Loss of HIV antibodies does not signal HIV clearance, even two years later

Disappearance of HIV antibodies in people who received treatment soon after infection, and who had tested positive, is not unusual, but doesn’t indicate clearance of HIV infection, say researchers from the University of California San Francisco, reporting in the March 1^st edition of Clinical Infectious Diseases (now available online).

Occasional reports of people who lose HIV antibodies after a period of treatment have prompted claims that such people are `cured` of HIV infection. However, no one knows how frequent such cases might be, and whether they represent genuine HIV clearance.

Researchers at the Options Project, a cohort of San Francisco patients recruited during acute HIV infection to begin antiretroviral treatment, identified 87 individuals who had begun treatment within 28 days of being diagnosed with HIV infection through either viral load testing in the presence of a negative antibody result, or a positive HIV antibody test.

Twelve individuals had recent HIV infection (identifiable only by viral load testing or by an indeterminate antibody test and a positive viral load test). The remainder had early HIV infection, identified by a positive antibody test.

Blood samples from individuals on treatment were tested for HIV antibodies twice: after 48 weeks on treatment, and at the latest time point at which the individual had an undetectable viral load (a median of 105 weeks after starting treatment). In cases where loss of antibody reactivity (seroreversion) occurred, stored blood samples were tested more frequently, using a variety of antibody tests.

Antibodies testing was carried out using four different second and third generation ELISA tests (the standard first-line antibody test), and two different confirmatory Western Blot tests.

Of the twelve who lacked antibody responses before beginning antiretroviral treatment, all but one developed antibodies on standard tests. One patient, antibody negative at baseline but with a viral load above 500,000 copies/ml, tested positive only with a third-generation antibody test, and developed a positive Western blot result at week 16 only to revert to indeterminate Western blot status by week 24. However, third generation antibody test results remained persistently positive during follow-up out to 60 weeks of treatment.

Five individuals who had tested antibody positive at baseline experienced negative antibody results using second-generation tests on at least one test after starting treatment, and in all cases had negative antibody status at the time of their last undetectable viral load result. However positive Western blot results persisted over time for the HIV envelope protein gp160 and core protein p24, even if reactivity to other HIV proteins such as gp120 and gp41 was lost over time.

When three individuals who had demonstrated seroreversion were tested for HIV-specific T-cell responses, one was found to have no responses to any HIV proteins at baseline or at end of the follow-up period. This was the patient with a baseline viral load above 500,000 copies and indeterminate antibody responses on Western blot testing after week 24. This patient was continuing to take antiretroviral therapy at the end of the follow-up period.

However, the remaining five patients who experienced seroreversion on treatment all elected to stop antiretroviral therapy, and all experienced rebound of HIV viral load and re-emergence of antibody responses.

The incidence of seroreversion in the population studied was 7% during a median treatment period of 105 weeks, with a trend towards a greater likelihood of seroreversion with longer duration of treatment.

The only factor that was predictive of seroreversion was a low antibody titre, or value, in the initial `detuned` antibody assay. A `detuned` assay is designed to pick up early traces of HIV antibodies for surveillance purposes, blood donor screening and research, but is not used as a standard diagnostic test.