The impact of unfit virus

Nelfinavir (Viracept)-resistant HIV is significantly slower to replicate than HIV that has never been exposed to drugs before. It is also slower to replicate than virus that has developed resistance to other protease inhibitors (PIs) such as saquinavir (Invirase / Fortovase) and indinavir (Crixivan).

Dr Richard D'Aquila and colleagues tested viruses with specific mutations known to be associated with drug resistance to see how quickly they replicated in the test tube. They found that virus with the D30N mutation, which is usually the first sign of nelfinavir resistance, grew 30% more slowly than virus not exposed to any drugs (Martinez-Picado 1999).

Two teams at the Chicago Retroviruses and Opportunistic Infections Conference reported similar findings in February 1999. A joint Franco-American team reported that virus highly resistant to saquinavir did not cause T-cell depletion in a mouse model after 28 days, despite low-level viral replication.

Another French team reported on the loss of viral fitness in 14 people receiving PIs who experienced viral rebound. During 20 months median follow-up, viral fitness declined by 28% despite an average viral load increase of 1.4 log₁₀. The degree of decline in viral fitness showed a strong inverse association with CD4 cell count increase; in other words, the less fit the virus after 20 months, the bigger the CD4 cell count increase. There was no association between viral load and the size of the CD4 cell count increase, so in two individuals with the same viral load, the only factor which was significantly associated with the level of the CD4 cell count was the fitness of the virus, not the level of the viral load (Faye 1999; Stoddart 2001).

Reduced replicative capacity has been clearly linked to long-term CD4 cell count improvements despite detectable viral load. A study of 17 patients with a mean of 34 months of detectable viral load after failure of a PI-containing regimen showed that an initial CD4 cell count gain of approximately 150 cells/mm³ was sustained, whilst replicative capacity declined by an average of 8% (Goetz 2003).

Another study, of 30 patients who had a CD4 cell count below 50 cells/mm³ or viral load above 50,000 copies/ml prior to commencing antiretroviral therapy, found that those with a discordant viral load and CD4 cell count response were more likely to have a reduced replicative capacity (Sufka 2003).

These findings suggest that viral load levels in people with less fit virus might not give an accurate prediction of the long-term risk of CD4 cell count decline. This theory is borne out by several large studies of people on antiretroviral therapy, which show that a significant proportion of people who experience viral rebound will maintain high CD4 cell counts for long periods, despite rising viral load. See Discordant CD4 and viral load responses in Anti-HIV therapy: Restoring the immune system for more details of these studies.

Clearly, treatment continues to benefit people by limiting the ability of HIV to replicate and thus boosting CD4 cell counts (Deeks 2001). American researchers from the University of California, San Francisco, have reported that replicative capacity slowly declines as drug resistance increases in people with long-term virologic failure who remain on treatment (Barbour 2002).

These studies also suggest that some protease mutations make HIV less harmful to CD4 T-cells even when viral load is rising. By reducing the pathogenicity of HIV, PI therapy may preserve CD4 T-cell populations. However, Faye and colleagues warned that eventually, if CD4 T-cell numbers kept rising and drug pressure was removed, wild type virus would come back very quickly because of the large number of target CD4 T-cells available for infection.

Timing of rebound and fitness

The San Francisco research team has found that the timing of viral rebound may be critical in determining whether individuals have long term CD4 cell count increases. Steven Deeks and colleagues studied 119 individuals who experienced virologic rebound after an initial period of undetectable viral load. They found that those who rebounded after less than three months had greater subsequent increases in CD4 cell counts, and were less likely to have experienced a viral load rebound back to baseline than those who experienced viral load rebound more than three months after starting treatment. The researchers argue that individuals who fail quickly may be experiencing the rapid selection of pre-existing viral mutants which are less fit, whilst those who fail after a lengthier period of viral suppression may be experiencing the selection of virus which is more fit and more likely to affect CD4 T-cell survival and production (Deeks 1999).

The same group found that for any given level of viral load, the turnover of CD4 T-cells and CD4 T-cell activation was lower among patients with drug-resistant virus when compared with patients who had wild type virus. This difference was seen even in patients with viral load above 30,000 copies/ml. Patients with drug-resistant virus had been on treatment for more than two years since virological rebound at the time of sampling, and had experienced a median CD4 cell increase of 109 cells/mm³ since starting PI therapy. There was little difference in rates of CD4 T-cell turnover between those with detectable drug-resistant viremia and those with undetectable viral load. The researchers suggested that immune reconstitution is a function of decreased CD4 T-cell turnover rather than increased production, and that drug-resistant virus is likely to maintain decreased CD4 T-cell turnover both by reduced infection of thymus tissue, hence increasing production of naive cells, and by reduced replicative capacity in thymic tissue (Deeks 2002).

