Update on Antiretroviral Therapy
Report from the 4th Conference on Retroviruses and Opportunistic Infections
Paradigm shifts can't happen twice a year. As rapidly as the field of HIV therapy is changing, we couldn't expect the headlines that we saw with last summer's international conference in Vancouver. Instead, this year's national meeting was characterized by incremental but important progress. In the area of antiretroviral therapy, we learned more about new agents, new combinations of agents, resistance, and drug interactions.
New Antiretroviral Agents:
Nucleoside and Nucleotide Analog Reverse Transcriptase Inhibitors
1592U89 is a new carbocyclic nucleoside analog being developed by Glaxo Wellcome. Potent for a nucleoside and able to penetrate the CNS, this drug appears to be well-tolerated, and its pharmacokinetics allow for twice daily dosing. In a dose escalation trial, patients on monotherapy with 1592 had a viral load reduction of over 1.5 logs, and 60% of patients treated with 1592/AZT achieved undetectable levels of viral RNA [Harrigan et al., Abstract 15]. Resistance to 1592 may lead to cross-resistance to 3TC, but cross-resistance to AZT or d4T does not appear to occur.
MKC-442 is a nucleoside analog developed by Triangle Pharmaceuticals. Like 1592U89, it penetrates the CNS. At the highest dose tested thus far, one log declines in viral load have been seen, but trials with higher doses are in progress. This drug exhibits in vitro synergy with AZT, ddI, and saquinavir, and while it acts more like a non-nucleoside reverse transcriptase inhibitor (NNRTI) than a nucleoside analog with respect to its binding properties, it is active against nevirapine-resistant strains [Moxham et al., Abstract LB1].
Adefovir, a nucleotide analog manufactured by Gilead Pharmaceuticals, has activity not only against HIV, but also against HSV, CMV, and hepatitis B virus. Pharmacokinetics allow for once daily dosing, and the drug is currently in Phase III trials at a dose of 125 mg daily.
The Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTIs)
Loviride is an NNRTI manufactured by Janssen. Two large trials involving loviride were presented at this conference. In the CAESAR trial pre-treated patients with advanced disease experienced no additional benefit when loviride was added to a regimen that included AZT/3TC [Cooper et al., Abstract 367]. However, in the AVANTI-1 trial antiretroviral-naive patients had a better response to AZT/3TC/loviride than those on AZT/3TC alone [Rozenbaum et al., Abstract 368]. The magnitude of benefit was less than has been seen with AZT, 3TC and a protease inhibitor, however.
Delavirdine (Rescriptor) remains in limbo at the FDA in part because of unimpressive efficacy data. However, researchers from Pharmacia-Upjohn presented results from two trials which suggested that the use of delavirdine may be associated with "re-sensitization" to AZT, not unlike the effect seen in patients taking AZT and 3TC [Wathen et al., Abstract 12]. Drug interactions associated with delavirdine are discussed below.
DMP-266 (DuPont Merck) has potent antiretroviral activity and a pharmacokinetic profile that allows for once-daily dosing. At a late-breaker session, the results of a trial involving DMP-266 plus indinavir were presented [Riddler, Abstract LB2]. After 24 weeks, patients on combination therapy had a 2.2 log decrease in viral load, compared to a 1.5 log reduction in patients taking indinavir monotherapy. CD4 counts increased by approximately 100 cells/mm3 in both groups. Viral loads were undetectable (
Nelfinavir (Viracept) promises to be the next FDA-approved protease inhibitor. It has a unique first mutation at D30N that does not appear to be associated with significant cross-resistance to other protease inhibitors, characteristics which may make it an attractive first-line agent [Patrick et al., Abstract 10]. Nelfinavir is dosed three times daily with meals, and its principal side-effect is mild-to-moderate diarrhea. The results of the three pivotal trials involving nelfinavir were presented by William Powderly [Abstract 370]. These trials included a monotherapy trial (508), a trial comparing d4T to d4T/nelfinavir (506), and a trial comparing AZT/3TC to AZT/3TC/nelfinavir (511). The latter trial, in which the best results were seen, was presented in the greatest detail. At 24 weeks, those taking AZT/3TC/nelfinavir had viral load reductions of approximately 2.5 logs, compared to 1.4 logs in the AZT/3TC group (p100,000 copies/ml did better on the higher dose of nelfinavir. There was no significant difference in toxicity between the two doses.
