New Testing for Very Early HIV Diagnosis

The commonly used HIV antibody test does not catch a very recent HIV infection, because it takes a few weeks for the body to create enough antibodies to show in the test. Other tests can detect the virus directly, but because they are expensive and pick up relatively few cases, they have not been routinely used in HIV testing. Now there is a public-health push to get the earlier results. This is because persons who are newly infected but have not developed antibodies yet are likely to have extremely high viral loads, and therefore have a high risk of transmitting the infection to others; a significant fraction of new infections may be transmitted this way. And when public-health experts can see immediately where new infections are headed, they can target prevention campaigns more effectively. Also, patients can get medical care for the early ("primary") HIV infection.

A report from North Carolina, where the state's 110 public HIV testing sites have offered the enhanced HIV testing since November 2002, provided the following information.¹ Of 109,788 persons tested, 622 were found to be HIV infected; of these, 21 were negative on the regular antibody test but found to be infected by directly testing for the virus. Nineteen of them (including one pregnant woman) began medical care and were offered antiretroviral treatment. And the testing program gave the first evidence of an epidemic of HIV infection among college men, mostly African Americans, at least in North Carolina, where the information was collected² -- allowing better targeting of prevention efforts.

A different approach to testing for recent HIV infection is STARHS (Serologic Testing Algorithm for Recent HIV Seroconversion). In STARHS, persons who have tested positive with the usual antibody test have a portion of their serum saved for a second antibody test, deliberately made less sensitive (it is sometimes called the "detuned assay"). Those who were infected recently, within about the last year, may not have enough antibodies yet to test positive on the detuned test, even though they tested positive on the regular HIV test. Note that STARHS only applies to persons who have tested positive with the regular HIV test -- not those picked up only by a direct test for the virus, whose serum would certainly show up negative on the detuned test used for STARHS.

Notes

• STARHS testing is not very reliable for an individual -- but more accurate for a population, since it wrongly classifies about the same number of people as early or late infection.

• STARHS is needed in addition to the direct viral test, because it picks up many more new infections, helping epidemiologists better understand where the epidemic is moving now.

• The public-health use of STARHS was delayed when Abbott Laboratories withdrew its test from the market. Now another detuned test is available.

References

1. C Pilcher, E Foust, J McPherson, and others. The screening and tracing active transmission program: Real-time detection and monitoring of HIV incidence. 11th Conference on Retroviruses and Opportunistic Infections, San Francisco, February 8-11, 2004 [abstract 20].

2. LB Hightow, P MacDonald, CD Pilcher and others. Transmission on campus: Insights from tracking HIV incidence in North Carolina. 11th Conference on Retroviruses and Opportunistic Infections, San Francisco, February 8-11, 2004 [abstract 84].

ISSN # 1052-4207

Copyright 2004 by John S. James. Permission granted for noncommercial reproduction, provided that our address and phone number are included if more than short quotations are used.