The authors summarized the experience of 2,674 Swiss residents (730 of them women) who began a protease inhibitor-containing combination between 1995 and 1998. About 43% had had no prior treatment and the rest had received prior nucleoside analogs. The members of this group were fairly advanced: The median CD4 count when starting therapy was only 192 cells/mm3. The initial viral load was 31,000 copies/ml even though a majority of the cohort was already receiving some treatment.
After six months on HAART, 82% of those who were treatment-naïve had achieved viral loads below 400 copies/ml, the bottom limit of the viral load assay in use. This figure gradually fell to 66% for those who had been on HAART for 30 months, and 60% of these people had changed therapies at least once. This leaves a 30-month success rate for initial HAART therapies of only 40%. The 30-month success rate for persons with nucleoside analog treatment before starting HAART was still lower -- 14.5% and depended on the number of new drugs introduced along with the first protease inhibitor. Other factors that increased the risk of failure include high baseline viral load, low CD4 count, advanced disease and the use of the protease inhibitor Invirase (the old, poorly absorbed version of saquinavir).
Members of the Swiss cohort did not have to reduce their viral loads below the limits of quantification to lower their risk of new AIDS-defining conditions or death. The 30-month rate for such disease progression was 6.6% in the persons with viral loads remaining below 400 copies/ml and just a bit more, 9.0%, in people with HIV that had rebounded to higher viral loads. For persons that had never seen their viral loads go below 400 copies/ml, the 30-month incidence of new AIDS-related illness or death was 20%. These rates were substantially lower than those of Swiss cohort members with the same CD4 counts who started single and dual nucleoside analog therapy in earlier epochs.
The researchers related this difference to the major sustained increases in CD4 count exhibited by people on HAART regardless of the extent of viral suppression. Their observations support considerable speculation that HAART-resistant HIV is somehow less fit or less able to cause disease.
The source of these inhibitory effects has been controversial. The active molecule could be the beta subunit of HCG, which is a protein made up of alpha and beta subunits. Alternatively, the ameliorating effects might stem from one of the other proteins present in commercial HCG preparations. (These products are extracts from the urine of pregnant women.) A report analyzing the antiviral effect of two such enzymes has just been published (S. Lee-Huang et al., Proceedings of the National Academy of Sciences USA, March 1999, pages 2678-81).
The authors fractionated commercial HCG, separating out proteins and HCG subunits on the basis of charge and size. They then characterized the amino acid sequences of the collected fractions. The isolated proteins comprised the following: urinary beta-core HCG (a degradation product from the subunit), human RNase U, urinary lysozyme C. Human RNase A exists as part of the HCG beta-core and was difficult to purify.
When the authors screened these proteins for anti-HIV activity in cell lines chronically infected with HIV-1, there were a few surprises. beta-core HCG did not reduce production of the HIV core protein (p24), whereas the isolated lysozyme C and RNase U reduced p24 production by up to seven-fold in a concentration-dependent manner. Bovine pancreatic RNase A showed similar activity, as did lysozymes from human milk, human neutrophils and chicken egg whites.
RNases typically break up RNA (ribonucleic acid) in cells, whereas lysozymes attack the complex sugars (polysaccharides) in bacteria cell membranes. The authors tested whether the purified HCG-associated proteins performed in this fashion. When total cellular RNA from HIV-1 infected lymphocytes was incubated with individual protein fractions and the products run on an RNA separating gel, the human RNase U fraction, but not the HCG beta-core, degraded the RNA. Likewise, the isolated human lysozyme C lysed the bacteria Micrococcus lysodeikticus.
These anti-HIV proteins are comparable in molecular weight to active agents isolated from HCG preparations, but not further identified, in previous studies by the University of Maryland Institute of Human Virology led by Robert Gallo. (See Y. Lunardi-Iskandar et al., Nature Medicine, April 1998, pages 428-34). Since they occur naturally and are not toxic to cell cultures at the doses tested, they may be keys to new and safer strategies for treating HIV infection.
In an interview, Dr. Gallo noted that several research teams in the past year have tried to identify the active antiviral agents in HCG preparations. He said, for example, that the molecules identified by a research group from New York University "might have an anti-HIV effect, but they're not what we are looking for." The so-called HAF, or HCG-associated factor, that his lab previously isolated has now been identified, and a report on its characteristics will be written later this year. The natural function of that molecule seems related to early fetal development and the killing of superfluous cells.
Researchers studied a total of 60 persons from four different EU cohorts. The cohorts comprised a diverse population: 63% Caucasian, 31% African-American and 6% Hispanic as well as 35% female. All participants reported regular, unprotected sexual contact with HIV-positive partners. Two individuals also shared syringes with HIV-positive persons. All 60 had remained HIV-negative and had no HIV-specific antibodies in their blood. Only 56% were also negative for anti-HIV antibodies in their urine, indicating that contact with HIV had been sufficient to provoke mucosal membrane antiviral immune activity in some.
The vast majority of the people studied (58 of 60) lacked the CCR5 deletion mutation. CD4 cells from eight randomly selected EUs and four unexposed controls were equally susceptible in lab tests to infection by HIV isolates, including those taken from sexual partners. Anti-HIV CTLs did not seem to be playing a decisive role in preventing HIV replication, either. In one subgroup, 14 of 15 people had no sign of an anti-HIV CTL response.
One mechanism that affords protection from disease progression in HIV-infected individuals is a strong non-cell-killing anti-HIV CD8 cell response that suppresses HIV replication by means of an as yet unidentified secreted factor. Almost half (46%) of EU study participants exhibited enhanced CD8 cell suppression of HIV replication compared to unexposed controls during laboratory cell culture tests. CD8 suppressor activity was confirmed when protection was abrogated by excluding CD8 cells from the assay cultures.
Since immune responses follow exposure to foreign antigens, the researchers classified EU participants according to their HIV exposure history: very high, moderate and low risk (multiple exposures in the past six months, between six months and a year, and greater than a year, respectively). Higher HIV exposure category correlated with increased suppression of the HIV isolates tested. The researchers argued that regular exposure to HIV enhanced antiviral immunity in the EU cohort members, in a fashion comparable to a booster vaccine.
Back to the GMHC Treatment Issues March 1999 contents page.