141W94, the Glaxo Wellcome protease inhibitor known as VX-478 during its development by Vertex, is now in Phase II/III clinical trails. It is synergistic with AZT, ddI, and ddC and can be dosed twice daily. In the type of study referred to by Dr. Joep Lange at this conference as "incestuous," Glaxo Wellcome administered this drug in combination with its investigational nucleoside analog 1592U89 and found mean viral load reductions of over 2 logs, with 71% of patients achieving undetectable viral loads after four weeks [Schooley et al., Abstract LB3].
ABT-378 is Abbott's newest foray into the world of protease inhibitors. While it has not yet even entered Phase I trials, the presentation on this drug generated a great deal of interest because it is ten times more potent than Abbot's current agent, ritonavir, no withering flower itself. It appears to be highly active against ritonavir-resistant strains, and has the "advantage" of being highly sensitive to blockade of the p450 enzyme CYP 3A4. Thus, we should probably tolerate a little incest when we observe that truly homeo-pathic doses of ritonavir are sufficient to significantly increase ABT-378 levels and to prolong its half-life. As for toxicity, so far the rats have no complaints.
Combination Therapy and Strategies of Antiretroviral Therapy:
After the Washington conference we can feel comfortable using a number of new drug combinations. Dual protease inhibitor regimens will become increasingly common, and NNRTIs and protease inhibitors can be combined safely provided drug interactions are considered. Both developments are god-sends for the problematic patient in need of protease inhibitor therapy who has been heavily pre-treated with nucleosides. In addition, various mixtures of ddI, d4T, and hydroxyurea appear to be safe and effective. Finally, the pipeline continues to contain a variety of new agents in all classes, many of which promise to have unique resistance patterns.
- In addition to his comments on incest, Dr. Joep Lange from Amsterdam gave a talk on the future of clinical trials that was sometimes inflammatory, frequently controversial, and always entertaining. He urged common sense with respect to treatment decisions: "When to start? Whenever the patient is ready. HIV is enough of a reason to treat. When to switch? When the drugs aren't working any more. When to stop? Also, when the patient is ready" [Lange, Abstract S54].
- A number of clinical trials addressed combination therapy using currently available agents. The CAESAR trial was presented at the Interscience Conference on Antimicrobial Agents and Chemotherapy last fall, and was reported
on in The Hopkins HIV Report (Vol 8, No. 3). Briefly, it demonstrated that in patients taking AZT-containing regimens (monotherapy or AZT/ddI or AZT/ddC), the addition of 3TC was associated with a 55% decrease in the risk of progression or death, and a 58% decrease in mortality [Cooper et al., Abstract 367].
The ALTIS trial was presented by Dr. Christine Katlama from Paris [Katlama et al., Abstract LB4]. In this non-comparative trial, antiretroviral naive patients treated with d4T/3TC had a viral load reduction of 2 logs at 4 weeks, decreasing to 1.7 logs at 24 weeks. CD4 counts increased by approximately 100 cells/mm3. In con-trast, those pretreated with nucleosides other than d4T and 3TC (AZT, ddI, ddC) had viral load reductions of 1.4 logs at 4 weeks that fell to 0.66 logs at 24 weeks, and CD4 count elevations were less than 50 cells/mm3. While d4T/3TC has been a popular regimen for patients ex-perienced with other nucleosides, this effective regimen, like others, may be less effective in pre-treated patients.
- The results of ACTG 193A were presented [Henry et al., Abstract LB6]. This large clinical endpoint trial, which began shortly before the International AIDS Conference in Berlin and ended in June 1996, now seems dated in this era of more aggressive antiretroviral therapy. The study found that in patients with advanced HIV disease (CD4
- Since ddI/d4T continues to look good, as does ddI plus hydroxyurea, it follows that ddI/d4T/hydroxyurea would not be far behind. Various combinations of these three agents were the subjects of a number of posters, all of which showed promising declines in viral RNA in small, uncontrolled studies.
