Detection of HIV-1 antibodies in blood specimens spotted on filter-paper.

Detection of HIV-1 antibodies in blood specimens spotted on filter-paper F. Lillo,' O.E. Varnier,1 E. Mantia,2 A. Terragna,2 G. van der Groen,3 1. Van kerckhoven,3 P.R Mortimer,4 J.V. Parry,4 G. Bayliss,4 & H. Tamashiro5 Described are the results of an international collaborative study to evaluate the use of whole blood samples spotted on filter-paper (BSP) for the detection of antibodies to human immunodeficiency virus type 1 (HIV-1). BSP samples were collected from 40 patients at risk for HIV-1 infection and tested blindly using commercially available HIV antibody test kits, either specifically manufactured or modified for this purpose. Parallel serum samples were also collected, and the antibody reactivity was defined and confirmed by Western blot. The results demonstrate that recovery of antibodies from BSP samples after elution can be compa- rable to that from serum. Some kits can be easily adapted to test BSP samples, while others cannot. At present, detection of HIV antibodies in BSP samples should therefore be carried out using kits specifi- cally manufactured for this purpose or by the development of a modified protocol using a panel of BSP and their corresponding serum specimens. Introduction The identification of human immunodeficiency virus (HIV) infection is mainly based on the detection of specific antibodies to viral antigens in samples of serum or plasma (1). Although much progress has been made in developing diagnostic reagents, anti- body reactivity still requires confirmation; multiple samples are therefore often needed to complete testing (2). Epidemiological studies can be hampered by the difficulty in obtaining adequate samples of blood, particularly from infants or elderly patients, and storage of aliquots may be difficult in devel- oping countries. In the last 20 years newboms have been screened for metabolic disorders by testing blood spotted on filter-paper (BSP) (3). This method has also been used to diagnose several diseases, such as measles (4), viral hepatitis B (5), and, more recently, to detect antibodies to HIV (6-11). ' Laboratory of Human Retrovirology, Institute of Microbiology, School of Medicine, V. le Benedetto XV, 10, 16132 Genoa, Italy. Requests for reprints should be sent to Dr Lillo at this address. 2 First Infectious Diseases Clinic, School of Medicine, Genoa, Italy. 3 Institute of Tropical Medicine, Department of Microbiology, Antwerp, Belgium. 4 Virus Reference Laboratory, Central Public Health Laboratory, London, England. 5Diagnostics Unit, Office of Research, Global Programme on AIDS, World Health Organization, Geneva, Switzerland. Reprint No. 5282 The collection of whole blood on filter-paper for antibody assay has unique advantages over the use of serum samples. Equipment requirements are minimal - inexpensive sterile lancets and filter-papers re- place the syringes, tubes, centrifuges, refrigerators, and freezers that are needed for serum collection and sto-rage. The filter-cards used are light, hold up to eight aliquots, cannot be broken or spilt, can be stor- ed at room temperature for several weeks, require minimal storage space, and can be sent by mail. The BSP technique is therefore particularly suitable for use in screening programmes in developing countries. Use of the BSP technique to screen for HIV antibodies needs to be validated to verify whether the results obtained in different assays are com- parable to those obtained by analysis of serum or plasma. The present international collaborative study therefore assessed the usefulness of different HIV antibody assays of BSP samples, using a panel of 40 paired samples of BSP and the corresponding serum samples. Methods Specimens Forty blood samples (10 ml each) were collected in 1989-90 by venepuncture using heparinized syringes from patients (25 intravenous drug addicts, 11 sexual partners of HIV-l-positive subjects, and 4 homo- sexual males) who were attending the Clinic of Infectious Diseases, Genoa. Whole blood was spot- Bulletin of the World Health Organization, 70 (3): 323-326 (1992) © World Health Organization 1992 323 F. Lillo et al. ted slowly to avoid splashing and to saturate the filter-paper (Schleicher & Schull, grade 903) until blood was visible on the reverse side of the test card and filled the delineated circle of 1-cm diameter (about 250 ,l per spot). It was not possible to collect the samples by finger-stick because of the need to prepare several spotted test cards (6 coded cards with 8 BSP each) to distribute to the three collaborating laboratories. The cards were kept at room tem- perature until the blood had dried, stored in sealed plastic bags, and sent to the laboratories wrapped in double plastic envelopes that were placed in a box marked "biohazard". Serum was separated from the blood samples and stored frozen until tested at the Institute of Microbiology, Genoa. The antibody specificity was confirmed by Westem blot analysis (HIV-1 Western blot, Du Pont) using WHO interpre- tation