But is the decline in fitness a short-term phenomenon? It has been suggested that mutations gradually occur which improve fitness. A study presented in early 2000 studied the replicative capacity of laboratory-derived mutant virus and patient-derived mutant virus. With patient-derived virus, a greater number of resistance mutations correlated with a decline in fitness. However, the lab-derived virus had fitness restored by subsequent mutations such as L63P, V77I and Gag cleavage site mutations (Wrin 2000). This study demonstrates that mutant virus does not replicate as efficiently as wild-type virus, although the extent to which subsequent mutations may restore replicative capacity remains unclear. As noted above, long-term follow-up of 20 people on failing therapy has led the Steve Deeks' team to suggest that HIV has a limited ability to become both highly resistant and highly fit (Barbour 2002).

Measuring fitness

An assay to measure replicative capacity or fitness has now been developed by Virologic, although studies are still ongoing to evaluate its usefulness in clinical practice.

Individuals recently infected with drug-resistant virus were found to have higher CD4 cell counts in a study of 130 drug-naive seroconverters identified in San Francisco, and those with PI resistance mutations had significantly lower viral replication capacity (Grant 2002).

A study of 28 individuals on failing therapy found that resistance was associated with R5 tropism, and that individuals with R5 strains had significantly lower replicative capacity and significantly higher CD4 cell counts. The lowest replicative capacity was detected in viruses with greater than five resistance mutations, whilst the K103N and Y181C mutations associated with non-nucleoside reverse transcriptase inhibitor (NNRTI) resistance were associated with high replicative capacity. Those with replicative capacity close to wild type invariably had viral loads above 30,000 copies/ml (Andreoni 2002).

Persistent low-level viral load without rebound is associated with severely reduced replicative capacity according to a study of 37 patients. Individuals with viral load that remained below 1000 copies/ml at two separate and consecutive visits to the clinic, four months apart, had virus with an average replicative capacity of just 9% of that of wild type virus. Four of the 23 subsequently experienced rebound above 1000 copies/ml (Karlsson 2002).

Another study of patients on failing PI therapy found that those with virus showing replicative capacity of 10% or less had an average CD4 cell count 82 cells/mm³ higher than those with virus with a replicative capacity equivalent to wild type (Haubrich 2002). Replicative capacity at baseline also predicted the likelihood of viral suppression after six and twelve months of a salvage regimen, and of maintaining a higher CD4 cell count.

Reduced fitness has also been reported in viruses resistant to T-20 (Lu 2002).

Taken together these studies provide some evidence to support the view that a leading factor contributing towards stable CD4 cell levels in people with treatment failure is reduced replicative capacity. Further research will be needed to define whether a measurement of replicative capacity in people with low level but detectable viral load is a useful additional test for monitoring when to change treatment.

Further research will also be needed to define whether reduced replicative capacity is maintained when the selective pressure of particular agents is removed, such as after changing treatment.

Fitness and infectiousness

A decline in viral fitness due to drug resistance does not appear to reduce the ability of HIV to infect certain types of cell in the laboratory, but a recent study comparing the prevalence of resistant virus with the number of seroconverters with resistant virus suggested that resistant virus may be less infectious (Leigh Brown 2000).

It is speculated that reduced fitness could restrict the types of immune cells which can be infected, with important knock-on consequences for CD4 T-cell production. Although drug-resistant viruses isolated from individuals with discordant viral load and CD4 cell count responses do not appear to be impaired in their ability to infect peripheral CD4 T-cells, it is possible that their ability to infect more productive precursors, such as thymic T-cells, may have lessened (Clavel 1999; Liegler 2001).

Is 3TC resistance beneficial?

Another drug-associated mutation which has been shown to reduce replicative fitness is the M184V mutation seen with 3TC (lamivudine, Epivir) treatment. A European study called Colate is now testing whether it is beneficial to continue with 3TC treatment after viral rebound on a 3TC-containing combination, or whether it is better to drop 3TC at the same time as changing all the other drugs in the combination.

Non-nucleoside reverse transcriptase inhibitor resistance and fitness

Test tube research suggests that virus with NNRTI-associated mutations is also less fit. When cultured in the absence of drug, wild-type virus grew faster than the mutant virus. Importantly, when cultured alone, virus with the K103N mutation did not return to wild type (Paul 1999).

Continuing therapy with an NNRTI when mutations are present is not currently recommended as a salvage strategy, as further NNRTI mutations may be selected, damaging a person's chance of responding to other NNRTIs in the pipeline.

Multi-drug resistance and fitness

It is usually assumed that a virus that is resistant to many drugs will replicate poorly. Test tube data from a team from the National Institutes of Health in the United States suggest that sometimes this is the case. However, sometimes the mutant virus replicates as well as the wild-type virus in the absence of drug pressure (Kosalaraksa 1999).

Implications for current practice

These findings highlight the discordance between virological and immunological responses to antiretroviral therapy that has been widely reported. They also raise questions about which events should trigger changes in therapy, and how that change should be managed.

Some experts, such as Steven Deeks, have proposed that the best way to manage patients with multi-drug resistance and virological failure is to be guided by the CD4 cell count when making decisions about switching treatment, rather than risk the toxicity of a mega-HAART regimen. Thus, those who maintain CD4 cell counts above the pre-treatment level, or above 200 cells/mm³, or who experience slower than expected CD4 cell declines, should stay on their current regimen if it is well tolerated, on the grounds that it is probably maintaining reduced replicative capacity and is thus less damaging to the immune system than wild type virus.

