Primary HIV Infection: Resistance, Treatment and Transmission issues

     Reported for NATAP by Christopher D. Pilcher, M.D. of the UNC Center for AIDS Research, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

What is it and why is it important to "the rest of us"?

Primary HIV infection (sometimes abbreviated PHI) is what takes place immediately following initial infection by HIV. Although many individuals are asymptomatic throughout their first months and years of infection, probably a majority of patients develop some symptoms within 2-4 weeks following their exposure to the virus-a mononucleosis-like illness also called the "acute retroviral syndrome". When present, symptoms are variable but are often characterized by fever, headache, sore throat, swollen glands, body aches and/or general malaise. Oral or genital ulcers are frequently present. These symptoms typically resolve within days to weeks, and when symptomatic individuals do present for evaluation their complaints are often attributed Epstein-Barr Virus infection or other common viral illnesses. Even if clinicians are alert to the possibility of HIV infection, diagnosis can be difficult because the antibodies that are detected by routine HIV tests (ELISA or confirmatory Western blot) are only just beginning to be formed during primary HIV infection and these tests come back negative. By contrast, viral loads in PHI are usually extremely high-so less routine (and more expensive) HIV RNA or p24 antigen tests are the only reliable means of detecting infection in this setting.

By focusing on patients with PHI---by definition, new HIV infections---researchers can get answers to some important but otherwise unanswerable questions. These include: How many new infections are being caused by drug resistant viruses in different parts of the world? Are people exceptionally infectious during PHI because of their very high viral loads? Moreover, are there any ways to intervene during PHI that can potentially alter patients' long-term prognosis if they are identified? We have already discussed this last point with regard to STIs, but this retrovirus conference has provided more new data relevant to all of these issues.

Infection with Resistant Viruses: Why is there now more in the US but less in Switzerland?

One can only find out how much resistant virus is getting transmitted by looking at patients with primary infection because resistance mutations can mutate back to drug-sensitive after a few months when patients aren't treated and the virus is allowed to evolve in the body. If one waits to test for resistance until people were diagnosed with chronic HIV infection years later, most people living with HIV appear to have drug-sensitive virus. Of course, once in, resistance mutations will always be present under the surface in a small number of viruses and can always rear their ugly heads again once patients begin drug therapy.

Worrisome Trends in Transmission of Drug Resistant Virus

Most studies have historically used genotype alone (i.e., the specific drug resistance mutations in the virus) to define resistance in patients with PHI, but some are now looking at phenotypic resistance as well (where viruses are actually tested for susceptibility to inhibition by different drugs. Susan Little from San Diego [Abstract 756] reported a hugely important study that surveyed phenotypic resistance (defined as a >10 fold decrease in susceptibility to any drug) in 408 newly infected patients in North American cities (Birmingham, Dallas, Denver, Los Angeles, San Diego, Seattle, Montreal and Vancouver) and found an overall prevalence of 8% resistance, with 4% being resistant to two classes. What was more worrisome, they noted that the overall rate of resistance has increased markedly between the 1995-8 period and 1999-2000 (from 3 to 14%; 1 to 6% for NNRTIs and 1 to 7% for PIs). Little and colleagues were able to pool data from enough patients to show that the presence of this resistance (>10 fold to any ARV) did increase the risk of virologic failure on therapy for all comers. Although most clinicians do not routinely take these type of data into account except when dealing with patients that have primary infection, it is critical to remember that burgeoning resistance today could have significant implications for success of antiretroviral therapy for chronically infected individuals when they are diagnosed years from now-since their bodies will still harbor resistant viruses from the time of their infection today. The issue is clearly not going to go away in the years to come, as more and more people living with HIV and potentially transmitting virus begin to fail more and more regimens with more and more drugs.

(Editorial notes: In the program abstract, Little suggests that NRTI resistance prevalence has not significantly changed. There is a little data suggesting response to therapy may not be as good when transmitted resistant virus is present).

