We found the quantitative DPS procedure to be specific and to yield p24 concentrations that were equal to those derived from venous plasma. The linear regression equation that relates the p24 concentrations established from DPS to those from venous plasma, y = 1.0 x + 0.04, indicates that the two procedures give equivalent results. There was also a fair correlation with HIV-1 RNA concentrations ( r = 0.67), which confirms recent results, such as in the study of Li et al. (2005) who observed a Spearman rank correlation of r = 0.74. However, only 90% of the HIV-1 RNA positive samples were found to contain measurable p24 when using the OPD substrate of the p24 antigen ELISA kit (Perkin-Elmer), whereas the sensitivity reached 96% when TMB SureBlue Reserve was used as substrate. By comparison, Li et al. (2005) observed a sensitivity of 80%, which might reflect a less effective elution of p24 antigen from the spots. The fact that two patients were negative for p24 and only a modest correlation was observed with HIV-1 RNA does not necessarily imply a lower suitability of p24 for treatment monitoring. Earlier studies have demonstrated similar predictive power for both p24 and HIV-1 RNA regarding CD4 T-cell decline, clinical progression and survival ( Ledergerber et al., 2000; Sterling et al., 2002 ). Several studies noted a higher correlation between CD4 counts and p24 than between CD4 counts and HIV-1 RNA ( Schupbach et al., 2003, 2005; Stevens et al., 2005; Tehe et al., submitted for publication ). It is of note that one of these studies found that the CD4 slopes observed over time in treated patients with stably suppressed HIV-1 RNA were correlated inversely with the concomitant slopes of p24 ( Schupbach et al., 2003 ), and another observed that the CD4 decline observed after treatment cessation in a structured treatment interruption study was correlated significantly with the p24 increase, but not with the much more impressive increase in HIV-1 RNA ( Schupbach et al., 2005 ). There are also situations in which the sensitivity of p24 exceeds that of HIV-1 RNA. A recent study demonstrated that one-third of treated patients, who had maintained HIV-1 RNA concentrations below 50 copies/ml for a median duration of 24 months, still contained low, but detectable concentrations of p24, which were furthermore predictive of the p24 rebound after treatment cessation ( Schupbach et al., 2005 .). This surprising result is compatible with the fact that most of the p24 present in plasma of chronically infected patients is located outside of the virus particles; disappearance of the particles (and HIV-1 RNA) under ART is apparently not always paralleled by a complete disappearance of extraviral p24, thus suggesting ongoing virus expression in these patients. The encouraging results obtained by determining p24 antigen in DPS in a subtype B setting should now be confirmed in a larger study that would include patients infected with different HIV subtypes. This might require additional improvements of the test, such as introducing a larger panel of p24-recognizing antibodies or further increasing its sensitivity. In summary, this study demonstrates that p24 antigen ELISA can be used quantitatively with DPS with equal sensitivity and specificity to venous plasma and may represent a useful tool for treatment monitoring in resource-constrained countries. Notably, this simplification of the pre-analytical steps may not only facilitate centralized testing, but also enable inter-continental collaborations or quality control programs. Standard-of-care in the management of HIV-1 infection can thereby be achieved more easily. Acknowledgments We are grateful to Christina Grube for excellent patient care and blood collection, to Lucia Bertodatto, Antonietta Baumgartner, and Cyril Shah for preparing the DPS and plasma samples. We are also indebted to the patients for their participation. This work was financed by a grant from the Propter Homines Foundation, Vaduz, Principality of Liechtenstein. References Behets et al., 1992 F. Behets M. Kashamuka M. Pappaioanou T.A. Green R.W. Ryder V. Batter Stability of human immunodeficiency virus type 1 antibodies in whole blood dried on filter paper and stored under various tropical conditions in Kinshasa Zaire J Clin Microbiol 30 1992 1179 1182 Burgisser et al., 2000 P. Burgisser P. Vernazza M. Flepp J. Boni Z. Tomasik U. Hummel Performance of five different assays for the quantification of viral load in persons infected with various subtypes