• High Replication Capacity Is Associated With High Baseline Viral Load in Untreated Subjects With Primary HIV Infection (Oral 152)
  Authored by S.J. Little, S.D.W. Frost, J.P. Routy, A.C. Collier, J.B. Margolick, E.S. Daar, R.A. Koup, B. Conway, L. Wang, T. Wrin, C.J. Petropoulos, N.S. Hellmann, D.D. Richman, S. Holte
  View the original abstract

Dr. Susan Little opened her presentation by changing the title of her abstract to "High Replication Capacity Is NOT Associated With High Baseline Viral Load in Untreated Subjects With Primary HIV Infection." This is always the danger when submitting abstracts before all the data is available and you wind up with a different conclusion on the final analysis compared with the interim results.

Unfortunately I was unable to see her presentation as I was covering the metabolic abstracts, but I did ask a colleague (who just happens to be my husband, P.T. Cohen, M.D., Ph.D.) to take notes. Dr. Little graciously sent me her slide presentation and offered to answer any questions by e-mail. The purpose of this study was to evaluate whether drug susceptibility and viral replication capacity (RC, e.g., fitness) are associated with baseline plasma viral load levels (VL) in antiretroviral-therapy (ART)-naive subjects with primary HIV infection. The researchers also wanted to characterize the natural history of RC among ART-naive subjects and to assess the relative virologic set point among ART-naive subjects relative to drug susceptibility, VL and RC.

I have discussed RC in a previous review of an abstract presented at the IDSA meeting last fall. RC has been purported to be a surrogate marker for viral fitness. It would be difficult to measure the actual RC because it is dependent on an individual's immune response and exposure to antiretrovirals, which cannot be mimicked in a laboratory test. Dr. Little emphasized this point as well in her talk. RC does not equal fitness.

Nevertheless, some providers have been using the RC assay results in order to continue an antiretroviral regimen that is "failing" in the setting of an unfit virus. Another purported use is to measure the RC in ART-naive chronically infected individuals to gauge HIV progression. If the RC were low, then this possibly would be associated with a slower progression. Further studies are warranted exploring the clinical utility of RC. Dr. Little and colleagues sought to define the clinical utility of this assay in primary infection.

They evaluated baseline plasma viral load (VL), drug susceptibility, RC, and pol gene sequence analysis for 516 ART-naive subjects. Drug resistance results have been previously reported in part for these subjects. Drug susceptibility and RC were measured using PhenoSense HIV. ABI sequence analysis of pol was used to identify drug resistant mutations. Virologic set point was estimated for a subset of 194 subjects who remained ART naive for a period of five-24 months after the estimated date of HIV infection. Of note, subjects with higher baseline VL are more likely to have begun therapy, therefore this subset most likely represents those with lower baseline VL. Longitudinal phenotype and RC data were available (median 28 months) for a subset of nine untreated subjects.

Baseline characteristics for the 322 excluded subjects and the 194 set-point subjects were similar with respect to age (mean 35 years), sex (91 percent men) and white, non-Hispanic race (75 percent). Patients included in the set-point analysis presented to care later after primary infection and with significantly lower baseline viral loads. The mean time from estimated date of HIV infection to ART was 78 days in the excluded subjects (VL 4.94 log10) compared with 340 days in the set-point subjects (VL 4.50 log10). The baseline CD4 T-cell counts and RC were similar between the two groups. The prevalence of transmitted drug resistance was equivalent among the set-point group and the excluded subjects with approximately 10 percent demonstrating phenotypic resistance. Rates of genotypic resistance were similar to those reported by Dr. Little and colleagues in previous publications.

Dr. Little and colleagues followed a subset of nine patients who remained ART naive for 11-40 months (median 28 months). They noted a significant correlation between protease inhibitor (PI) hypersusceptibility (HS) (subject IC50/reference IC50 of <0.4) and low RC. The mean RC was 30 percent for subjects with HS to any one or more PI compared with RC 50 percent for PI susceptible (p<0.0001). PI HS is more prevalent (33 percent) in primary infection than in established infection (18 percent) among subjects infected with a drug sensitive virus.

The mean RC in primary infection is significantly lower than in established infection and that low RC virus of primary infection may evolve to higher RC virus of established infection as a result of host immune selective pressure (see abstract 549, Brown et al.). Dr. Little and colleagues postulated that subjects infected with drug-resistant HIV would have a lower virologic set point than patients infected with a drug-sensitive virus. They reported a 0.34 log10 VL increase among those subjects with all class resistance, 0.18 log10 VL increase among subjects with NRTI resistance, 0.61 log10 VL increase among subjects with NNRTI resistance and 0.19 log10 VL increase among subjects with PI resistance compared with the fully susceptible subjects for whom the VL remained stable over the course of follow up.

The investigators concluded that higher baseline RC is not associated with higher baseline VL and that the mean RC is lower and remains stable among subjects with primary infection compared to those with established infection. The set point VL is higher among subjects infected with drug resistant virus, despite comparable baseline VL and appears to be most linked to transmitted NNRTI resistance. They estimated that the median time to disease progression in this set-point cohort is likely >10 years. The clinical significance of RC needs further study (see editorial comments below). The question remains whether one could use RC in a newly diagnosed HIV seropositive patient (whether acute or chronically infected) to assist in identifying patients who should begin ART immediately or defer until evidence of immunologic decline particularly among patients whose CD4 T-cells linger in the 300-500 range.

An issue I have difficulty understanding is if the RC is lower and remains stable among subjects with primary infection compared with those who have "established" infection, then how come the RC is increased in established infection? At some point, it cannot remain stable and it has to increase. When, why? Also, when does "primary" infection become "chronic" or "established" infection? The subjects above were followed a median of 28 months (range 11-40 months). I would consider those subjects to have developed chronic infection. Further studies are needed to follow RC from the time of acute HIV seroconversion longitudinally. Whether the RC value can assist with timing of ART when one sees an increase in RC remains to be determined.