New HIV drugs, resistance, pharmacogenomics

Mike Youle MD, for NATAP

Drug resistance

There was hardly a feast of new data for the treatment ferrets at this year’s ICAAC. The 11 September debacle had allowed data to be presented at Athens at the 8th European Conference on Clinical Aspects and Treatment of HIV infection held in October. Even allowing for some repetition of studies of tenofovir and atazanavir little else was evident.

However, the importance of having a pipeline of agents to deal with HIV resistance was emphasized by the late-breaker of Doug Richman and the HCSUS study [1]. This revealed surprisingly high rates of resistance to at least one antiretroviral in a very carefully selected sample of patients bled in 1999. PhenoSense assays, performed by ViroLogic, San Francisco gave 1080 (93%) successfully phenotyped in 1167 subjects who had measurable virus above 500copies/mL from 1647 plasma samples.

This sample group can be extrapolated to represent 132,442 (63%) of the 209,000 US adults under care in 1996 who survived to 1999 with a detectable HIV viraemia. The prevalence was higher in subjects who were on antiretroviral therapy (87%) versus those not (41%), p=0.0001. 78% of patients with detectable HIV had any drug resistance. Richman said this represents 50% of the entire US HIV-infected population under care if you assume patients with <500 copies of viral load had no drug resistance. Resistance was more common in those who had more advanced disease and with lower nadir CD4 T cell counts as well as in the population whom had been most exposed to antiretrovirals (non-Hispanic, white gay males who were highly educated and privately insured).

Of course, since this is not a randomised study and resistance is probably the price to be paid for survival in this group of subjects. One wonders, however what the relative importance of minor amounts of antiretroviral exposure as in prevention of mother to child transmission, bare-backing and viral spikes due to poor absorption, adherence or infections.


There is a great need for new targets against HIV. A German consortium from Axxima Pharmaceuticals and the University Erlangen-Nuremburg showed data on an exciting compound, AXD-455 [2]. This works by blocking the activity of a chemical messenger eIF-5A that transports RNA from the nucleus of the cell to the cytoplasm for further processing and creation of new viral particles. No other agents that act in the nucleus are currently at an advanced phase of development, although Merck has recently reported that they are moving ahead with phase I/II studies of an integrase inhibitor.

The drug appears to have significant activity in the test tube against clinical viral strains resistant to all available antiretrovirals and blocked both M- and T-tropic strains and is now in early clinical trials in Europe although few details were given. From what was shown it would seem that the median effect of this agent is a less than a 1-log suppression from baseline of HIV-RNA. However it does open a new front on the battle against the virus.


More information on tipranavir was presented by Charles Farthing from Los Angeles who unfortunately appeared not to have been given all of the available data regarding this Boehringer Ingelheim’s tipranavir study BI 1182.4 [3]. In the trial subjects who had taken a variable amount of HAART containing a single protease inhibitor were randomised to receive two doses of tipranavir (500mg bid [n=20] or 1250mg bid [n=21]) with ritonavir 100mg bid as a pharmacokinetic enhancer; or saquinavir 400mg + ritonavir 400mg bid [n=21]. All arms exhibited a median 1.3-2 log drop in viral load with the best response in the high dose tipranavir arm.

Small numbers of participants precluded any significant differences between any of the arms with regard to surrogate marker responses and clinical or laboratory adverse events. The major problems across all arms were diarrhoea, nausea and rise in lipids, which has been seen in all ritonavir boosted or based protease inhibitor regimens. From this trial it really was not possible to determine the importance of these findings especially as no data was presented on the resistance patterns of responders or non-responders, and 40-50% of patients reportedly had no PI resistance.


The atazanavir study that produced significant new information concerned a comparative study of a novel combination of protease inhibitors atazanavir 400mg or 600mg once daily with saquinavir 1200mg once daily compared to saquinavir 400mg + ritonavir 400mg twice daily [LB-16]. This of course would have more relevance to the current therapeutic milieu if the comparison arm had been saquinavir boosted with ritonavir 100mg once or twice daily.

