New data suggest that in combination with other antiretroviral agents, the NNRTIs are potent inhibitors of viral activity. Studies presented at the World AIDS Conference in Geneva indicated that they are a viable choice either in first-line regimens or in treatment-experienced patients. However, many of the studies were unable to provide conclusive findings due to small numbers of participants and the fact that data were collected from chart reviews rather than controlled clinical trials. The manufacturers of the NNRTIs need to commit to undertaking larger, well-controlled clinical trials to obtain clear-cut data on durability of antiviral effect. Head-to-head comparisons of the NNRTIs to each other and to the protease inhibitors are also needed.
Nevirapine, manufactured by Roxane, a subsidiary of Boehringer Ingelheim Worldwide, and delavirdine, manufactured by Pharmacia and Upjohn, were the first two approved NNRTIs. Dupont's efavirenz was just approved in September (for more on this agent, see articles "What Price Efavirenz? Sparing the Protease for Some Spoils the Price for All" and "An Even Dozen: Another Expensive Drug for Your Cocktail" in this issue). The NNRTIs appear to be highly cross-resistant. A mutation at position 103 on the HIV reverse transcriptase gene confers resistance to each of the agents in the class. However, this mutation does not confer resistance to drugs in other classes.
A poster presentation in Geneva, by Patrick Robinson, M.D., of Boehringer Ingelheim, examined protease inhibitor sensitivity to 24 highly nevirapine-resistant strains of HIV using phenotypic assays (abstract 12231). Dr. Robinson and colleagues found that in the lab all nevirapine isolates were fully susceptible to the four FDA-approved protease inhibitors.
In addition, the investigators conducted a retrospective study of the clinical outcomes of 59 participants who had been treated with first-line nevirapine- and nucleoside analog-containing regimens and had subsequently switched to a protease inhibitor combination due to treatment failure, adverse events or study discontinuation. Median baseline CD4 count was about 150 cells and median viral load was about 40,000 copies/ml. Fifty of 51 evaluable participants (98%) experienced mean viral load decreases of 1.52 log (97%) after switching therapies. The authors conclude that first-line therapy with nevirapine and nucleoside analogs does not render patients unresponsive to subsequent protease inhibitor treatment.
A small study of delavirdine treatment failures had similar results. Michael Para, M.D., and colleagues, of Ohio State University, conducted a retrospective study of 18 protease inhibitor-naïve clinical trial participants who had used delavirdine-containing regimens and subsequently switched to a protease inhibitor and new nucleoside analog combination after treatment failure (abstract 12236). Median time on delavirdine was 19 months. At time of the switch median baseline CD4 count was 337 cells and median viral load was 38,331 copies/ml. After six months on the new regimen, 14 of the 18 participants (78%) achieved viral load levels below 400 copies/ml. These results suggest that first-line NNTRI-based regimens preserve future treatment options.
More data are emerging on the long-term efficacy of nevirapine plus nucleoside analogs in treatment-naïve populations. The 52-week INCAS Study compared twice-daily (bid) regimens of nevirapine/AZT/ddI, nevirapine/AZT and AZT/ddI in 151 participants in Italy, the Netherlands, Canada and Australia (J Montaner et al. JAMA. March 25, 1998; 279(12):930-7). Baseline demographics were similar across the three treatment groups for median CD4 count, ranging from 345 to 395 cells. However, median viral load was significantly lower in the triple combination arm, 17,732 copies/ml, compared to 32,163 copies/ml in the AZT/ddI arm and 48,654 copies/ml in the AZT/nevirapine arm.
Even with a lower baseline viral load, participants in the triple-combination group experienced a significantly greater reduction in viral load than did those in the double-combination arms. In an intent-to-treat analysis using the ultrasensitive PCR assay, the AZT/nevirapine group had a 0.9 log (87%) reduction in viral load; the AZT/ddI group had a 1.55 log (97%) drop; and the triple-combination group had a 2.18 log (99.3%) drop at 8 weeks that was basically sustained throughout the 52 weeks of the study. At week 52, 51% of participants on the triple-combination arm maintained viral loads below 20 copies/ml, compared to 12% in the AZT/ddI group and 0% in the nevirapine/AZT group. This is an impressive antiretroviral effect for the nevirapine/AZT/ddI group.
Treatment rash occurred in 22% of 98 patients in the nevirapine treatment groups and 4% developed severe rash. There were no episodes of Stevens Johnson syndrome (a serious, sometimes fatal, disorder that can be triggered by an allergic reaction to nevirapine). There were no reported cases of lipodystrophy.
