Antiretroviral Susceptibility of Non-B HIV-1 Subtypes
Antiretroviral Susceptibility of Non-B HIV-1 Subtypes
I am treating a husband and wife who both have clade AG HIV-1 infection. Both started antiretroviral therapy (ART) 1 year ago this month. The husband started on ritonavir-boosted lopinavir, and later switched to ritonavir-boosted atazanavir plus abacavir/lamivudine. The wife started directly on the ritonavir-boosted atazanavir regimen. After 1 year, HIV RNA remains detectable in both. They both started with HIV RNA > 100,000 copies/mL, had a slow decline, and have spent the last 6 months with HIV RNA in the 200-1000 copies/mL range. T cells have come up nicely. I do believe that they're both very adherent.
At the 6-month point he was still at a viral load of 1200 copies/mL. Sensitivity testing and genotyping with the PhenoSense GT assay showed that his virus had no mutations and that the phenotype was actually hypersusceptible to most drugs. I don't understand why the virus is still not undetectable, and I'm bothered by it. I might infer that their clade is less susceptible, but the phenotype certainly doesn't support that. Any thoughts?
Graeme J. Moyle, MD, MBBS
Medical Director, St. Stephen's HIV Research Centre, Chelsea and Westminster Hospital, London, United Kingdom
It is often difficult to draw conclusions regarding susceptibility of different viral clades to highly active antiretroviral therapy (HAART), and studies have produced conflicting results. Clade B is, or course, the most common HIV-1 subtype in North America and Europe, and a major criticism of available data on non-subtype B viruses is that datasets are too small, specifically with regard to less common clades such as AG.
Group O viruses (as with HIV-2) are thought to be naturally resistant to nonnucleoside reverse transcriptase inhibitors (NNRTIs), and group N viruses have not been phenotypically studied in detail. For the M group, data have suggested that protease inhibitor (PI) mutations/polymorphisms are more common at baseline in non-B subtypes vs B subtypes, with a decreased response to HAART. In a Zambian study examining the prevalence of drug resistance and evolutionary changes in sequence in HIV-1 subtype C viruses (n = 28), only secondary but no primary drug-resistant mutations were observed. In an Italian study, preexisting protease mutations at codons 10 and 36, present in 39.3% and 40% of individuals, respectively, in whom treatment had subsequently failed, were the strongest predictors of virologic rebound (n = 248; P = .004). Palmer and coworkers determined the susceptibility of subtypes A to E to zidovudine, lamivudine, didanosine, nevirapine, and ritonavir. All subtypes tested had similar susceptibilities, apart from D subtypes, which had higher levels of resistance to all classes. Further studies have shown no differences in subtype responses to tenofovir, adefovir, and zidovudine, with no differences in responses to reverse transcriptase inhibitors among subtypes A through to F. Recent data that evaluated susceptibility of the Ghanaian subtype CRF02_AG strains suggested that they may be less susceptible to nelfinavir than are subtype B strains. Of note, specific data addressing viral subtype susceptibility to atazanavir have not been reported.
Two UK studies have failed to demonstrate that the presence of particular subtypes, individual polymorphisms, and mutation combinations are associated with either a decreased response to HAART or a worse prognosis. In the first of these studies, the relationship between viral subtype, disease progression, and response to HAART was examined (N = 867). For B (n = 457) and non-B subtypes (n = 410), there was a similar rate of pretreatment CD4 decline and similar time to viral undetectability. Within the "non-B" subtype category, 19% were A; 36% were C; 9% were D, and 4.4% were infected with recombinant strains (the most common being CRF02_AG). A study at Imperial College sequenced the reverse transcriptase and protease genes in African immigrants (n = 79) and showed that the non-B subtypes that were present in the study population were all similarly clinically susceptible to HAART. A similar study in therapy-naive individuals (n = 479) demonstrated that although baseline CD4+ cell counts were lower in individuals harboring the non-B subtypes (P = .02), there were no significant differences in virologic response up to 18 months.
A further consideration involves assays for detection of resistance. Subtype B isolates appear to be sequenced with a greater degree of success than non-B isolates. Because the pattern of drug resistance mutations may differ between subtypes, one recombinant virus phenotypic study (n = 32) has shown that data from drug-resistant B subtypes may not always be applicable to non-B subtypes.
It is not certain what to do with the individuals you describe, because they should have completed their antiviral responses by now. Ongoing viral replication clearly risks emergence of the reverse transcriptase resistance mutation M184V and potentially other nucleoside reverse transcriptase inhibitor resistance mutations (NRTI) as well as PI resistance mutations. I think it is worth considering modification of the regimen to an alternative NRTI and ritonavir-boosted PI combination to evaluate whether this leads to a more complete virologic response.