Geographic differences in transmission of resistant viruses can be dramatic, however. Chaix et al from Paris [Abstract 755] reported similar rates to the US estimates with a 10% prevalence of overall genotypic resistance. In contrast, the resistance situation is apparently rapidly changing in Switzerland: Luc Perrin from Geneva [Abstract 754] explained a decrease in the proportion of new infections with drug resistant viruses in his country (from 15% in 1997 to a low of 5% in 1999) in part by an increasing number of new infections with "non-subtype B" viruses (i.e., viruses with recent origins outside of Europe, mainly from Africa) and in part by the increased effectiveness of antiretroviral therapy at reducing viral loads in Swiss patients between 1997 and 1999.

Transmission Issues-Does PHI Drive the Epidemic?

There has been a small cadre of epidemiologists who have long insisted that the only way to explain the explosive growth of the HIV epidemic was that HIV must travel quickly from one person to another-implying that the highest sexual infectivity must occur during the time of primary infection. There are many other infectious diseases, from measles to chickenpox, that operate exactly that way, producing lots of virus shedding into body fluids right at the time of earliest infection to maximize their chance of spread. The stakes of these types of assertions about hyper-infectiousness in PHI are enormous, however. They might imply that public health HIV prevention efforts should be radically expanded to target early diagnosis of PHI and urgent tracing of sexual contacts of newly infected patients with PHI. It might also provide a rationale for antiretroviral treatment in PHI, since we know that ART reduces shedding of virus into semen and vaginal fluids during this period. To the chagrin of the proponents of the hyper-infectiouness theory, has been no clear evidence of the importance of transmission specifically on the part of individuals with PHI. In fact, such transmission had never actually been documented at all before this Conference.

Sexual transmission was shown to occur in 5 cases in a poster on Tuesday presented by myself but representing cases from both the US and Switzerland [Abstract 411]-in three cases, transmission occurred prior to the onset of symptoms in the transmitter. As to the importance of this phenomenon, Luc Perrin from Geneva gave some preliminary indications in a presentation on Wednesday afternoon [Abstract 754] when he presented an analysis looking at the genetic sequences of the viruses from 197 individuals-all of the patients identified with PHI in Switzerland over the past 4 years. By phylogenetic analysis (methods constructing "evolutionary trees" reflecting genetic relatedness of different strains) he showed that 29% of the viruses "clustered" together with at least one other virus, i.e., 29% of the infected individuals shared basically identical viruses with one of the others in this cohort of PHI patients. They went on to show that by tracing sexual and injection drug use contacts they could definitively establish transmission from one person with PHI to a partner in a third of those patients (or 9% of the total number). Because of the notorious difficulty in establishing such connections, the true magnitude of the public health hazard represented by newly infected, completely unaware individuals passing on HIV will still ultimately have to be guessed at.

Previous data from several groups have shown that semen shedding is established quite early in PHI, and by some reports at significantly higher levels than later in chronic infection. Perhaps as important as any biological explanation for increased transmission on the part of people with PHI, however, is the fact that they tend to be currently engaged in activities that unwittingly put others at risk. In a study of 63 patients with PHI referred to the LA County Health Department reported in a poster on Monday [Abstract 216], 75% of 59 men surveyed responded that they had used some public sex venue at least once in the previous 6 months (bathhouses, group sex parties, etc.) Such figures should remind us that in 2001, safe sex messages are clearly not reaching the young people that are at the highest risk of acquiring HIV infection.