In Geneva, Julio Montaner, M.D., of the University of British Columbia in Canada, presented data on 87 participants who completed the INCAS protocol and continued on a nevirapine-containing regimen for a total of 42 months through an open-label follow-up study, BI 1036 (abstract 12368). Thirty-four participants were from the original INCAS triple combination group. Twenty-one to 24 months after enrolling in BI 1036, 62% of this group (13 of 21) maintained viral suppression below 20 copies/ml. No significant new or unexpected adverse events were observed.
A variation on the regimen studied in INCAS substitutes d4T for AZT. This combination of d4T/ddI/nevirapine is also dosed on a twice-a-day schedule. An Australian group conducted a retrospective study of 39 patients "stepping down" to this combination from a protease inhibitor-containing regimen due to elevated viral load, problems with the dosing regimen or side effects (abstract 22357). Mean duration of protease inhibitor therapy was 57 weeks. Mean baseline CD4 count was 367 cells and mean viral load was 190,500 copies/ml.
Of those patients with viral loads below 400 copies/ml at time of switch, 73% remained below the limit of quantification through 26 weeks of treatment. Among the participants with viral loads above 400 copies/ml at time of switch, 54% went below the limit of quantification at some time during the 26 week follow-up period. Four participants who were naïve to all three drugs in the combination achieved viral loads below the limit of quantification. The authors concluded that for patients experiencing difficulties with protease inhibitors, "stepping down" to a more convenient regimen is supported by the findings of this study and warrants further investigation in controlled trials.
Martin Fisher, M.D., of the Royal Sussex Hospital in England, presented the results of a retrospective case review of 502 participants from the U.K. expanded-access program for nevirapine (abstract 32248). The study group was made up of 141 treatment-naïve participants and 361 participants who had experienced a median of 15 months of prior antiretroviral treatment, 124 of whom had used protease inhibitors.
Median baseline CD4 count was 217 cells and median baseline viral load was 69,000 copies/ml in the naïve group. For the 54 participants who completed six months of nevirapine-containing regimens, median viral load reduction was over 2 logs (99%) and over 70% achieved a viral load below the limit of quantification (unfortunately, the six centers involved in this study used different viral load assays so it is not possible to state a definitive lower limit of quantification).
Median baseline CD4 count was 200 cells and median baseline viral load was 28,183 copies/ml in the treatment-experienced group. At the time of initiation of nevirapine, 228 participants were able to commence at least one new antiretroviral agent. For the 157 participants who completed six months of nevirapine-containing regimens, the median viral load drop was about 0.8 logs (84%) and about 40% achieved a viral load below the limit of quantification.
Duncan Churchill, M.D., one of the investigators, stated that the study aimed to determine the efficacy of nevirapine in the "real world" of clinical practice as opposed to in the artificial atmosphere of a clinical trial. He concluded that the drug had performed well in this setting.
Nevirapine is an inducer of cytochrome P450 (a family of enzymes in the liver that metabolizes drugs), which could lead to an enhanced clearance rate and lower blood levels of other concurrently administered drugs that are metabolized by these enzymes, such as the protease inhibitors. There have been some small studies conducted on the interactions between nevirapine and the protease inhibitors.
Gail Skowron, M.D., of Brown University in Rhode Island, reported on the combination of d4T/nelfinavir/nevirapine in 24 NNRTI- and protease inhibitor-naïve participants (abstract 12275). Mean baseline CD4 count was 372 cells and median baseline viral load was 31,620 copies/ml. Dr. Skowron and colleagues found that there was no evidence of any interaction between the three agents when administered together at the standard dose of each drug. In a preliminary on-treatment analysis (which does not include dropouts) at least 82% of participants achieved viral load suppression below 400 copies/ml during weeks 21 to 45 (over 50% were below 50 copies/ml).
This regimen represents a "triple-class" combination because it includes a nucleoside analog, an NNRTI and a protease inhibitor. Although it proved quite potent in this study, there is some concern that using all available classes in a first-line regimen potentially leaves no options for salvage.
Robert Murphy, M.D., of Northwestern University in Chicago, examined the interaction between nevirapine and indinavir in a prospective open-label study of 24 protease inhibitor-naïve participants (abstract 22404). Nevirapine decreased indinavir plasma concentrations while indinavir had no significant effect on nevirapine plasma levels. The reduced indinavir plasma levels did not go below the threshold of antiviral activity. Dr. Murphy concluded, "the take home point is that you do not have to change the dose of indinavir when using nevirapine based on this study." (Many clinicians still use an increased dose of indinavir in combination with nevirapine.)