In the study 85 subjects with a viral load between 1000 and 100,000copies/mL and a CD4 cell count >100cells/cumm were followed on assigned medication for 24 weeks in the first instance. The data for lipid changes was perhaps the most important finding from this study. Mean fasting triglycerides (mg/dL) at baseline were 223 in the 400mg atazanavir group, 177 in the 600 mg arm, and 191 in the ritonavir + saquinavir group. Mean total cholesterol at baseline ranged between 181 and 202mg/dL.

Total cholesterol rose by 10% in the ritonavir + saquinavir arm whilst remaining the same as baseline in the 400 mg atazanavir arm, and decreased by 10% in the 600 mg atazanavir arm (p<0.05) over the study period of six months. Fasting triglycerides increased by over 90% in the ritonavir +saquinavir group, but declined by approximately 25% in the atazanavir arms. Jaundice was reported in 13-19% of the atazanavir arms and was not associated with major alterations in liver enzymes and decreased over time. This problem was seen in the early days of indinavir and was mooted as a big potential problem, but this has not really materialised in any significant way so perhaps it is best to take a watching brief as to its importance.


The other protease inhibitor that was paraded was the earliest data from TMC114 an analogue of the drug TMC126 that Tibotec-Virco tested last year. This compound has nanomolar range activity against HIV (potent). Ronald Van der Geest presented data from HIV negative human volunteers. The 18-subject study could find no significant toxicity whilst plasma concentrations from single doses of 2400mg and 3200mg lay between 26.2 and 27.8microM [5]. This formulation contained polyethylene glycol as a filling agent, which is known to cause some problematic diarrhoea, not expected to be a problem with later formulations.

Another presentation on this drug evaluated the in vitro capacity to suppress both wild-type virus and that of 261 randomly selected recent recombinant clinical isolates, among which 32% showed a greater than 10 fold increase in EC50 for at least one protease inhibitor [6]. The drug has a median activity against 90% of viruses of 10.3nM and there appeared to be no greater than five-fold increase in the EC50 when tested against a series of 20 HIV variants resistant to available protease inhibitors.


Murabutide is a novel synthetic immunomodulator developed and tested in France. It has activity in vitro against HIV infected macrophages, dendritic cells and T4 macrophages and has reduced HIV expression in SCID-HU mice [7 and 8]. In this study, which was not randomised, 22 subjects were followed as controls whilst 20 subjects received subcutaneous injections of 7mg murabutide/day five days/week for six weeks.

After treatment immunological improvement a significantly better median CD4 rise from 280 to 350 cells/mm3 compared a decline from 280-250 cells/mm3 in the control group (p=0.002). No significant effect was seen on viral load suppression although 40% of the treated group had undetectable viral loads compared to 14% in the untreated controls. A definite agent to follow, especially since the French have been the only country to embrace interleukin-2 to the extent of making it available, as an immunomodulator, to those with CD4 cell counts <200/cumm.

Treatment interruption

Michelle Parish and Joel Gallant reported a cohort study that examined what happened to 48 subjects who discontinued HAART [9]. The median CD4 cell count before their treatment was commenced was 426/cumm and the median pre-HAART viral load was 23,575copies/mL. At the time of starting therapy 38 (79%) met the criteria for commencing treatment (CD4 <500 and HIVRNA>20,000copies/mL). However only 15 (31%) were within the 2001 DHHS guidelines which were more rigid i.e. CD<350 and HIV RNA>55,000copies/mL.

The major reason for coming off therapy was a perception of lack of need (20/48) whilst toxicity was the reason in 8/48 and non-adherence in 11/48. At the time of presentation 12 subjects (25%) had restarted treatment and those who did so were more likely to have a faster CD4 cell decline (44 vs 18cells/month) and a lower CD4 at discontinuation (592 vs 733). Of these individuals 10/12 restarted due to falling CD4 or rising viral load whilst 2/12 was for patient request alone.

It would appear that a treatment interruption was not detrimental to a high proportion of subjects who had commenced antiretrovirals when it was deemed more important to commence therapy early in the course of infection. However it will remain an important issue to examine more carefully using a degree of subject randomisation.

Pharmacogenomics – insights into an important field

Amalio Telenti from Lausanne, Switzerland opened his talk on the pharmacogenomics of antiretrovirals with an anecdote concerning King Mithradates who allegedly took small drafts of poison thereby inducing his liver function that subsequently protected him from future exposure from an attempted assassination [10].

