The FDA Recommends the Approval of Atazanavir, a New Protease Inhibitor
Also: BMS Lipid Studies on Atazanavir Presented to the FDA
June 10, 2003
The FDA Antiviral Drugs Advisory Committee met on May 13th from 8am to 5pm and voted unanimously 15-0 to recommend that the FDA approve Atazanavir. The FDA said that since BMS had provided extensive 48 week data the FDA will provide full traditional approval rather than just accelerated approval, if it agrees with the recommendation from the panel, which I presume they will.
Last night I sent out the first report from this very informative hearing and it should be posted and archived on the NATAP website at www.natap.org/2003/may/051403_2.htm if you have not seen it.
BMS reported that 1000 antiviral-naive subjects have been studied on Atazanavir (ATV) and over 500 patients treated for over 2 years. 500 antiviral treatment-experienced patients have been studied. 9 phase II/III studies have been conducted. An early access program was provided to 3600 patients.
Two studies were reported yesterday in treatment-experienced patients: ATV 400 mg compared to Kaletra in patients with 1 PI failure and Kaletra performed better; the second study compared once daily ATV 300 mg plus ritonavir booster of 100 mg compared to once daily ATV 400/saquinavir 1200 mg and Kaletra. After 24 weeks ATV 300 mg/RTV 100 mg appeared comparable in antiviral efficacy to Kaletra. But ATV 400 mg/SQV 1200 mg appeared inferior. The details and data were in yesterday's report.
In the phase III study comparing ATV+AZT/3TC to efavirenz+AZT/3TC, subjects in this trial generally were modestly immuno-compromised based on CD4 cell count. The median HIV RNA level and CD4 cell count for all treated subjects were 4.88 log10 c/mL and 282 cells/mm3, respectively, and were comparable between regimens. Forty-two percent of all treated subjects had baseline HIV RNA levels greater than 100,000 c/mL.
Comparisons of the treatment regimens within the subsets of baseline HIV RNA levels (<30,000 c/mL, 30,000-<100,000 c/mL, >100,000 c/mL) and CD4 cell counts (<200 cells/mm3, 200-<350 cells/mm3, >350 cells/mm3) were also consistent between treatment regimens, slightly favoring ATV. Apparently, regardless of CD4 count and if viral load was over 100,000 c/ml responses as evaluated by achieving <400 copies/ml were comparable between EFV and ATV. The data reported was that for patients with >100,000 copies/ml of viral load at baseline 61% of patients receiving EFV had had <400 copies/ml and 64% of patients receiving ATV (ITT) had <400 copies/ml. They did not report yesterday the percent of patients for each drug <50 copies/ml for patients with >100,000 copies/ml at baseline nor for patients with <200 CD4s.
BMS reported yesterday that patients who were receiving nelfinavir in study 034 could rollover into substituting ATV in study 044. After 24 weeks total cholesterol declined by 36 mg/dl, LDL-C declined by 27 mg/dl, non-HDL-C by 33 mg/dl, and triglycerides by 37. BMS said lipid returned to baseline levels after switching to ATV.
Atazanavir is a new drug with about 48 weeks of efficacy data and lipid data out as far as 108 weeks. With all new drugs additional safety and efficacy information is revealed after a few years following approval. The benefits of the drug are its once daily, low pill burden (2 pills once daily), it apparently does not raise lipids nor glucose. ATV can cause bilirubin elevations which appear not harmful so far in studies. Experts I have spoken with do not feel it is a problem medically but cosmetically it is an issue as some patients will develop jaundice and yellow eyes. One expert told me by laying in the sun or going under a sun lamp you can prevent the yellowing or get rid of it. One study of patients with lipid elevations on NFV saw reversal of lipid levels to baseline after switching to ATV. The cardiac effects of ATV are for prolongation of PR and QT intervals. Regarding QT intervals at the meeting yesterday it was said that it appears as if ATV has the same effect in this area as other protease inhibitors. It was said that some patients could be at greater risk if for example they had heart disease or were on medications that also effect QT intervals. For them it was suggested they might want to consider an EKG before starting therapy but opinions on the need for an EKG were mixed. As I reported yesterday studies show ATV was comparable antivirally to NFV and to EFV. Since this is a new drug more information will be revealed as doctors start prescribing it.
Here is information from BMS on studies exploring lipids and ATV presented to FDA Committee.
The results of five Phase II/III comparative studies conducted across ARV treatment-naive and treatment experienced individuals consistently demonstrated that ATV treatment, as part of a HAART regimen, resulted in significantly less hyperlipidemia and hypertriglyceridemia as compared with other PIs (eg, NFV, LPV/RTV) and as compared with the NNRTI, EFV. The favorable lipid and triglyceride profile for ATV was consistently demonstrated when ATV was combined with a variety of NRTI backbones (ZDV, d4T, ddI, 3TC, TDF). One study (AI424043) met the co-primary objective to demonstrate superior lipid parameters for ATV compared with the PI LPV/RTV as assessed by low density lipoprotein (LDL-cholesterol) measurements.
