The 3^rd European Symposium on the Clinical Implications of HIV Drug Resistance, held in Frankfurt, Germany, on 23-25 February 2001 was an outstanding success. This conference, organized annually by Drs. Veronica Miller and Schlomo Staszewski, has now become a staple among those interested in HIV drug resistance and related fields.

Issues of viral fitness
Viral fitness was addressed in a number of presentations, among which were those by Swiss and American groups. It was demonstrated by C. Suné et al. that the generation of single- and multiple-site mutants of the HIV protease gene could be used to examine issues of viral fitness (Abstract 2) [1]. In general, these scientists demonstrated that variants of HIV containing distinct point mutations associated with resistance to PI can have impaired replication fitness. The investigators suggested that these findings may have clinical utility, particularly when drugs are used under conditions when drug toxicity may limit dosage. In particular, they demonstrated that viruses containing mutations such as G48V and L90M do not have an altered fitness profile, while other combinations of mutations, such as A71V/V82T and 184V, can lead to diminished fitness. Additional data suggested that the maintenance of diminished fitness might necessitate using the same drugs that selected these codon changes, since reversions to wild-type codons might otherwise take place in the context of ongoing viral replication. In several instances, the addition of the L63P substitution was shown to actually increase viral replication kinetics, raising concern about resistance-conferring mutations that may serve an important compensatory function in regard to viral replication.

Data on pertaining to this topic were also presented by R.M. Grant et al. (Abstract 3) [2]. They examined weekly samples from 13 subjects. Measurement of viral fitness was based on assessments of viral replication in vivo. The results showed that PI-resistant variants of HIV were markedly less fit than PI-susceptible viruses. During effective therapy, wild-type viruses appeared to persist at low levels. However, more effective therapy may suppress wild-type virus and select for highly resistant variants with impaired fitness capacity. These authors also concluded that some mutations could impact on viral replication fitness more than others and that the maintenance of some mutations could be contemplated in a therapeutic context.

A presentation on this theme was also made by Mark Wainberg (Abstract 5) [3]. The M184V mutation, associated with 3TC and some other drugs, is located within the active catalytic site of reverse transcriptase and can result in a number of altered enzymatic functions, including reduced processivity, increased biochemical fidelity, diminished pyrophosphrolysis, and diminished primer unblocking. While the individual effects of any of these changes may be minimal, their cumulative impact on HIV replication kinetics and susceptibility to antiviral drugs may be very high. Dr. Wainberg argued that the M184V mutation should be maintained, even after treatment failure involving 3TC and/or other drugs. He also summarized data to demonstrate that patients whose virus had the M184V mutation were less likely to develop a wide array of other NRTI-associated mutations than were patients who did not possess the substitution, and that the 184V mutation could, at least temporarily, lead to resensitization and/or enhanced sensitivity to AZT, d4T and some other drugs.

Nucleoside-associated mutations
Drs. Brendan Larder (Abstract 6) and Dominique Costagliola (Abstract 7) presented data on how our understanding of HIV drug resistance has undergone profound changes during the last several years with respect to nucleoside-associated mutations (NAM's) [4,5]. There is now a growing consensus that these mutations may, in fact, be common to multiple nucleoside compounds and that, when present in various combinations, can confer cross-resistance among a wide array of nucleoside drugs. Clearly, this paradigm change has and will continue to fundamentally affect the way in which we make treatment recommendations on the basis of genotypic resistance test results. This is especially true with regard to the use of abacavir, and presentations by Lanier et al. (Abstract 13) and by Schmidt et al. (Abstract 14) highlight this point [6,7]. It will become important in future to identify precisely which combinations of NAM's are associated with diminished sensitivity to abacavir and other drugs. It is also crucial to ask how specific combinations of mutations may impact the general use of NRTI in view of the problem of cross-resistance.

Differences among HIV subtypes
The issue of whether or not viruses of different subtypes or clades may respond in differently to antiviral drugs was highlighted in a presentation by C. Loveday et al. In their study, comparisons were made between patients who had undergone infection with HIV-1 of subtype B vs a variety of non-B subtypes (Abstract 15) [8]. A case-control study of 100 patients showed that few or no differences existed in regard to either initial or long-term virologic responses to a variety of antiviral regimens.

The question of viral subtypes was also addressed in a discussion of EuroGuidelines with regard to drug resistance. Dr. Luc Perrin of Geneva presented data showing that a high proportion (30-40%) of new HIV infections in Europe represent recombinants among subtype B viruses and viruses of other subtypes. The issue of viral recombination is extremely important with respect to drug resistance, since it is well understood that recombination among viral genomes can contribute to both elevated levels of drug resistance as well as the presence of genotypes that confer resistance to a variety of drug classes.

In this context, presentations from several groups highlighted the increased prevalence of drug-resistant viruses in primary infection in a number of European countries, including Portugal, Poland, France, Germany, and Spain. In particular, the group of Soriano et al. demonstrated a high rate of PI-associated substitutions, including natural polymorphisms in drug-naive subjects who had been infected with non-B subtypes in Spain (Abstract 63) [9]. This highlights a need for further information on this topic and for guidelines based on HIV-B subtypes that may be of use toward guiding clinical therapy with other HIV subtypes.

Predictive value of HIV genotyping
As stated above, the presence of multiple NRTI-associated mutations is an important factor affecting decisions that might be made for future therapy based on genotypic assessments. Further data on this topic were provided by the group of B. Clotet and the Havana study group, who provided data after six months of analysis based on studies conducted on ART-experienced patients (Abstract 34) [10]. The investigators reported by multivariate regression analysis that both the use of genotyping and expert advice were associated with diminished viral loads, and that genotyping plus expert interpretation were associated with improved virologic outcomes and successful virologic control after 24 weeks of evaluation. The group of Callegaro et al. also presented data on this topic to suggest that highly pre-treated patients, who have been exposed to multiple anti-HIV drugs, might represent a group in whom genotyping and expert interpretation might have limited value (Abstract 22) [11]. These data are consistent with earlier reports.

See Dr. Charles Boucher's report from the 3^rd European Symposium on the Clinical Implications of HIV Drug Resistance.