He posed the question: Was it really the induction of the cytochrome p450 system or was it due to an underlying genetic advantage he might have had? During the ensuing talk he presented tantalizing data, to be published in the Lancet in January 2002, suggesting that much of the variability in antiretroviral drug levels seen in clinical practice is due to genetic variations in the expression of enzymes in the liver and transporter chemicals such as p-glycoprotein (PGP). When nelfinavir and efavirenz levels in the blood in clinical samples were correlated with the genotype of CYP2D6 and CYP3A4 in a cohort of individuals the former showed a significant effect on serum drug levels (p=0.04) whilst the latter did not (p=0.17). There was a greater effect due to the expression of MDR 1 3435 such that those individuals with the CC genotype had higher drug levels that those with TT genotype.

In addition an unexpected relationship was revealed in that TT genotype predicted almost a 50% greater increase in CD4 T cell count in HAART treated individuals compared to the CC genotype (p=0.005). This puts MDR genotype up there as a major factor and the first genetic marker associated with CD4 change under antiretroviral therapy surpassed only in magnitude by baseline HIV RNA. He listed several agents that are already available that are inhibitors of PGP such as cyclosporin A and verapamil, as well as a number of new drugs in development such as LY-335979 [Lilly] and GF 120918 [GlaxoSmithKline] that could potentially modify the genetic consequence.

Finally, some disturbing data were shown suggesting that since there is a major variability in MDR genotype by ethnicity that this could translate into an important determinant factor for the pharmacokinetics of HIV therapy.

It would seem that by the end of next year we will all be experts in pharmacogenomics… a new skill to learn and a wide but obviously important area to research.


  1. Richman DD, Bozzette S, Morton S, et al. The prevalence of antiretroviral drug resistance in the US. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Abstract LB-17.
  2. Bevec D, Hauber I, Kratzer F, et al. AXD455 – a novel inhibitor of antiretroviral drug-resistant HIV by pharmacological interference with nuclear export of viral mRNA. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Abstract LB-14.
  3. Slater L, Farthing C, Jayaweera J, et al. Safety and efficacy of tipranavir (TPV), a novel non-peptidic protease inhibitor, plus ritonavir (RTV), in PI-failure patients. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Abstract LB-15.
  4. Haas D, Zala C, Schrader S, Thiry A, McGovern R, Schnittman S. Once-daily atazanavir plus saquinavir favorably affects total cholesterol (TC) and fasting triglyceride (TG) profiles in patients failing prior PI therapy (Trial AI424-009, Wk 24). Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Abstract LB-16
  5. Van der Geest R, van der Sandt D, Gille D et al. Safety, Tolerability and Pharmacokinetics of Escalating Single Oral Doses of TMC114, a Novel Protease Inhibitor (PI) Highly Active against HIV-1 Variants Resistant to Other Pis. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Presentation 1934
  6. de Béthune M, Wigenrick P, Jonckheere H, et al. TMC114, a highly potent protease inhibitor (PI) with an excellent profile against HIV variants highly resistant to current PIs. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Abstract F-1677.
  7. Darcissac EC, Truong MJ, Dewulf J, et al. The synthetic immunomodulator murabutide controls human immunodeficiency virus type 1 replication at multiple levels in macrophages and dendritic cells. J Virol (United States), Sep 2000, 74(17) p7794-802
  8. George M. Bahr, Edith C. A. Darcissac, Nathalie Castéran, Corinne Amiel, Cécile Cocude, Marie-José Truong, Joëlle Dewulf, André Capron, and Yves Mouton Selective Regulation of Human Immunodeficiency Virus-Infected CD4+ Lymphocytes by a Synthetic Immunomodulator Leads to Potent Virus Suppression In Vitro and in hu-PBL-SCID Mice J. Virol. 2001 75: 6941-6952.
  9. Parish MA, Raines C, Higgins M, Gallant JE. Treatment discontinuation (DC) in patients with marginal indications for HAART. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy, December 16-19, 2001; Chicago, Illinois. Abstract I-673.
  10. Telenti A. Pharmacogenetics of antiretrovirals. Program and abstracts of the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy; December 16-19, 2001; Chicago, Illinois. Abstract 423.

Source: NATAP

Links to other websites are current at date of posting but not maintained.