Another study (AI424008/44) confirmed the long term durability of achieving lower lipid concentrations with ATV treatment through 108 weeks and demonstrated a regression in hyperlipidemia and hypertriglyceridemia when NFV was switched to ATV. In addition, ATV demonstrated no clinically important effect on insulin/glucose metabolism. The magnitude of the difference between LDL-cholesterol concentrations observed for the population of subjects treated with ATV relative to comparator PIs and to the NNRTI EFV is clinically relevant based upon NCEP criteria. In the short-term, fewer ATV-treated patients meet requirements for lipid lowering interventions. The long-term benefits of avoiding hyperlipidemia have been established in the general population, and current clinical practice for HAART-treated patients presumes a similar association between hyperlipidemia and CV risk. For example, 18-30% of patients treated with PIs have a co-morbidity condition of hyperlipidemia, and 8% of patients initiate a lipid lowering therapy.
Results from three Phase II/III studies conducted in ARV treatment-naive HIV-infected subjects consistently demonstrated an improved lipid profile for ATV-treated subjects compared to NFV and EFV treated subjects as assessed by total cholesterol, fasting LDL-cholesterol, and fasting triglycerides changes from baseline.
Studies AI424007 and AI424008 were randomized studies comparing ATV regimens to NFV regimens in ARV treatment-naive subjects. Differences between ATV and comparator regimens were observed by Week 4 and continued throughout the treatment period. Analysis of mean percent changes from baseline indicated that subjects on the NFV treatment regimens had substantial increases in LDL-cholesterol by Week 12 and that these increases were sustained through 72 weeks. ATV-treated subjects experienced little or not increases in LDL-cholesterol. Based on NCEP, 80 - 87% of subjects had baseline LDL-cholesterol in the optimal or near optimal range (<130 mg/dL). However, at Week 72 the proportion of subjects who achieved LDL-cholesterol concentrations that were classified as high or very high ( 160 mg/dL) was greater on NFV compared with ATV (17% vs 5%).
Study AI424034 was randomized blinded Phase III study comparing ATV with EFV, each in combination with fixed dose ZDV. Through 48 weeks of treatment, ATV did not result in increases from baseline in fasting LDL-cholesterol, total cholesterol, or insulin, and resulted in a statistically significant decrease of 9% in fasting triglycerides and a statistically significant increase of 13% in HDL-cholesterol p <0.05 for comparisons to baseline. Baseline LDL-cholesterol values assessed by NCEP categories were comparable for ATV and EFV regimens. At 48 weeks, the proportion of ATV- treated fasting LDL-cholesterol concentrations outside the NCEP-defined desirable range was unchanged from baseline. Among ATV-treated subjects, LDL-cholesterol was >130 mg/dL in 13% of subjects at baseline and 13% at Week 48. LDL-cholesterol was >160 mg/dL in 2% and 3% of ATV-treated subjects at baseline and Week 48, respectively. In comparison, the proportion of EFV-treated subjects with LDL-cholesterol >130 or >160 mg/dL increased from baseline. Nineteen subjects (2%) were administered lipid reduction pharmacological therapy while on-study, five (1%) on ATV and 14 (3%) on EFV.
Long-term lipid benefits were maintained through Week 108 for those subjects who enrolled into Studies AI424041 and AI424044. The following fasting LDL-cholesterol analyses were conducted on the cohort of subjects who were treated on the extended dosing studies that followed completion of Studies AI424007 and AI424008.
AI424007/041 provided long term lipid data for subjects treated with ATV+ddI+d4T versus a NFV comparator. Through Week 108, subjects treated with the NFV regimen show substantially higher mean fasting LDL-cholesterol levels compared to ATV-treated subjects, who show no substantial mean increase from baseline. The ATV regimen was associated with a minimal mean percent increase from baseline in fasting LDL-cholesterol of 1% whereas, the NFV regimen was associated with a 22% increase from baseline.
AI424008/44 provided long term lipid data (108 weeks) for subjects continuing on treatment with ATV+d4T+3TC and demonstrated that after prolonged therapy (median 76 weeks) with another PI (NFV), reduction in serum lipid and triglyceride concentrations to pre-ARV levels was achieved within four weeks of the switch to ATV and sustained for 24 weeks following the switch.
Subjects who enrolled into AI424044 and were initially treated with NFV maintained an increase from baseline in fasting LDL-cholesterol of approximately 30% through the time they rolled over. Changes from baseline for the corresponding subjects treated with ATV were minimal through Week 108 (ATV 400, 7%; ATV 600, 8%).