- In another trial involving patients with advanced disease (CD4
- Another hot topic was dual protease inhibitor therapy. John Mellors reviewed combinations in current or upcoming testing, including ritonavir/saquinavir, nelfinavir/saquinavir, nelfinavir/indinavir, and ritonavir/nelfinavir [Mellors, un-numbered abstract]. The ritonavir/saquinavir combination has been discussed in previous issues of The Hopkins HIV Report (Vol. 8, Nos. 3 & 4) and is already widely used. In an Abbott-sponsored trial updated at this conference, over 80% of patients taking this combination (at doses of 400 or 600 mg bid of each drug) had undetectable viral loads at 24 weeks of therapy.
- Finally, we learned a lot more about using NNRTIs, especially nevirapine, because of the wealth of drug interaction data that were presented. Fears of drug interactions have kept us from using NNRTIs with protease inhibitors, but those fears may now be allayed. Nevirapine can be administered safely with ritonavir without the need for dose adjustment. Because the AUC of indinavir is decreased by 28%, clinicians should consider increasing the dose of indinavir to 1,000 mg q8h. Saquinavir's AUC is also reduced when used with nevirapine. Since the bioavailability of the currently available hard-gel formulation of saquinavir is already poor, the two drugs should probably not be combined unless the newer soft-gel formulation is being used, or unless the
bioavailability of saquinavir is being increased by coadministration with ritonavir [Murphy et al., Abstract 374]. In a trial of indinavir, nevirapine, and 3TC in a group of patients with advanced disease, combination therapy was associated with a 3 log reduction in viral load [Harris et al., Abstract 234].
Delavirdine has the opposite set of drug interactions. It increases drug levels of saquinavr and indinavir, while no effect is observed in combination with ritonavir. An indinavir dose of 400 mg q8h is equivalent to standard dose when combined with delavirdine [Cox et al., Abstract 372].
Like a dark cloud threatening to rain out a picnic, the specter of drug resistance reminds us that our highly effective therapy is only highly effective as long as it lasts. Resistance was a major focus of the Washington conference.
One of the most important lessons we learned at this meeting is that the viral nadir achieved after starting antiretroviral therapy is a stronger predictor of the durability of response than either baseline viral load or the magnitude of decline [Kempf et al., Abstract 373]. While we may be satisfied with "acceptable responses" when using nucleoside analog combinations, the goal of suppressive therapy using protease inhibitors must be to attain and maintain undetectable viral loads, preferably less than 200 copies/ml. Patients who don't achieve that goal are more likely to develop resistance and fail therapy.
Dr. Douglas Richman, in his review of antiretroviral resistance, stated that our goal has changed from contending with resistance to preventing it. He reiterated the now hackneyed but often unheeded maxim that "adding one drug to a failing regimen is substandard practice." He also made the more novel suggestion that drugs for which a single point mutation leads to high-level resistance (3TC and the NNRTIs) should be reserved for use in suppression-inducing regimens. In other words, by combining 3TC or nevirapine with protease inhibitors, their potent antiviral effect is maintained. When used with nucleoside analogs, however, resistance may occur rapidly and their utility will be short-lived. Finally, he stated that the goal of therapy is complete viral suppression, and that therapy should be started early when there is less genetic diversity among viral strains and less immunologic deterioration. As we know, compliance on the patient's part is critical; therefore, the effective use of antiretroviral therapy requires more of the clinician than writing the right prescription. Richman introduced the concept of the "robustness" of a drug regimen. Unlike potency, the antiretroviral efficacy of a regimen under optimal experimental conditions in highly motivated and compliant patients, robustness measures efficacy in the real world. It is a function not only of the potency of the regimen, but also of the prior antiretroviral experience of the patients and of the convenience and tolerability of the combination [Richman, Abstract S53].
In her review of new developments in antiretroviral therapy, Ann Collier closed with a quotation from Voltaire: "Perfection is attained by slow degrees. She requires the hand of time." At conferences like this one we may come away overwhelmed by the dizzying pace of our progress, but a few days later and a few steps back, we remember that perfection is a long way off.