icon-folder.gif   Conference Reports for NATAP  
  9th Conference on Retroviruses and Opportunistic Infections
Seattle, Washington, February, 2002
Back grey_arrow_rt.gif
Reported by Christopher D. Pilcher, MD of the University of North Carolina at Chapel Hill Center for AIDS Research
- Treatment & STIs during acute infection
- Eradication
- Risk of transmission increased during acute infection
- Increased rates of drug resistant virus
- Superinfection: transmission of multiple viruses
- Limits of HIV antibody test
As its name implies, primary HIV infection (PHI) is the first phase in the natural history of HIV infection. It begins the moment a virus invades, and lasts as long as it takes for the infected person's immune response to finally fight back. Primary infection is generally considered over after about 4 months in, by which time an equilibrium has been established between the virus' ability to proliferate and the host's immune control: the relative strength of these two forces determines the "set point" or steady state viral load that then sets the course for each individual's infection.
Most people are symptomatic beginning about 14 days after exposure and typically suffer from days to weeks of fever with or without muscle and joint aches, fatigue, headache, sore throat, swollen glands and sometimes rash. This "acute retroviral syndrome" is rarely diagnosed since it is absolutely indistinguishable in most cases from infectious mononucleosis or other very common ailments. The rarity of diagnosis makes it a very difficult area to study and makes the observations gained from small numbers of patients precious. Recent research is showing that identification of infected persons during PHI can be a critical window of opportunity on several fronts, emphasized by presentations at the 9th Retrovirus conference.
One of the most exciting developments in the field has come from the seminal observations of Bruce Walker, Eric Rosenberg and their Harvard colleagues that a series of patients treated with HAART very early in PHI and then submitted to a series of brief treatment interruptions were eventually able to control their viremia at unusually low levels for many months without drugs. The explanation for this finding has been that since CD4 cells are key components of the host's immune response to HIV, shielding a patient's CD4's from destruction during the early phase can boost one's innate ability to respond vigorously to the virus: the theory of "immune rescue". Two PHI treatment trials presented as posters at this year's conference presented important insight into the potential of this strategy.
Hoen and colleagues (Abstract 530-M) presented preliminary results from PRIMSTOP, a 29 patient trial in which patients with PHI are treated with ddI, d4T, HU and NFV, then have initially brief but progressively longer STI's starting about 9 months into therapy. STI's are 2 weeks long at month 9, 4 weeks long at month 12 and 6 weeks long at month 15. Ultimately a final "readout" stop of therapy is planned when patients reach 24 weeks. Results through the 3rd stop are available for 12 patients. These patients started therapy with a median VL of only 190,000-almost 10-fold lower than the Rosenberg cohort, indicating they were somewhat further out from their infections when therapy was initiated. Nevertheless, by the 3rd stop 8/12 patients spontaneously controlled viruemia to <10,000 copies/ml and 5/12 were <1,000 copies without drugs. These findings are preliminary and the trial is ongoing; but they are extremely encouraging that the STI strategy may be able to benefit somewhat less highly selected patients than those studied by the Harvard group.
Miro and colleagues (poster 529-M) presented contrasting data from a Spanish trial enrolling patients with previously treated PHI. The protocol plans 14 successive 2-4 month STI's, each separated by 2 months back on therapy; results were reported for the first 12 patients to complete 3 STIs. These 12 subjects were treated much later in their infections (enrolled a median of 11 weeks after onset of acute retroviral symptoms-as compared to 3 weeks for PRIMSTOP, and even less time for the Harvard study). The patients were also treated with HAART for longer before their 1st STI (24 months on average) and were stopped for a much longer time (2-4 months rather than weeks) at each STI. Results were dismal: overall there was no trend to improvement by the 3rd stop. Although one patient failed to rebound even after 4 months off treatment in his 3rd stop, and 3 others had lower peak viremia during their 3rd than 2nd stop, most patients still continued to rebound quickly to high levels of viremia, steadying out at somewhere between 20 and 100,000 copies per ml by the end of each stop.
The message from these two trials, then, is mixed. On the one hand, repeated, brief and rapid interruptions over a year or two may actually enable patients with PHI to control their own virus without drugs, at least in the short term. The PRIMSTOP experience suggests that this may even be possible for patients presenting several weeks out from the onset of their acute retroviral symptoms. However, the durability of this type of virologic control as well as the optimum time to initiate HAART and then interrupt it remain critically important and totally unresolved questions. This particular window is still open.
A second reason that PHI may represent a window of opportunity is the hope that patients treated very early might be prevented from establishing the levels of long-lived "reservoirs" of latently infected cells that are felt to be the reason for failure of HAART to eradicate HIV in patients with chronic infection. On Tuesday, Strain and colleagues from UC San Diego presented data suggesting that early treatment may indeed radically reduce the size of these reservoirs (Abstract 97). They studied 29 subjects: 19 "acute" patients (identified before Western blot seroconversion) and compared these to 10 "recently" infected patients who were identified as having been infected within the previous 6 months. Patients were treated with various HAART regimens. Investigators looked at the decay in HIV reservoirs in these patients using both the Roche HIV DNA kit and sophisticated CD8-depleted quantitative microculture assays. Decay was much more rapid in the patients treated prior to seroconversion, but rates of decay in long lived (latent) reservoir slowed for the faster group after one year on therapy. These very preliminary results offer hope that active eradication strategies can be found to mimic the rapid decay in latent reservoirs that can occur in patients treated with PHI.