See Discordant CD4 cell count and viral load responses in Anti-HIV therapy: Restoring the immune system and Anti-HIV therapy: Changing treatment for further details.

Replicative capacity and immunologic response

Replicative capacity refers to the ability of a virus to replicate compared to a virus regarded by the manufacturers of the replication capacity assay as having 'average' replication capacity. Certain drug resistance mutations have been shown to reduce replicative capacity, and low replicative capacity has been correlated with a slower rate of CD4 T-cell loss, or stable CD4 cell counts, in individuals who have experienced viral rebound on treatment.

References

Andreoni M et al. Biological phenotype, replicative capacity and drug resistance of HIV strains isolated from patients failing antiretroviral therapy. Antivir Ther 7: S45, 2002.

Barbour JD et al. Evolution of phenotypic drug susceptibility and viral replication capacity during long-term virologic failure of protease inhibitor therapy in human immunodeficiency virus-infected adults. J Virol 76: 11104-11112, 2002.

Clavel F et al. Can resistance-associated loss of viral fitness explain discordant CD4 and plasma HIV RNA evolution following protease inhibitor failure? Antivir Ther 4: S91, 1999.

Deeks SG et al. Shorter length of viral suppression prior to virologic failure is associated with greater post-failure CD4 accumulation: implications for increased target cell availability and/or reduced viral fitness in virological failure. Antivir Ther 4: S93, 1999.

Deeks SG et al. Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia. N Engl J Med 344: 472-480, 2001.

Deeks SG et al. CD4+ T cell kinetics and activation in human immunodeficiency virus-infected patients who remain viremic despite long-term treatment with protease inhibitor-based therapy. J Infect Dis 185: 315-323, 2002.

Faye A et al. Viral fitness in patients with discordant CD4 and plasma HIV RNA evolution following protease inhibitor failure. Sixth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 331, 1999.

Goetz MB et al. Long term durability of impaired HIV replication capacity in patients maintained on a stable treatment regimen despite low-level virological failure. Second International AIDS Society Conference on HIV Treatment and Pathogenesis, Paris, abstract 271, 2003.

Grant RM et al. Transmission of drug resistant HIV-1 exhibiting lower replication capacity is associated with higher CD4 cell counts. Antivir Ther 7: S41, 2002.

Haubrich R et al. Replication capacity as a predictor of immunological and virological benefit despite virological failure of an antiretroviral regimen. Antivir Ther 7: S101, 2002.

Karlsson AC et al. HIV specific immunity, T cell activation and replicative capacity in antiretroviral treated adults experiencing intermittent versus low-level viral replication. Antivir Ther 7: S43, 2002.

Kosaloaraksa P et al. Comparative fitness of multi-dideoxynucleoside-resistant human immunodeficiency virus type 1 (HIV-1) in an in vitro competitive HIV-1 replication assay. J Virol 73: 5356-5363, 1999.

Kuritzkes DR et al. Continued lamivudine versus delavirdine in combination with indinavir and zidovudine or stavudine in lamivudine-experienced patients: results of Adult AIDS Clinical Trials Group protocol 370. AIDS 14: 1553-1561, 2000.

Leigh-Brown AJ et al. Transmission and fitness of drug resistant strains of HIV. Fifth International Congress on Drug Therapy in HIV Infection, Glasgow, abstract PL9.6, 2000.

Liegler T et al. Phenotypic characterization of HIV-1 from virologic failures with continued CD4 T cell accumulation following HAART. Antivir Ther 4: S94, 1999.

Liegler T et al. Protease inhibitor-resistant HIV-1 from patients with preserved CD4 cell counts is cytopathic in activated CD4 T lymphocytes. AIDS 15: 179-184, 2001.

Lu J et al. Fitness of HIV-1 clinical isolates resistant to T-20 (enfuvirtide). Antivir Ther 7: S56, 2002.

Martinez-Picado J et al. Replicative fitness of protease inhibitor resistant mutants of human immunodeficiency virus type 1. J Virol 73: 3744-3752, 1999.

Paul MA et al. Relative fitness of efavirenz (EFV, DMP 266)- resistant mutants of HIV-1. 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, abstract 438, 1999.

Stoddart C et al. Impaired replication of protease inhibitor-resistant HIV-1 in human thymus. Nat Med 7: 712-718, 2001.

Sufka SA et al. Prolonged CD4+ cell/virus load discordance during treatment with protease inhibitor-based highly active antiretroviral therapy: immune response and viral control. J Infect Dis 187: 1027-37, 2003.

Wrin T et al. Measuring the replicative fitness of recombinant HIV-1 vectors expressing protease and reverse transcriptase derived from patient viruses. Seventh Conference on Retroviruses and Opportunistic Infections, San Francisco, abstract 233, 2000.

Youle M et al. Hydroxyurea in a late stage salvage study. Second Meeting of the Research Institute for Genetic and Human Therapy (RIGHT), Washington, 1999.