Question
I am treating a husband and wife who both have clade AG HIV-1 infection. Both started antiretroviral therapy (ART) 1 year ago this month. The husband started on ritonavir-boosted lopinavir, and later switched to ritonavir-boosted atazanavir plus abacavir/lamivudine. The wife started directly on the ritonavir-boosted atazanavir regimen. After 1 year, HIV RNA remains detectable in both. They both started with HIV RNA > 100,000 copies/mL, had a slow decline, and have spent the last 6 months with HIV RNA in the 200-1000 copies/mL range. T cells have come up nicely. I do believe that they're both very adherent.
At the 6-month point he was still at a viral load of 1200 copies/mL. Sensitivity testing and genotyping with the PhenoSense GT assay showed that his virus had no mutations and that the phenotype was actually hypersusceptible to most drugs. I don't understand why the virus is still not undetectable, and I'm bothered by it. I might infer that their clade is less susceptible, but the phenotype certainly doesn't support that. Any thoughts?
Response From Expert
Graeme J. Moyle, MD, MBBS
Medical Director, St. Stephen's HIV Research Centre, Chelsea and Westminster Hospital, London, United Kingdom
It is often difficult to draw conclusions regarding susceptibility of different viral clades to highly active antiretroviral therapy (HAART), and studies have produced conflicting results. Clade B is, or course, the most common HIV-1 subtype in North America and Europe, and a major criticism of available data on non-subtype B viruses is that datasets are too small, specifically with regard to less common clades such as AG.
Group O viruses (as with HIV-2) are thought to be naturally resistant to nonnucleoside reverse transcriptase inhibitors (NNRTIs), and group N viruses have not been phenotypically studied in detail. For the M group, data have suggested that protease inhibitor (PI) mutations/polymorphisms are more common at baseline in non-B subtypes vs B subtypes, with a decreased response to HAART. In a Zambian study examining the prevalence of drug resistance and evolutionary changes in sequence in HIV-1 subtype C viruses (n = 28), only secondary but no primary drug-resistant mutations were observed. In an Italian study, preexisting protease mutations at codons 10 and 36, present in 39.3% and 40% of individuals, respectively, in whom treatment had subsequently failed, were the strongest predictors of virologic rebound (n = 248; P = .004). Palmer and coworkers determined the susceptibility of subtypes A to E to zidovudine, lamivudine, didanosine, nevirapine, and ritonavir. All subtypes tested had similar susceptibilities, apart from D subtypes, which had higher levels of resistance to all classes. Further studies have shown no differences in subtype responses to tenofovir, adefovir, and zidovudine, with no differences in responses to reverse transcriptase inhibitors among subtypes A through to F. Recent data that evaluated susceptibility of the Ghanaian subtype CRF02_AG strains suggested that they may be less susceptible to nelfinavir than are subtype B strains. Of note, specific data addressing viral subtype susceptibility to atazanavir have not been reported.
Two UK studies have failed to demonstrate that the presence of particular subtypes, individual polymorphisms, and mutation combinations are associated with either a decreased response to HAART or a worse prognosis. In the first of these studies, the relationship between viral subtype, disease progression, and response to HAART was examined (N = 867). For B (n = 457) and non-B subtypes (n = 410), there was a similar rate of pretreatment CD4 decline and similar time to viral undetectability. Within the "non-B" subtype category, 19% were A; 36% were C; 9% were D, and 4.4% were infected with recombinant strains (the most common being CRF02_AG). A study at Imperial College sequenced the reverse transcriptase and protease genes in African immigrants (n = 79) and showed that the non-B subtypes that were present in the study population were all similarly clinically susceptible to HAART. A similar study in therapy-naive individuals (n = 479) demonstrated that although baseline CD4+ cell counts were lower in individuals harboring the non-B subtypes (P = .02), there were no significant differences in virologic response up to 18 months.
A further consideration involves assays for detection of resistance. Subtype B isolates appear to be sequenced with a greater degree of success than non-B isolates. Because the pattern of drug resistance mutations may differ between subtypes, one recombinant virus phenotypic study (n = 32) has shown that data from drug-resistant B subtypes may not always be applicable to non-B subtypes.
It is not certain what to do with the individuals you describe, because they should have completed their antiviral responses by now. Ongoing viral replication clearly risks emergence of the reverse transcriptase resistance mutation M184V and potentially other nucleoside reverse transcriptase inhibitor resistance mutations (NRTI) as well as PI resistance mutations. I think it is worth considering modification of the regimen to an alternative NRTI and ritonavir-boosted PI combination to evaluate whether this leads to a more complete virologic response.