The Body Covers: HIV Genetic Diversity: Emerging Clinical Issues
Report on Genetic Diversity Conference
June 30, 1999
Presentation I: Epidemiology of Transmission of HIV-1 Non-B Types and HIV-2 in the United States
Coverage provided by Jay Dobkin, M.D.
A small evening conference held in New York last week addressed the great importance of non-subtype B strains of HIV around the world and increasingly in the United States. Three formal talks and a series of cases presented by two clinicians drove home this message: HIV strains have diverged widely in different areas and the relevance of diagnostic, treatment or pathogenesis notions based on studies of the subtype B strains predominating in the US and Western Europe must be questioned. In particular viral load assays may give misleading results when applied to a patient infected with non-B subtypes.
The conference entitled "HIV Genetic Diversity: Emerging Clinical Issues" was jointly sponsored by the African Services Committee, a social service agancy working with African immigrants and refugees, the New York City Health Department (NYCHD) and the Treatment Action Group (TAG) and funded by Bayer Diagnostics, the company which acquired the branch chain DNA assay from Chiron. It was held at the New York Academy of Medicine on June 30.
The discussions all revolved around the classification of HIV isolates. HIV-1 and HIV-2 are members of the genus Lentivirus. HIV-1 is subdivided into groups: M (for main) and O (for outlier). Within group M are the vast majority of HIV-1 strains subdivided into subtypes (currently 10: A-J) based on genetic variation in the env and gag regions of between 15 and 30%. Definitive classification requires DNA sequencing but simpler methods such as immunoassays which can detect subtype-specific peptides are increasingly utilized.
The distinction of HIV-1 from HIV-2 is relatively widely known among clinicians. Mostly, the HIV-1 strain is found in West Africa and nearly all US cases have had a historical connection with this area. Specific EIA and Western blot assays for HIV-2 make evaluation of patients with this infection relatively straightforward if the diagnosis is suspected. Far less widely known is the existence of non-B subtype HIV-1 infection. Again, most of the recognized cases have had a link to Africa, but little surveillance data is available to assess the prevalence or spread of these strains and even the methodology to perform these studies is still developing. Most significantly, it seems, is the likelihood that patients will be improperly managed by providers unaware of the problem.
The first of two lectures was presented by Max Essex, Ph.D. of the Harvard School of Public Health, a pioneer in the study of HIV evolution in Africa. He reviewed the compelling evidence that most of the ongoing HIV-1 epidemic around the world is due to non-B subtypes, especially subtype C in sub-Saharan Africa which has grown explosively since 1990 while the relative prevalence of the other common subtypes (A, D and E) has declined. Subtype B is virtually absent from this region. The rapid spread of HIV-1C corresponds to an epidemic that is predominantly heterosexual and has led to astonishing seroprevalence levels in some areas (38.5% of pregnant women screened in Botswana in July 1998 were positive). Dr. Essex reviewed emerging evidence for biological features of non-B subtypes which account for their apparently enhanced infectivity. HIV-1 viral load measurements from cervico-vaginal lavage fluid were higher in women infected with C than A or D subtypes. Several genetic markers have also been identified that may account for the behavior of subtype C. A regulatory region enhancer sequence is repeated 3 or 4 times in C strains as opposed to twice in HIV-1B and once in HIV-2 strains. Responsiveness to TNF-alpha by increased replication is much greater for C than B strains, possibly making these strains more infectious in the setting of another concurrent sexually transmitted disease. The far greater rate of genetic diversity noted among subtype C than B strains also suggests a higher replication and/or mutation rate, Dr. Essex noted.
He also referred to the finding of more rapid HIV disease progression in members of a West African cohort infected with HIV-1C than A, G or D subtypes and commented that little is known about the sensitivity or therapeutic response to anti-retroviral drugs among non-B strains. He speculated that even if the initial response to treatment was similar, the tendency of these strains toward higher replication rates might lead to early resistance and treatment failure.
Presentation II: HIV-1 Non-B Subtypes and HIV-2 Laboratory Diagnosis, Monitoring and Clinical ManagementSpeaker: Sara T. Beatrice, Ph.D.