A late breaker presentation by Steven Deeks, M.D., of San Francisco General Hospital, reported impressive results from an open-label study comparing saquinavir/ritonavir/abacavir in combination with either a nucleoside analog (nucleoside analog group) or nevirapine (nevirapine group) in 20 indinavir-treatment failures (abstract LB 22490).
Participants were nelfinavir-, saquinavir- and/or NNRTI-naïve. The first ten patients in the nucleoside analog group had a median baseline CD4 count of 293 cells and a median baseline viral load of 31,620 copies/ml. At week 20, the median viral load reduction was 0.59 log (74%) with one of seven participants below 400 copies/ml. In contrast, the other ten participants in the nevirapine group, with median baseline CD4 count of 288 cells and median baseline viral load of 17,378 copies/ml, experienced a 2.67 log (99.8%) viral load decrease at week 20 and six of seven participants suppressed viral load to below 500 copies/ml. The use of nevirapine instead of a second nucleoside analog appeared to be strongly associated with a successful outcome in this patient population.
The possibility of a once-daily dosing schedule is extremely attractive. Because of its long half-life of 25-30 hours, nevirapine is a good candidate for this. Alex Dusek, of Boehringer Ingelheim, analyzed retrospective data from 123 participants in five clinical trials where nevirapine was administered at 400 mg once a day (abstract 12360). Participants were either taking nevirapine alone or in conjunction with nucleoside analogs. In this analysis, the nevirapine plasma levels achieved by once-daily dosing were 250-fold above the IC90 (the drug level required to block 90% of HIV replication) of wild type virus. The mean nevirapine trough levels of the participants in the once-a-day trials were comparable to those seen in nevirapine twice-a-day studies (INCAS and BI 1036). The adverse events in the once-daily group did not differ greatly from the twice-daily group. The authors suggest that 400 mg once-daily dosing, after a two-week 200 mg once-daily lead-in, should provide adequate drug exposure in combination regimens.
In fact, data from a triple-combination regimen administered on a once-a-day schedule were presented in Geneva. In addition to nevirapine, ddI and 3TC have the longest active intracellular half-lives of the approved antiretroviral agents (25 hours and 12 hours respectively). Annette Haberl, M.D., and colleagues, of Goethe University, Frankfurt, Germany, reported on the results of a prospective safety and efficacy trial of once-daily nevirapine/ddI/3TC in 70 IVDU participants (abstract 22398). Median baseline CD4 count was 247 cells and median viral load was 124,000 copies. At week 16, 60% of the 32 evaluable participants suppressed viral loads to below 500 copies/ml. At week 40, all of the 10 participants who had completed the study thus far had achieved viral loads below 500 copies/ml.
All study subjects were participating in a methadone maintenance treatment program. Where possible the antiretrovirals were administered at the same time as the methadone. Despite the comparative ease of the regimen, this was a difficult study population and there was a 47% drop-out rate. Because methadone is metabolized by the cytochrome P450 enzyme system, concurrent administration of nevirapine causes a decrease in methadone blood levels. Over 75% of participants required a mean methadone dose increase of 45%. Five participants discontinued the study due to methadone side effects. Boehringer Ingelheim is finally planning to conduct a formal methadone interaction study in the near future with investigators from this trial.
Delavirdine is not as easy a drug to take as nevirapine. It is administered on a thrice-daily (tid) schedule and should not be taken with many other drugs. When it was approved in April 1997, the data were unspectacular, but things are looking better a year later. Pharmacia and Upjohn conducted a study similar in design to INCAS called Protocol 0021 Part II (abstract 129/12219). The study compared a triple combination containing delavirdine and two nucleoside analogs, in this case AZT and 3TC, to two dual combinations, AZT/3TC or delavirdine/AZT, in 373 largely treatment-naïve participants (prior use of AZT for six months was permitted). Baseline demographics were similar across the three treatment groups. Mean baseline CD4 count ranged from 354 to 360 cells and mean baseline viral load ranged from 22,000 to 31,000 copies/ml.
