The diversity and prevalence of non-subtype B (NSB) viruses paint an increasingly complex picture, particularly in Europe and Africa, where along with complex circulating recombinant forms (CRF), they pose a growing challenge to our understanding of HIV drug susceptibility and resistance, as well as to the potential efficacy of experimental vaccines. There is increasing interest in using the phylogenetic results of resistance testing in epidemiologic surveys, clinical trials and clinical practice to monitor the frequency of NSB viruses and CRFs in order to identify clusters and avenues of transmission. Unfortunately, as efforts to extend ARV treatment in resource-poor, high prevalence regions increase, drug resistance in NSB viruses remains largely uncharacterized.

Methods for determining subtype and resistance
HIV-1 subtypes, which are typically classified according to phylogenetic differences in the envelope gene, have been identified in clinical studies through sequencing techniques designed for drug resistance assays. Studies presented at the 9^th CROI used the TruGene^TM (Visible Genetics), ViroSeq^TM(Applied Biosystems) or "homebrew" sequencing techniques to obtain RT and protease sequences and identify drug resistance mutations (for information on these assays, see HIVresistanceWeb's Test Information pages. Increasingly, determinations of HIV-1 subtype may be derived from results of RT and protease gene sequencing. While only Visible Genetics' TruGene 1.0 is FDA cleared for clinical use (the Applied Biosystems kit is currently before the FDA), the use of these kits for resistance testing in NSB viruses has not been systematically validated by comparing genotypic and phenotypic resistance test results across a range of NSB viruses.

In a comparison of test methodologies, Nelson Michael and colleagues presented sequencing results using the two test kits in a panel of 34 pre-treatment NSB isolates [1]. Two of 34 (one subtype C and one subtype G) were difficult to sequence with the TruGene assay, while all viruses were successfully sequenced with the ViroSeq kit. The authors commented, however, that there was a need for further testing and refinement of primers since there were segments of the RT and protease which yielded ambiguous sequence results with both kits.

Subtype distribution and drug resistance in subtype B vs. NSB
Genotyping studies from Europe have shown that approximately 50% of newly identified/newly infected patients are infected with NSB viruses. However, clinical studies to date have suggested that virologic responses to ARV therapy among patients with NSB viruses are equivalent to those of patients with subtype B viruses. In one such study, P. Hermans, N. Clumeck and colleagues analyzed virologic and immunologic responses to therapy in 121 patients in Antwerp: 64 had subtype B and 57 had non-B HIV. Virologic responses were found to be consistent those of the general clinic population in the US [2].

In another study, the heterogeneity of subtypes among HIV-infected children in Europe was shown by phylogenetic analysis of resistance test results in the PENTA V study [3]. The proportions of children infected with various HIV-1 subtypes are depicted in the Figure below. As in studies in adults, there was no evidence that HIV-1 subtype was associated with virologic response (either change in RNA or the proportion with RNA < 400 or 50 copies/mL at 24 and 48 weeks). However, the assay failure rates for genotypic and phenotypic tests were inversely related to HIV-1 RNA (genotype, P = 0.002; phenotype, P = 0.01) and were higher in non-B than B subtypes in genotypic and phenotypic assays performed by Tibotec-Virco (genotype, P = 0.01; phenotype, P = 0.03).

Figure. HIV-1 subtype distribution in the PENTA V study (adapted from D. Pillay et al, 9th Conference on Retroviruses and Opportunistic Infections, Abstract 813).


In one phylogenetic study from France which combined pol gene sequencing with envelope serology, a range of diverse viruses and CRFs were described in a group of recent seroconverters [4]. The protease polymorphisms K20I and M36I were found to be highly prevalent in subtypes CRF02, G, J, F2, and CRF06. Only three isolates were found to have NRTI- or NNRTI-associated resistance mutations: one subtype D isolate (M184I), one subtype J isolate (T69N, K70R), and one subtype A isolate (K103N).

In Spain and Portugal, subtype G viruses account for an increasing proportion of NSB viruses. Holguín, Álvarez, and Soriano reported that sequences of clinical NSB isolates from Iberian patients revealed that a majority were recombinant G variants [5].