Study AI424044 had a specific objective to assess changes in serum lipid concentrations associated with a switch from NFV to ATV. The sample size had sufficient statistical power to detect a mean decrease of 15% or greater in total cholesterol after 12 weeks of therapy. The 12-week endpoint was selected based on data from previous Phase II studies (AI424007 and AI424008) that showed ARV associated dyslipidemia can be detected as early as Week 4 after the start of HAART therapy. The mean time on NFV prior to a switch to ATV was 76 weeks. The mean time on ATV in Study AI424044 was 38 weeks among the 63 subjects switching from NFV to ATV. The aggregate time on ATV for those subjects continuing ATV from AI424008 was approximately 108 weeks. At the Week 12 endpoint, significant differences were observed for the mean percent change from entry in total cholesterol (-16%), HDL-cholesterol (+5%), fasting LDL-cholesterol (-21%), and fasting triglycerides (-28%) (Table 7.1.4). For all lipid parameters except HDL-cholesterol, these statistically significant differences were maintained through Week 24. Median Week 12 levels of total cholesterol (165 mg/dL), fasting LDL-cholesterol (94 mg/dL), and fasting triglycerides (86 mg/dL) in this cohort were comparable to median baseline levels in Study AI424008 (168 mg/dL, 91 mg/dL, and 93 mg/dL, respectively). Additionally, more than half of subjects in the NFV-ATV cohort who entered AI424044 with borderline high or high LDL-cholesterol no longer met these criteria by Week 12 (LDL-cholesterol >130 mg/dL in 55% at entry vs 22% at Week 12; LDL-cholesterol >160 mg/dL in 27% at entry vs 10% at Week 12).
Lipids in Treatment-Experienced Studies
Results in treatment-experienced HIV-infected subjects demonstrated a favorable lipid profile for ATV-treated subjects compared to RTV/SQV and LPV/RTV-treated subjects, as assessed by changes in total cholesterol, LDL-cholesterol, and fasting triglycerides.
Study AI424009 was a pilot study conducted in 85 ARV treatment-experienced subjects. At Week 72 on Study AI424009, the mean change in LDL-cholesterol for RTV/SQV treated subjects was 8% compared to a mean decrease of 2% for ATV 400/SQV and 30% for ATV 600/SQV-treated subjects. While the mean change in serum triglycerides for RTV/SQV-treated subjects was 95% compared to 27% for ATV 400/SQV-treated subjects and a decrease of 15% for ATV 600/SQV-treated subjects.
Study AI424043 was a randomized, open-label, study in ARV treatment-experienced subjects comparing ATV with LPV/RTV, each in combination with two nucleosides selected on the basis of resistance testing. In Study AI424043, the co-primary objective was to compare the magnitude of change in LDL-cholesterol at Week 24. The magnitude of changes in total cholesterol, HDL-cholesterol, glucose, and fasting triglycerides through Week 24 were analyzed as secondary study objectives. The ATV-containing treatment regimen was associated with a decrease from baseline in fasting LDL-cholesterol (6%), whereas LPV/RTV was associated with an increase from baseline (8%) at Week 24 The superior fasting LDL- cholesterol levels for ATV compared with LPV/RTV were demonstrated by comparing the mean percent change from baseline in fasting LDL-cholesterol (-14.2%, 97.5% CI: - 23.0%, -5.4%, nominal p <0.0001).
Atazanavir treatment was also associated with small decreases from baseline in total cholesterol (2%) and fasting triglycerides (2%) through 24 weeks. In contrast, at 24 weeks LPV/RTV was associated with substantial increases from baseline in total cholesterol (18%) and fasting triglycerides (57%). Mean percent changes from baseline were statistically superior for ATV for total cholesterol (-18.1%, 95% CI: -23.4%, -12.9%; p <0.0001) and fasting triglycerides (-36.1%, 95% CI: -46.7%, -25.5%, p <0.0001). The differences in LDL-cholesterol, total cholesterol, and fasting triglyceride levels were observed for ATV-treated subjects within four weeks of beginning therapy and persisted through Week 24. HDL-cholesterol increased for both treatment regimens, with slightly greater increases on the LPV/RTV treatment regimen. As a result, the use of serum lipid reducing agents was less common on the ATV treatment regimen (5%) than on the LPV/RTV treatment regimen (18%).
In the highly treatment-experienced population (AI424045), both ATV-containing regimens were associated with superior lipid parameters compared to LPV/RTV, as assessed by the mean percent change from baseline at Week 16 in total cholesterol (ATV/RTV: -7%; ATV/SQV: -10%; LPV/RTV: +5%) and fasting triglycerides (ATV/RTV: 2%; ATV/SQV: -15%; LPV/RTV: +34%). Mean percent changes from baseline in fasting LDL-cholesterol at Week 16 for ATV 400/SQV were superior compared to LPV/RTV (ATV/SQV: -10%; LPV/RTV: +1%). At Week 16, fasting LDL-cholesterol concentrations outside the desirable range (values >130 mg/dL) were comparable across the three treatment regimens. Thirty-eight subjects (11%) took serum lipid reduction therapy while on study, 7% on ATV 300/RTV, 12% on ATV 400/SQV, and 14% on LPV/RTV. More importantly, 3% of subjects each on ATV 300/RTV and ATV 400/SQV and 8% of subjects on LPV/RTV initiated serum lipid reduction therapy while on study.
Thank you to NATAP for the privilege of posting this article on our Ezine.
This article was provided by Seattle Treatment Education Project. It is a part of the publication STEP Ezine.