New compounds from all three classes (nucleosides, non-nucleoside RT inhibitors and protease inhibitors) were presented in the first session of the meeting.

In vitro data with the non-nucleoside quinoxaline GW420867X were shown [1]. Selection of drug resistant virus in the test tube using quinoxaline in combination with nucleosides (lamivudine or abacavir) resulted in some classical NNRTI mutations at positions (K101E, K103R, V106A, V108I, Y181C) and a novel substitution at L100V. Some of the results with one or more NNRTI mutations showed some sensitivity in vitro to efavirenz. The authors from the Rega-Institute and Glaxo suggested that combining quinoxaline with efavirenz might be worthwhile evaluating in vivo.

Agouron presented the resistance profile from their novel NNRTI AG1549 (formerly known as S1153) [2]. Previous experiments had shown that resistance can occur through combinations of mutations (V106A + F227L) or (K103T, V106A, L234I). Low level resistance (3-13 fold increase in IC₅₀) or no increases were observed with some of the classical NNRTI mutations (L100I (3x), K103N (1x), V106A (4x), Y191C (13x), Y188C (1,0), P236L (4x).

It seems that some of the newer NNRTIs have somewhat altered resistance profiles when compared to the compounds currently in the clinic. Researchers from the Southern Research Institute (Maryland, USA) presented a novel compound SJ-3366 [3]. This non-nucleoside RT inhibitor showed active against both HIV-1 and HIV-2. The mechanism of action of this compound is dual: inhibition of RT similar to all NNRTI (selecting for classical RT mutations) and inhibition of virus attachment. The effect of some RT mutations was discussed-- K101E 18 fold, K103N 50 fold, V108I 18-fold, Y191 10-100-fold, Y188 55-fold, P236 33-fold. A novel change in combination with Y181C created a million fold increase in IC₅₀. Further studies with this compound will be pursued. The patterns in these drugs are not dramatically different from those in current NNRTIs, and therefore it does not seem reasonable to call them second generation drugs. However, in vivo studies will reveal the potential value of these newer compounds.

The results of a phase I study with the non-nucleoside RT inhibitor (MKC-442) or emi .ne were presented by investigators from Triangle pharmaceuticals [4]. In two randomized trials using emivirine (in combination with stavudine + lamuvidine or stavudine + didanosine), analyses of therapy-naïve individuals were performed to investigate their resistance profiles. A total of 23 patients were studied and the classical changes were observed; K103N (13/3), G109A (2/23), Y191C (2/23), A98S + K101E (1/23), K101E + V106M/V (1/23), V108I (1/23) and E138K (1/23). In addition, two novel mutations E138Q (1/23) and A98S (1/23) were observed. The effects of these mutations on sensitivity to a variety of NNRTI were studied. It appeared that the A98S and E138Q mutations did not affect the sensitivity to delavirdine, efavirenz, or nevirapine. As expected, the K103N mutation conferred reduced sensitivity to all four NNRTIs tested. For all the other mutations at 101, 108, 138, 181 and 190, at least one of the four NNRTIs had retained activity (IC₅₀ increase < 4-fold in vitro).

New data for four protease inhibitors in various stages of development were presented. The novel Abbott compound 378 was evaluated in combination with ritonavir (ABT378/r) [5,6]. Patients failing a regimen of two nucleosides and a protease inhibitor, with viral loads of 10³ to 10⁵ copies RNA per ml, were entered in the trial. The protease inhibitor was switched to ABT378/r, either 400 mg/100 mg, twice daily or 400 mg/200 mg twice daily. On day 14 nevirapine was added and the nucleosides were altered to include one or more NRTI. During the first two weeks of the trial, effects from the protease switch were observed in most patients. In 66/70 patients a decrease of ^>/= 0.5 log₁₀ copies/mL of HIV RNA was observed (n=42) or a viral load value of <400 copies/mL (n=24). The response did not seem to be dependent on the previous protease inhibitor used. Analysis of protease resistance at the moment of switching revealed that an increase in IC₅₀ to ABT378 ranged between 0.7-26 fold. No relationship was found between the initial decline from baseline and the sensitivity to ABT378. The authors suggested the high serum levels of 378 in vivo might explain the observed effect even in patients with viruses with decreased susceptibility to 378 in vitro.

Three groups presented data on resistance profiles to tipranavir (PNU 140690) the new protease inhibitor developed by Upjohn [7-9]. Similar data were presented by all three groups showing that tipranavir has a relatively unique resistance pattern. The Virco investigators presented data from the largest study involving 125 isolates obtained from patients on a variety of protease inhibitors. From 85 isolates with a more than 10-fold increase in IC₅₀ to indinavir, ritonavir, nelfinavir and saquinavir, 74 (87%) remained completely susceptible to tipranavir. Only 3 isolates had IC₅₀ values > 10 fold that of the wild type control to tipranavir. An increase of more than four-fold in IC₅₀ was seen in seven isolates. The mutations involved in causing a reduction in sensitivity to tipranavir were a combination of 82T +84V+90M plus four to six secondary mutations. It will be interesting to determine the resistance profile from patients treated with tipranavir as the first Pl. In addition it remains to be determined whether tipranavir can be used to treat patients with resistance to any of the current protease inhibitors.