Coverage provided by Jay Dobkin, M.D.
The second speaker of the evening, Sara T. Beatrice, Ph.D., of the New York City Department of Health Bureau of Laboratories reviewed the technical issues raised in distinguishing infections with non-B subtypes of HIV-1 and measuring viral load with these strains. In a lookback program 8800 specimens submitted for HIV testing from 1993-97 which noted African birth were screened for non-B HIV out of a total of about 800,000 samples sent for HIV testing. 186 specimens were HIV-2 EIA and Western blot positive of which 43 proved to be from HIV-2 infected individuals and 143 from patients with HIV-1 infection, the great majority (117) of whom were infected with non-subtype B strains. Of the 43 HIV-2 patients 31 had been positive on the standard HIV-1 assays, 9 negative and 3 indeterminate.
Dr. Beatrice also reported that using an experimental EIA that is intended to distinguish HIV-1 subtypes, her lab this year has detected non-B strains in the majority of HIV positive immigrants from several areas: Africa (85%), Asia (89%) and South America (51%). She then addressed problems of viral load testing associated with non-B subtypes. All available HIV-1 viral load assays are unreliable for Group O HIV-1 and HIV-2 strains. Because it uses a single set of primers standardized on HIV-1B strains, pcr is the most likely to give inaccurate results with other strains. Assay results may underdeteted the amount of viral RNA from subtypes A, E, F, and G as well as group O. A new pcr assay in development (version 1.5) is designed to overcome some of these limitations. The NASBA assay may also underdetect subtypes A and E.
The branch chain DNA (bDNA) assay is generally thought to be least affected by strain differences because it utilizes a mixture of probes of varying sequences. However, as the subsequent case presentation demonstrated, even this approach is not foolproof.
Presentation IV: Antiviral Drug Susceptibility of HIV-1 Non-B Subtypes and HIV-2Speaker: Chou-Pong Pau, Ph.D.
Coverage provided by Jay Dobkin, M.D.
The first of the two clinician presenters was Elizabeth Jenny-Avital, M.D.of the Bronx Municipal Hospital Center and the Albert Einstein College of Medicine. She reported observations of a series of 7 African immigrant patients with surprisingly low viral load results by pcr most of whom had non-B subtypes. Comparisons of viral load data by simultaneous pcr and bDNA in these cases revealed striking discrepancies as large as 2 to 3 logs. In one case a pcr result of <400 copies contrasted with a result of 92,000 copies by bDNA. Two other patients who were nearly undetectable on pcr had high viral loads when bDNA was done (503 vs. 80,000 copies and 543 vs. 92,000 respectively). Dr. Jenny-Avital pointed out that mistakes in antiretroviral therapy are likely, given these misleading findings. She also reported results of a review of the 14 non-B infections among 16 African patients in her clinic. Four on no HIV therapy had comparable viral loads by both assays, 5 had low viral loads on treatment with both assays (4 were undetectable on both tests), and 5 others had significant underdetection of viral load by pcr compared to bDNA.
Several cases presented by another physician, Dr. John Corser of Metropolitan Hospital Center and New York Medical College, cast doubt on the reliability of the bDNA assay in some non-subtype B infections. One patient with persistently undetectable viral load by pcr despite a falling CD4 count was also less than 500 copies by bDNA.
The final lecture given by Chou-Pong Pau, Ph.D., of the Centers for Disease Control was "Antiviral Drug Susceptibility of HIV-1 non-B subtypes and HIV-2." Limited by the availability of samples, especially subtype C, he described a general pattern of baseline susceptibility by most non-B strains with the development of typical resistance mutations after treatment failure. HIV-2 and Group O HIV-1 strains may have inherent resistance to some drugs.
For a relatively modest event this conference conveyed several powerful messages. One is that clinicians clearly need to be vigilant regarding unusual treatment responses that could reflect infection with non-subtype B strains. More sobering still is the recognition that the American HIV epidemic and all that has been done to contain it may be a small island in a vastly larger and more dangerous process about which we know relatively little.
This article was provided by The Body PRO. Copyright © Body Health Resources Corporation. All rights reserved.