The findings and conclusions in these presentations have not been formally disseminated by the Centers for Disease Control and Prevention and should not be construed to represent any agency determination or policy.

Tuesday, May 9, 2006
126

Analysis of Index Value Using the Focus HerpesSelect® 2 ELISA IgG in Estimating HSV-2 Seroreversion in an Indian Population

Patricia Rizzo-Price1, Charlotte Gaydos1, David D. Celentano2, Suniti Solomon3, P. Balakrishnan3, Oliver B. Laeyendecker4, Thomas C. Quinn4, and NIMH Collaborative HIV/STD Prevention Trial Group5. (1) Division of Infectious Diseases, Medicine, Johns Hopkins University, 1156 Ross Bldg, 720 Rutland Ave, Baltimore, MD, USA, (2) Epidemiology, Johns Hopkins University, 615 North Wolfe Street, Room E-6008, Baltimore, MD, USA, (3) YRG Centre for AIDS Research and Education (CARE), 1, Raman Street, Tamil, Nadu, T. Nagar, Chennai, India, (4) NIAID, National Institutes of Health, 1159 Ross Bldg, 720 Rutland Ave, Baltimore, MD, USA, (5) NIMH Center for Mental Health AIDS Research, National Institutes of Mental Health, 6001 Executive Boulevard, Bethesda, MD, USA


Background:
The Focus HerpesSelect 2 ELISA IgG test detects human IgG antibodies to HSV-2 in human sera. An Index Value cut-off 1.1 determines positive results and may increase false positives in developing countries

Objective:
To evaluate the Focus® cut-off used to test a population in India, and estimate false positives.

Method:
Sera from 2,186 participants of The NIMH Collaborative HIV/STD Prevention Trial in Chennai, India were obtained from men and women. Samples were collected every12-months and tested using the Focus Kit. Index Values > 1.1 were positive for HSV-2; Index Values of 0.9 -1.1 were indeterminate, and index values < 0.9 were negative. Data comparing the Index Values at visits 1 and 2 were evaluated at the Johns Hopkins Reference Lab.

Result:
Of 2,186 results, 1,480 (67.6%) tested negative at both visits, and 511 (23.4%) tested positive at both visits. Of the 511 samples, 103 (20.1%) index values were < 3.5, and 408 (79.8%) were >3.5. Of the 2,186 participants, 86 (3.93%) tested positive at visit 1, and negative/indeterminate at visit 2. Of these 86 samples, 75 (87.2%) had Index Values < 3.5, and 11(12.8%) were > 3.5 at visit 1. Of 2,186 there were 32 (1.46%) negative/indeterminate results at visit 1 that tested positive at visit 2. Of the 32 results, 12 (37.5%) had Index Values > 3.5 at visit 2, possible seroconverters, whereas the remaining 20 (62.5%) had Index Values <3.5 at visit 2, possible false positives.

Conclusion:
Using 1.1 as a cut-off may increase the number of false positive results, and seroreverters in the Chennai population. Further evaluation is warranted, to determine if the cut-off is too low, estimate the possible effects on seroreversion/false positives, and confirm testing by Western Blot.

Implications:
When kit performance is as expected, an adjusted cut-off comparative to the population tested may increase testing accuracy, and PPV.