of HIV-1. Swiss HIV Cohort Study J Acquir Immune Defic Syndr 23 2000 138 144 de Baets et al., 2005 A.J. de Baets B.S. Edidi M.J. Kasali G. Beelaert W. Schrooten A. Litzroth Pediatric human immunodeficiency virus screening in an African district hospital Clin Diagn Lab Immunol 12 2005 86 92 Friedl et al., 2001 A.C. Friedl B. Ledergerber M. Flepp B. Hirschel A. Telenti H. Furrer Response to first protease inhibitor- and efavirenz-containing antiretroviral combination therapy. The Swiss HIV Cohort Study AIDS 15 2001 1793 1800 Ledergerber et al., 2000 B. Ledergerber M. Flepp J. Boni Z. Tomasik R.W. Cone R. Luthy Human immunodeficiency virus type 1 p24 concentration measured by boosted ELISA of heat-denatured plasma correlates with decline in CD4 cells, progression to AIDS, and survival: comparison with viral RNA measurement J Infect Dis 181 2000 1280 1288 Li et al., 2005 C.C. Li K.D. Seidel R.W. Coombs L.M. Frenkel Detection and quantification of human immunodeficiency virus type 1 p24 antigen in dried whole blood and plasma on filter paper stored under various conditions J Clin Microbiol 43 2005 3901 3905 Mascolini, 2002 Mascolini M. Transfer of HIV diagnostic and monitoring technologies into resource-poor settings, http://www.hivforum.org/publications/TechTransFinal.pdf ; 2002. Respess et al., 2005 R.A. Respess A. Cachafeiro D. Withum S.A. Fiscus D. Newman B. Branson Evaluation of an ultrasensitive p24 antigen assay as a potential alternative to human immunodeficiency virus type 1 RNA viral load assay in resource-limited settings J Clin Microbiol 43 2005 506 508 Respess et al., 2003 R.A. Respess A. Cachafeiro S.A. Fiscus D. Newman B. Branson O. Varnier Evaluation of a commercially available ultrasensitive p24 antigen viral load assay in samples from patients infected with genetically diverse HIV-1 from different geographic settings 10th conference on retroviruses and opportunistic infections Boston 2003 Ribas et al., 2003 S.G. Ribas P. Ondoa J. Schupbach G. van der Groen K. Fransen Performance of a quantitative human immunodeficiency virus type 1 p24 antigen assay on various HIV-1 subtypes for the follow-up of human immunodeficiency type 1 seropositive individuals J Virol Methods 113 2003 29 34 Schupbach, 2003 J. Schupbach Viral RNA and p24 antigen as markers of HIV disease and antiretroviral treatment success Int Arch Allergy Immunol 132 2003 196 209 Schupbach et al., 2003 J. Schupbach J. Boni L.R. Bisset Z. Tomasik M. Fischer H.F. Gunthard HIV-1 p24 antigen is a significant inverse correlate of CD4 T-cell change in patients with suppressed viremia under long-term antiretroviral therapy J Acquir Immune Defic Syndr 33 2003 292 299 Schupbach et al., 2005 J. Schupbach H.F. Gunthard B. Joos M. Fischer Z. Tomasik J. Boni HIV-1 p24 may persist during long-term HAART, increases little during short treatment breaks, and its rebound after treatment stop correlates with the CD4+ loss J Acquir Immune Defic Syndr 40 2005 250 256 Sherman et al., 2005 G.G. Sherman G. Stevens S.A. Jones P. Horsfield W.S. Stevens Dried blood spots improve access to HIV diagnosis and care for infants in low-resource settings J Acquir Immune Defic Syndr 38 2005 615 617 Sterling et al., 2002 T.R. Sterling D.R. Hoover J. Astemborski D. Vlahov J.G. Bartlett J. Schupbach Heat-denatured human immunodeficiency virus type 1 protein 24 antigen: prognostic value in adults with early-stage disease J Infect Dis 186 2002 1181 1185 Stevens et al., 2005 G. Stevens N. Rekhviashvili L.E. Scott R. Gonin W. Stevens Evaluation of two commercially available, inexpensive alternative assay used for assessing viral load in a cohort of human immunodeficiency virus type1 subtype C-infected patients form South Africa J Clin Microbiol 43 2005 857 861 Stevens et al., 1992 R. Stevens K. Pass S. Fuller A. Wiznia L. Noble S. Duva Blood spot screening and confirmatory tests for syphilis antibody J Clin Microbiol 30 1992 2353 2358 Sutthent et al., 2003 R. Sutthent N. Gaudart K. Chokpaibulkit N. Tanliang C. Kanoksinsombath P. Chaisilwatana p24 Antigen detection assay modified with a booster step for diagnosis and monitoring of human immunodeficiency virus type 1 infection J Clin Microbiol 41 2003 1016 1022 Tehe et al., submitted for publication Tehe A, Maurice C, Hanson DL, Borget MY, Abiola N, Maran M, et al. Quantification and evaluation of HIV-1 p24 by a highly improved ELISA as an alternative to HIV-1 RNA measurement in patients receiving antiretroviral therapy in Abidjan, Côte d’Ivoire, submitted for publication.