criteria. Each participating laboratory tested blindly the BSP cards and was given information on the sero- logical status of each sample only at the end of the study. Procedures The optimal performance of an anti-HIV assay requires strict application of the recommended pro- cedures, in particular the sample dilution and vol- ume. A paper disc of 5-mm diameter contains approximately 10 gl of serum, and therefore the elu- tion procedure was adjusted to obtain the appropriate dilution for each test. This was carried out by varying the number of discs added and/or the volume of the diluent. Elution protocol The protocol described below was used. * By means of a calibrated puncher, paper discs of 5-mm diameter were cut out from the BSP cards. To avoid cross-contamination between the samples, the puncher was carefully cleaned using a deter- gent solution. * Using a blank microtitration plate, the 5-mm filter-paper discs were placed flat at the bottom of the assigned well on the plate. * The appropriate amount of PBS was added to each well containing the filter-paper discs and the plate was then incubated overnight at 4 'C. * This "elution plate" was warmed up to room tem- perature and the eluates mixed thoroughly. * The appropriate amount of each eluate was trans- ferred from the elution plate to the assigned well of the reaction plate. HIV-1 kits The only commercially available kit that is manufac- tured to test BSP samples is the Du Pont HIV recom- binant (ENV9) enzyme-linked immunosorbent assay (ELISA), which on request is provided with a dedi- cated sample diluent. This test was adopted as the reference test by all the collaborating laboratories in the study. Since the other kits were not validated for testing BSP samples, partially modified protocols were used. Here only the results obtained with those kits that proved to be satisfactory for this purpose are reported. The manufacturers of the HIV assays that were unsatisfactory for testing BSP samples in our study will be informed and invited to develop speci- fic protocols or to advise their customers not to use their kits for this purpose until specific protocols have been developed. The following kits were satisfactory: - the Cellular Product Inc. ELISA prepared with HIV-1 virus lysate;a - the Behring "Enzygnost Anti-HIV Micro" com- petition ELISA;b - the Fujirebio "Serodia-HIV" particle agglutina- tion test;c - the IgG antibody-capture ELISA (GACELISA)d developed at the Virus Reference Laboratory, London, England, in collaboration with Well- come Diagnostics. This assay employs a recom- binant-antigen directly conjugated to alkaline phosphatase; and - the GACPAT,d a capture assay based on a modi- fication of the Serodia-HIV test (12). Results All the 40 BSP samples were tested using the Du Pont HIV recombinant (ENV9) ELISA, in parallel with the corresponding serum samples. The BSP results were concordant with those obtained for the sera, with the exception of one negative sample, which was initially reactive but negative when re- tested (Table 1). The Cellular Product Inc. ELISA detected the presence of HIV-1 antibodies in all the positive BSP and serum specimens. Five HIV-l-seronegative BSP samples were initially reactive in this test and two of these samples were also repeatedly reactive when retested: these two samples were reactive only for p24 in the HIV-1 Western blot. a Cellular Product Inc., Buffalo, NY, USA. b Behringwerke AG, Marburg, Germany. cFujirebio Inc., Tokyo, Japan. d Wellcome Diagnostics, Dartford, England. 324 WHO Bulletin OMS. Vol 70 1992 Detection of HIV-1 antibodies on filter-paper blood spots Table 1: Results obtained for the detection of HIV-1 antibodies in blood spotted on filter-paper (BSP) and serum samples using various commercially available test kits Serum samples BSP Initial test Repeat test Initial test Repeat test Concord- Concord- Kit No. +ve No. -ve No. +ve No. -ve ance (%) No. +ve No. -ve No. +ve No. +ve ance (%) Du Pont HIV-1 34 6 - - 100 35 5 34 6 100 recombinant ELISA Cellular Product 34 6 - - 100 39 1 362 4 95 Inc. ELISA Enzygnost Anti- 34 6 - - 100 34 6 - - 100 HIV Micro GACELISA 34 6 - - 100 34 6 - - 100 GACPAT 34 6 - - 100 34 6 - - 100 Serodia-HIVb 35 5 35 5 97.5 35 5 35 5 97.5 a Two HIV-1 negative samples were p24 reactive in the Western blot. b One HIV-1 negative sample was p24 reactive in the Western blot. The Behring Enzygnost Anti-HIV Micro ELISA detected HIV-1 antibodies in all the positive BSP and serum samples, without any false reactions and with a final 100% concordance. The same results were also obtained with the GACELISA and the GACPAT. With the Serodia-HIV particle-agglutination test only one HIV-1-negative serum sample, but not the corresponding BSP, was repeatedly positive. Wes- tern blot analysis of the serum revealed the presence of p24 reactivity. The Serodia-HIV was also used at the Virus Reference Laboratory to test the BSP panel: the results obtained were concordant with the exception of one HIV-1-negative sample, which was reactive (this serum sample was indeterminate