Agouron investigators discussed a novel protease inhibitor AG1776 (formerly JE2147, a petidomimetic inhibitor developed by the Japan-Energy Corporation) [10]. in vitro selection of drug resistant isolates resulted in a 15-fold reduction in susceptibility associated with mutations at L10F, I47V, V92I or I84V. Further evaluations will be required to fully understand the cross-resistance profile in vitro and in vivo.

A study of BMS 232,632 was presented by investigators from Bristol-Myers-Squibb [11]. Variants generated by extensive passage in vitro resulted in highly resistant isolates (79-188 fold). The basis for this high level resistance was in all cases a combination of mutations. In two cases a N88S change was observed (among other mutations) and I84V and L89M were observed in the other cases. These highly resistant isolates showed varying levels of cross resistance to other PIs in vitro.

Investigators from NCI showed data on a novel compound RS-344, designed with a 'resistant repellent' core [12]. The compound showed similar potency in both enzymatic and virological assays. A variety of HIV protease mutants, all containing the V82A mutation, remained sensitive to RS-344. In addition lower KI changes were observed towards other active site mutants, including I84V, V82F, I84V and G48V/L90M.The authors claim that the 'flexible core' of the inhibitor was responsible for the relative lack of cross resistance. It remains to be determined what the future of this compound will be.

Two nucleosides were discussed in this session. Researchers from Emory University presented data from studies they are performing to develop new nucleosides [13]. They presented data on the anti-HIV activity and cross-resistance profile of novel 2'-fluoro-2', 3' unsaturated D and L nucleosides (d4N). Most of the D-enantiomers remained active against 3TC- resistant M184V variants. Activity against the classical multi-drug-mutations was not reported. Further development will be made with this series of compounds. Investigators from Triangle Pharmaceuticals presented their data on DAPD, a novel nucleoside RT inhibitor [14]. DAPD is converted by the cell to another compound DXG (B-D-dioxolane guanine). in vitro selection with DXG resulted in a single L74V mutation, giving a four-fold increase in IC₅₀. The emergence of a K65R mutation, causing an 8-fold increase in IC₅₀, has also been described. Although a systematic analysis was not performed on cross-resistance patterns, some interesting observations were shown. The classical ZDV mutations, in different combinations and accompanied by the 3TC mutations, did not seem to result in great changes in IC₅₀s. Unfortunately the classical MDR resistance pathway (K65R, F116Y, Q151M) resulted in a 40-fold increase in IC₅₀. Further studies are warranted to evaluate the potential of this drug.

Two alternative approaches were presented for interfering with nucleosides metabolism in cells. Investigators from Innogenetics presented their studies on didox, a novel ribonucleoside reductase inhibitor, in an HIV-infected HuPBMC SCID mice [15]. Didox (3,4-dihydroxybenzo-hydroxamic acid) showed some interesting properties in this model system, indicating more significant activity than hydroxyurea.

Researchers from the Institute of Human Virology (Baltimore) and colleagues presented data on the combination of abacavir and mycophenolic acid (MA), an inhibitor of inosine monophosphate dehydrogenase [16]. This combination showed synergistic activity in vitro. Mycophenolic acid (cell-cept) a selective inhibitor of lymphocyte proliferation currently in use in organ transplantation, inhibits inosine monophosphate dehydrogenase, blocking the synthesis of guanosine monophosphate. The authors show that abacavir and MA inhibit HIV in activated peripheral blood mononuclear cells and macrophages. MA displayed significant antiviral activity against 3TC resistant isolates (L74/M184V). Interestingly, the combination of MA and zidovudine or stavudine was antagonistic, likely due to inhibition of thymidine phosphorylation by MA. The a .rs conclude that clinical trials are warranted to investigate the potential clinical role for containing MA with abacavir.

Two papers on potential new treatment strategies for treatment of HIV were presented. Investigators from IrisCaixa (Barcelona, Spain) and the Rega-Institute presented studies on the potential for inhibition of earlier stages of HIV-replication, namely HIV absorption, binding to the host-cell, and virus-cell fusion [17]. An integrated overview on the role of the action of AMD3100 was presented. This compound of the bicyclam-class selected for multiple mutations in the envelope gene in vitro. Two mechanisms were shown to be involved in virus binding to CD4 and chemokine reception dependent virus entry.

Finally, investigators from Trimeris presented data on the selection of drug resistant virus in patients treated with T20 [17]. T20 is a linear peptide fusion inhibitor currently in Phase II clinical trials. Previously, in vitro selection of T-20 resistant isolates resulted in changes in gp41, the transmembrane glycoprotein. The viruses resistant to T20 were shown to be sensitive to a second-generation fusion inhibitor. This inhibitor, T1249, is a hybrid synthetic peptide consisting of 39 amino-acid sequence derived from a highly conserved regimen of the gp41. Further studies will be performed to establish the activity and resistance profile of T1249.

In conclusion, active basic research and clinical studies are being performed in all areas of HIV-therapy. Novel nucleoside, non-nucleoside RT inhibitors as well as protease inhibitors are in development. Research is focussing, among other things, on cross-resistance profiles, and the opportunities to exploit the potential lack of complete cross-resistance.

The field of fusion inhibitors is maturing; Phase I/II studies are currently being performed, and appropriate studies on resistance development are incorporated in these programs. Whereas some of the compounds presented at this meeting may have a promise based on claims of lack of (complete) cross-resistance, it seems too early to draw selective conclusions. For all of those compounds it remains to be seen what their value for clinical use will be. Once they have been analyzed in (large) clinical trials definite conclusions can be made.