Evidence for and mechanisms of subtype-associated differences in drug resistance
The prevalence of subtype G and CRFs in Portugal may have significance in the choice and sequencing of PI. Gomes et al reported that among 30 patients failing nelfinavir-containing therapy as their first PI regimen, of whom 19 were infected with HIV-1 subtype B and 11 with subtype G, nelfinavir-associated resistance mutations differed significantly depending on viral subtype [6]. As would be expected, the majority (7 of 11) of subtype B samples with genotypic evidence of nelfinavir resistance had the D30N mutation, although there was one with the L90M mutation. In contrast, all 10 of the subtype G-infected samples with genotypic evidence of nelfinavir resistance exhibited the L90M pathway or another set of mutations. Further, the I54V mutation was present in six of 10 subtype G-infected patients but only in one of 11 subtype B genotypes. These results suggest that the two distinct pathways (D30N and L90M) which lead to nelfinavir resistance may be driven by subtype-specific polymorphisms and backbone of the protease gene, and that subtype G infected patients could be at much higher risk of developing broad PI cross-resistance when nelfinavir is used as the initial PI.

The basis for a differential response to PI among HIV subtypes may be explained by the presence of multiple polymorphisms in subtype A and G viruses, many of which are located near classically described PI mutations. In a study from Nigeria, researchers from the CDC used gp41 serologies to identify subtype and performed both phenotypic and genotypic resistance tests on samples from untreated patients [7]. Their analysis of mutations associated with resistance to ARV drugs revealed no known primary resistance-associated mutations. However, many of the isolates contained between nine and 17 amino acid substitutions within the protease region, all of which would be classified as secondary mutations or polymorphisms.

In the same study, an interesting sequence from an untreated individual demonstrated low -level reductions in susceptibility to nelfinavir (3.8-fold) and ritonavir (3.4-fold) relative to the drug-sensitive control [7]. This sequence included myriad amino acid changes relative to the subtype B consensus sequence, including L10I, I13V, K14R, I15I/V, K20I, E35Q, M36I, N37D, R41K, R57K, L63L/P, I64I/M, C67E, H69K, T74T/S, V82I, and L89M. V82I is of particular interest as previous studies have demonstrated an association between this mutation and moderately reduced susceptibility to PI (see Stanford HIV RT and Protease Sequence Database for complete references and analysis of phenotypic and clinical drug resistance associated with V82I in untreated patients infected with subtype B and NSB HIV).

In conclusion, the increased use of resistance testing is extending the wealth of knowledge gained from geographic surveys of HIV subtype distribution. As increasing diversity is recognized, it is likely that additional sequence- and phenotype-associated differences in both RT and protease will be documented. At present, HIV subtype is yet another uncertain factor when making clinical decisions about ARV treatment, although new data and analyses may soon help to elucidate important subtype-dependent differences that affect virologic response to ARV therapy. Growing information on the responses of NSB viruses to ARV drugs may, along with genetic studies, pharmacogenomics and virus load testing, help to drive the expansion of drug treatment efforts while simultaneously contributing to their effectiveness across different geographical regions and HIV subtypes.

References

L. Jagodzinski, J. Cooley, S. Kelly, N. Michael. Performance of the TRUGENE HIV-1 Genotyping Kit and the Applied Biosystems HIV-1 Genotyping System in Sequence-Based Analysis of Non-B HIV-1 Subtypes. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 594.
P. Hermans, J-C. Schmit, K. Kabeya, E. O'Doherty, S. De Wit, B. Sommereijns, N. Clumeck. Virological Response to Salvage Therapy at 6 Months in Patients with B or Non-B Subtypes. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 427.
D. Pillay, A. S. Walker, D. M. Gibb, A. de Rossi, S. Kaye, M. Ait-Khaled, M. Munoz-Fernandez, A. Babiker. The Impact of HIV-1 Subtypes on Virological Response and Emergence of Resistance in the PENTA 5 Trial. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 813.
H. Fleury, A. Caumont, M. Faure, J. C. Plantier, P. Roques, E. Couturier, F. Simon. HIV-1 Diversity in France, 1999-2001: Molecular Characterization of non-B HIV-1 Subtypes and Potential Impact on ARV Susceptibility. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 758.
Á. Holguín, A. Álvarez, V. Soriano.Increasing Prevalence and Heterogeneous Nature of HIV-1 Subtype-G Recombinants Spreading in Southern Europe. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 759.
P. Gomes, I. Diogo, M. F. Gonçalves, P. Carvalho, J. Cabanas, M. C. Lobo, R. Camacho. Different Pathways to Nelfinavir Genotypic Resistance in HIV-1 Subtypes B and G. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 46.
S. M. Agwale, C. Zeh, E. Paxinos, L. Odama, D. Pienazek, C. Wambebe, M. L. Kalish, R. Ziermann. Genotypic and Phenotypic Analyses of Drug Susceptibility in HIV-1 Isolates from Drug-Naïve Patients in Nigeria. 9th Conference on Retroviruses and Opportunistic Infections. 24-28-4 Feb 2002, Seattle, WA. Abstract 461.