in the Western blot with a p24 reactivity). Conclusions The results that we have reported confirm and extend previous observations that whole blood collected on filter-paper can be effectively used instead of serum samples for HIV-antibody testing. The method can easily be carried out by individuals who have received only a little training, taking the following storage precautions: the BSP need to be carefully dried after collection and stored wrapped in sealed plastic bags in a dry, dust-free place. Other HIV assays should be investigated to determine whether they are suitable for use with BSP specimens. Some commercial kits may easily be adapted for this purpose as shown by our results, while others may not be suitable for use in their original form or may require to be modified, e.g., by the provision of adequate elution diluents by the manufacturers as accessory reagents. At present, however, the detection of HIV antibodies in BSP samples should be performed using only specifically manufactured kits. Whenever there is the need to test BSP samples and such kits are not available, either the kits reported here to be satisfactory should be used or a modified protocol should be developed using a panel of BSP and their corresponding serum specimens. Acknowledgements The HIV-antibody kits evaluated in this study were kindly provided by the WHO Global Programme on AIDS, Geneva, Switzerland. Resume Detection des anticorps anti-VIH-1 dans des pr6levements sanguins recueillis sur papier filtre Le diagnostic de l'infection par le virus de l'immu- nod6ficience humaine type 1 (VIH-1) s'appuie principalement sur la detection d'anticorps speci- fiques dans le serum des sujets infectes. Bien que de grands progres aient ete faits dans la pre- paration des reactifs d'identification de ces anti- corps, les resultats positifs doivent toujours etre confirmes et plusieurs echantillons de s6rum sont souvent necessaires. Les ponctions veineuses sont parfois difficiles a realiser chez les nourris- sons ou les personnes agees et la conservation des prelevements sanguins peut presenter des difficult6s dans les pays en developpement. WHO Bulletin OMS. Vol 70 1992 325 F. Lillo et al. L'etude collective presentee ici visait a veri- fier si des 6chantillons de sang total recueillis sur papier filtre pouvaient effectivement etre utilises pour diagnostiquer l'infection a VIH-1 et pour obtenir des donnees epidemiologiques dans les pays en developpement. Des echantillons de s6rum pr6lev6s de fa,on classique et des 6chan- tillons de sang recueillis sur papier filtre ont ete obtenus aupres de 40 patients de l'Institut des maladies infectieuses de Genes et envoyes aux laboratoires participants. Ces echantillons ont ete test6s a I'aide de trousses commerciales d'anti- corps anti-VIH conques specialement pour les essais sur papier filtre ou adaptees a cette fin. Les resultats des essais qui se sont rev616s satis- faisants sont resumes ci-apres. La trousse Du Pont ENV9 (sp6ciale pour essai sur papier filtre), 1'essai immuno-enzymatique par competition de Behring, le GACELISA avec capture d'IgC et le GACPAT (modification de l'6preuve d'agglutina- tion de particules de Serodia) ont donn6 des r6sultats concordants pour les serums et les 6chantillons de sang sur papier filtre. L'essai immuno-enzymatique sur lysat de virus de Cellu- lar Product Inc. et l'epreuve d'agglutination de Serodia ont donne des taux de concordance de 95% et 97,5% respectivement, en raison des faux positifs. Les r6sultats de cette 6tude confirment et completent les observations anterieures selon lesquelles des 6chantillons de sang recueillis sur papier filtre pouvaient effectivement remplacer les echantillons de serum pour la recherche des anti- corps anti-VIH. Toutefois, les trousses utilisees aux fins de diagnostic doivent etre specialement conques en fonction de cette technique ou v6ri- fiees avec une serie d'6chantillons de sang sur papier filtre bien caracterises et avec les serums correspondants. References 1. Sarangadharan, M.G. et al. Science, 224: 504 (1984). 2. Varnier, O.E. et al., ed. Recents advances in AIDS and Kaposi's sarcoma. Basel, Karger, 1987, p. 180. 3. Guthrie, E. & Susi, A. Pediatrics, 32: 338 (1963). 4. Wassilak, S.G.F. et al. Pediatric infectious diseases, 3: 117 (1984). 5. Farzadegan, H. et al. Lancet, 1: 362 (1978). 6. Farzadegan, H. et al. Journal of infectious diseases, 155: 1073 (1987). 7. Lindhardt, B.O. et al. Journal of virological methods, 18: 73 (1987). 8. Grady, G.F. et al. Abstracts of the IlIl International Conference on AIDS, Washington, DC, 1-5 June 1987. Abstract THP 175, p. 192. 9. Evengard, B. et al. British medical journal, 297: 1178 (1988). 10. Varnier, O.E. et al. AIDS research and human retroviruses, 4: 131 (1988). 11. George, J.R. et al. Serologic assays for human immunodeficiency virus antibody in dried blood spe- cimens collected on filter-paper from neonates, (Centers for Disease Control, Monograph) August 1989. 12. Parry, J.V. & Mortimer, P.P. AIDS, 3: 173 (1989). 326 WHO Bulletin OMS. Vol 70 1992