A third reason that PHI may represent real opportunity is that it is a time of heightened infectiousness, during which identification of infected patients can open up the possibility of preventing secondary infections. The Duke UNC Emory group in collaboration with Swiss colleagues (Abstract 366-M) presented data on paired semen and blood HIV viral load concentrations measured at baseline (prior to initiating HAART) in a group of 30 men with PHI. By displaying these levels as a function of estimated time from infection, investigators showed that viral loads appear to be highest in semen at the precise time that they are highest in blood, and that changes over time in viral loads appear to occur in parallel in the two compartments. They then used a series of calculations to estimate that the likelihood of sexual transmission from a given man to a given woman would be increased about 20-fold during PHI as compared with the same couple having the same sex act 4 months later. These observations suggest that changes in semen viral load are significant during PHI and may put sexual partners at truly high risk-emphasizing the public health opportunity PHI represents.
Previous reports have discussed data on the apparent growth in the scope of drug resistance among people newly infected with HIV. At the 9th Retrovirus conference, Bennett and colleagues from CDC presented new data from 5 US cities involving 1078 patients newly diagnosed with HIV between 1998 and 2000 (Abstract 372-M); of these, 182 were identified as having been infected within 6 months. Prevalence of any drug resistance among these recent infections rose from 10% in '98 to >12% in 2000-numbers that are strikingly similar to the previous findings of Little, et al from last year's conference and reiterated in a number of posters from France, Spain and Australia (369-M, 370-M, 371-M).
Susan Little and colleagues presented some preliminary data in an oral presentation Wednesday (Abstract 95) that transmitted NNRTI resistance persisted in the blood (i.e., did not "revert" to "wild-type" virus) for up to a year in a few of 10 patients found to have been infected with such NNRTI-resistant virus and not going on HAART therapy. Dr. Little concluded that genotyping may actually be useful in detecting such mutations for a year or greater, and so may have a greater role in managing newly diagnosed HAART naive patients for whom the time of infection is not clear.
On the issue of transmitted resistance, Steven Taylor from the UK presented a poster with potentially important implications for this frightening scenario (Abstract 374-M). The presentation detailed a "transmission chain" of three individuals; as each was treated (unsuccessfully), the virus gained resistance at each step. Dr. Taylor examined the genetic relatedness of viruses in semen and blood in these individuals, and found that the viruses in the blood and semen of one patient were quite distinct-potentially reflecting different penetration of these "compartments" by antiretroviral drugs. More striking, he found that the highly resistant viruses isolated from the blood of the next patient in the chain were highly related to the first patient's semen (and not blood) viruses. This presentation highlighted the critical and underappreciated role that drug penetration of the genital fluids must play in the evolution and transmission of drug resistance in the population. Until drugs are widely used that reliably penetrate the genital tract, this type of scenario may be expected to continue to play out tragically for thousands.
HIV prevention efforts hinge on many assumptions about HIV transmission; several groups studying PHI presented findings that bear on these basic assumptions. Julie Overbaugh's group from the University of Washington have conducted a number of studies in a population of commercial sex workers in Mombasa, Kenya that have yielded intriguing evidence that patients may be infected by more than one virus strain. In two oral sessions on Wednesday (Abstracts S23,100), Overbaugh and colleagues presented data on 1400 Kenyan women enrolled in the Mombasa cohort that became HIV infected and had blood drawn at the time of PHI and in followup. They examined the viruses present beginning around month 4 after infection in these women using a laboratory technique called the heteroduplex mobility assay (HMA). HMA can distinguish genetically distinct subpopulations in "swarm" of viruses. Their most striking finding was that 60% of the women studied appeared to be infected by genetically diverse, or heterogeneous populations; in other words, mixtures of viruses with different origins. When they compared viral loads at set point in these women with the 40% that appeared to have been infected by a more homogeneous population, the multiply-infected women had significantly higher and more rapidly increasing virus loads over time. They were unable to see evidence of such multiple infection in a group of recently infected men. The only factor that appeared to be associated with the acquisition of multiple variants in more detailed analysis was hormonal contraception use; Dr. Overbaugh suggested that changes in vaginal biology induced by these contraceptives might provide one explanation for the acquisition of multiple viruses-but the implications of her findings for prevention are not at all clear.
Interestingly, Stephanie Freel and colleagues from North Carolina (Abstract 353-M) found a similar situation among 18 US patients, of whom 16 were men-suggesting that whatever the presence of these diverse variants represents, it may be a fundamental characteristic of HIV infection rather than a happenstance finding in a small population of African women. In an oral presentation Wednesday, Eric Daar (Abstract 96) from UCLA presented evidence that at least one of his patients with PHI had been infected by two completely distinct virus strains-one drug resistant and one drug sensitive. The patient admitted to multiple episodes of risky sex around the time of infection, making it difficult for Dr. Daar to conclude whether this dual infection represented transmission from two separate individuals (super-infection) versus transmission of two variants from one individual (co-infection), although the former certainly seems more likely. All in all, findings from around the 9th conference seem to confirm the long-held suspicion that the infection of one individual with HIV can often involve multiple virus strains. The clinical implications of this are entirely unclear.
In all of this, there may be hope for expanding PHI diagnosis. The North Carolina group presented a study (Abstract 359-M) in which all patients coming in for routine HIV testing in the state over one month and found to be antibody negative were then tested for HIV RNA. HIV RNA testing was actually done first on pools of 90 specimens, reducing the number of tests done to screen the large population. Of the total of 8,341 patients tested, 37 were HIV+ by antibody tests and 5 were HIV RNA+ but negative by antibody tests. 4 of the 5 were found to have true acute infection and 1 was "false positive". The 4 patients with PHI thus represented 10% of all HIV+ patients that came in for testing in NC that month. All of the acute infections identified were in women who were worried about risky exposures and came in for testing at STD clinics. The preliminary cost estimates from the poster suggest that testing using pooling may be highly cost-effective in this setting. If such strategies could be expanded, it is possible that the "window of opportunity" for better HIV treatment and prevention could get significantly wider.