As would be expected, the triple-combination arm did significantly better than the two dual-combination groups. In an interim, on-treatment analysis using the ultrasensitive assay, 75 participants in the triple-combination arm had completed 24 weeks and achieved a greater than 2 log (99%) reduction in viral load. This was sustained at 52 weeks in the 34 participants out to that point. At 24 weeks, 61.3% of the 75 participants from the triple-combination group suppressed their viral loads to below 40 copies/ml. This was sustained in about 60% of the 34 participants out to week 52. The main toxicity was rash, reported in 31% of participants in the triple-combination group (3% discontinued treatment). These results are very similar to those seen in the INCAS group and suggest that either nevirapine or delavirdine are viable choices for first-line protease-sparing regimens.
Delavirdine is a cytochrome P450 inhibitor, which could be advantageous by boosting protease inhibitor blood levels. Some small studies looked at regimens containing delavirdine/protease inhibitor combinations. Pharmacia and Upjohn Protocol 0063 compared delavirdine/indinavir/AZT using two reduced doses of indinavir (400 mg and 600 mg, three times a day) to AZT/3TC/indinavir at the standard doses in an open label study (abstract 22398). Mean baseline CD4 count was 377 cells and mean viral load was 199,500 copies/ml. The study enrolled 19 NNRTI-, protease inhibitor-, 3TC-naïve participants with a maximum of one month's prior use of AZT. All three arms experienced large decreases in viral load after 16 weeks on therapy.
Using the ultrasensitive assay, the delavirdine/400 mg indinavir/AZT group achieved a 2.7 log (99.8%) decrease but only about 18% of participants went below 40 copies/ml (83% went below 400 copies/ml). Participants in the delavirdine/600 mg indinavir/AZT group experienced a 2.3 log (99.5%) drop and 25% were below 40 copies/ml (50% below 400 copies/ml). In the AZT/3TC/indinavir group, there was also a 2.3 log (99.5%) decrease in viral load and 45% went below 40 copies/ml (71% below 400 copies/ml).
Although the indinavir doses were reduced in the delavirdine-containing arms, the plasma indinavir concentrations were equivalent to or greater than those seen with the standard 800 mg dose of indinavir without delavirdine. Pharmacia and Upjohn is now enrolling a new protocol, 0072, to look at twice-a-day dosing of the two drugs (600 mg delavirdine bid and 800 mg indinavir bid) in combination with AZT.
While delavirdine increases nelfinavir plasma levels, nelfinavir decreases delavirdine plasma levels. Protocol 0073 was another small open-label trial looking at delavirdine at the regular 400 mg dose and an increased 600 mg dose in combination with the standard doses of nelfinavir, ddI and d4T in 19 NNRTI-, protease inhibitor-, d4T-naïve participants with less than six month's use of ddI. Average baseline CD4 count was 354 cells and viral load was 138,000 copies/ml. After 24 weeks of therapy, viral load was reduced by over 2.8 logs (99.8%) in both arms and 57% to 63% of all participants had viral loads below 40 copies/ml.
Participants treated with 600 mg of delavirdine plus nelfinavir had delavirdine trough concentrations nearly equivalent to the standard 400 mg tid dose. Participants treated with 400 mg of delavirdine plus nelfinavir had delavirdine trough concentrations nearly equivalent to 300 mg tid. According to Pharmacia and Upjohn, 300 mg tid of delavirdine has similar antiviral activity to 400 mg so there is no need to increase the standard dose with nelfinavir. Concentrations of the active nelfinavir compound were 70% higher with both doses of delavirdine.
The tid-dosing schedule of delavirdine is problematic and numerous studies are underway to test various bid regimens. Protocol 0073 Part II is looking at delavirdine at 600 mg bid in combination with nelfinavir at 1250 mg bid plus ddI and/or d4T. Preliminary results in the first few participants demonstrate a greater than 2 log (99%) reduction in viral load across all treatment arms. ACTG 359 is studying bid dosing of delavirdine (or adefovir) in combination with saquinavir and nelfinavir or ritonavir in indinavir-experienced participants.
Larry Lyle, M.D., of San Diego, presented a retrospective analysis of 24 heavily pretreated patients, 75% of whom had previously used at least one protease inhibitor (abstract 12329). Nelfinavir/indinavir/delavirdine was administered in conjunction with 3TC and d4T in the majority of patients. Mean baseline viral load was 239 cells and mean baseline viral load was 17,780 copies/ml. Viral load decreased by 1.12 logs (92%) and 50% of patients went below 400 copies/ml after six months of the combination regimen. This study suggests that even for patients with extensive treatment histories, salvage therapy with drugs from each available class may be beneficial.