Treatment for PHI? Antiretroviral and adjunctive therapies

It is truly not known how best (or even if) to treat people with PHI. With viral loads and CD4 counts in dramatic flux during the period of PHI, there is little basis for predicting how individual patients with PHI will do once the period of PHI has passed. HAART, on the other hand, is a known quantity. Long-term toxicities, expense and effects on quality of life are all part of the picture. So, HAART treatment during PHI can only be justified if it is aimed at improving long-term prognosis. In his "State-of-the-art Lecture" on STIs Tuesday, Bruce Walker from Harvard discussed the preliminary success of STI protocols in the Mass General Hosp/Brigham/Fenway trial at bringing a small series of patients under immunologic control after therapy. However this success has not been reproduced in other trials conducted at different centers, with different regimens or with patients that differ only slightly from the MGH group. (see separate report on STIs for a summary). Walker and others emphasized repeatedly that while this approach is extremely important to explore, the situation is still unclear and there is no indication for anyone with PHI to be treated outside of research protocols.

Immunotherapies in PHI: hydroxyurea; cyclosporin

Several groups noted treatment success with both PI based and PI sparing HAART regimens in a variety of poster sessions. The role of hydroxyurea in PHI treatment was called into question by two presentations showing adverse effects of HU on CD8 anti-HIV immune responses [Abstracts 356 and 408]. These presentations basically contradict previous results from Franco Lori's group, and the reasons for the discrepancy are totally unclear.

(Edit comments from Jules Levin: in Abstract 408, 13 subjects received non-PI HAART and 1 person did not receive HU during PHI. They were followed for about 1+ years. Study investigators reported after discontinuing HU immune responses markedly improved: "following discontinuation. of HU in most individuals by week 16-24, a more significant rise in T-cell subsets could be appreciated......Minimal to undetectable HIV-1-p24-specific proliferative responses were present in those subjects receiving HU, whereas strong reactivity was readily detectable in one subject not receiving HU.....Following cessation of HU, the majority of individuals exhibited significant HIV-p24-specific proliferative responses".

Abstract 356 appears to be 52-week follow-up for 32-week data from the same study reported at last Summer's Resistance Workshop. 20 chronically infected receiving d4T+3TC+IDV for 52 weeks & <20 copies/ml for at least 32 weeks were randomized to receive d4T/ddI/IDV with or without HU. The Retrovirus program abstract reports after 4 cycles of STI relatively similar CTL responses were seen in both groups, and after the 5th STI 1/6 in the ART group and 5/5 in the HU group maintained CD4 HIV-1-specific response.

Although I did not read the entire poster the conclusions by the authors said HU alone was used during the STIs starting after the third STI and HU helped keep viral replication down. I don't think this has any clinical implications, but it may spur further research. At the Resistance Workshop, Felipe Garcia reported on 3 STIs and said "rebound in plasma viral load was detected in all cases" after each stop in therapy, and 5/9 in both HAART and HU groups had strong CTL response and CD4 lymphocyte proliferative response to Hiv- antigens after the 2nd & third stops (which was not present at baseline or after the first stop).

Garcia concluded at Resistance Workshop that HU associated with STI did not prevent viral load rebound or viral load dynamics after interruption. As well, HU was not effective in increasing control of viral replication after 3 STIs nor in inducing a higher proportion of specific immune responses against HIV-1 antigens).

One particularly interesting talk came from Rizzardi from Italy [Abstract 759], who reported on a single arm pilot study using HAART with the addition of cyclosporin A (a drug used commonly in transplant patients for immunosuppression, to prevent rejection) for the first 8 weeks. The drug was reported vaguely to be "well tolerated" over this short course of treatment. They compared the virologic responses and CD4 recovery of their patients to a "control group" of 22 patients from a similar but separate trial of HAART in PHI in Italy. Despite the apparent lack of rigor in their approach, they noted that the CD4 rises in their group were dramatically higher (1443 vs 712 cells/cu.mm at week 64) than in the HAART alone study. Based on limited data that CD4 rises did not correlate with CD4 proliferation in their subjects (which would have implied new cell production as the source of the increase), they postulated that the cyclosporin might be blocking the trapping of CD4 cells in the lymph nodes during PHI. The observations of this pilot study clearly warrant a randomized trial, and if confirmed might prove important both as an adjunct to treatment and in providing clues to immunopathogenesis in PHI.