NDA 21567 Atazanavir

1
NDA 21-567 Atazanavir

• Kendall A. Marcus, M.D.
• Medical Reviewer
• Division of Antiviral Drug Products

2
Presentation Outline

• NDA Submission Overview
• Efficacy Summary - Tom Hammerstrom, Ph.D.
• Clinical Virology - Lisa Naeger, Ph.D.
• Safety Issues
• Hyperbilirubinemia
• Lipid Profiles
• Effects on the QT and PR Interval
• Conclusions

3
NDA Overview

• Submission Date December 20, 2002
• Proposed Dosage Atazanavir 400 mg once daily
• Proposed Indication Treatment of HIV infection

4
Phase 2 Dose-Finding Studies Treatment Naïve
Patients

• AI424-007 N420
• ATV 200 mg, 400 mg, and 500 mg
• NFV 750 mg tid
• Each given with d4T/ddI.
• AI424-008 N467
• ATV 400 mg and 600 mg
• NFV 1250 mg bid
• Each given with d4T/3TC.

5
Phase 3 Studies

• AI424-034 Treatment-naïve subjects (n810)
• ATV 400 mg daily
• EFV 600 mg daily
• Each given with AZT/3TC (Combivir?)
• AI424-043 Subjects failing PI based regimens
  (n300)
• ATV 400 mg daily
• LPV/RTV bid
• Optimized background of 2 NRTIs based on
  phenotypic testing.

6
Phase 3 Studies

• AI424-045 Patients failing at least 2 regimens
  containing drugs from all three classes (n358)
  
• ATV 300 mg/RTV 100 mg once daily
• ATV 400 mg/SQV 1200 mg once daily
• LPV/RTV bid
• Background therapy of tenofovir and one NRTI.
• 16 week data on roughly 33 patients/arm
  submitted with initial NDA submission.

7
Other Supportive Studies

• AI424-041 and AI424-044
• Rollover studies for Phase 2
• PACTG 1020-A
• PK and safety study in infants, children, and
  adolescents
• AI424-900 - Expanded access protocol
• AI424-009 (N85)
• Small Phase 2 study of treatment experienced
  patients comparing ATV/SQV to RTV/SQV

8

• Atazanavir Resistance
• Lisa K. Naeger, Ph.D.
• Antiviral Advisory Committee Meeting
• May 13, 2003

9
In vitro Selection
3 Different Strains of HIV-1 were serially
selected with ATV for 4-5 months (200 - 500 nM)
Virus strain mutations RF V32I,
L33F, M46I, A71V, I84V, N88S LAI
L10Y/F, I50L, L63P, A71V, N88S NL4-3
V32I, M46I, I84V, L89M
Fold ATV Resistance 183 93 96
10
I50L Mutation

• ATV resistance corresponded to the presence of
  I50L and A71V in the protease of recombinant
  viruses from 8 clinical isolates.
• 2- to 17-fold decreases in ATV susceptibility
  were observed in viruses containing the I50L and
  A71V mutation

11
I50L Mutation

• Viruses containing the I50L mutation either alone
  or in combination with A71V remained susceptible
  to APV, IDV, NFV, and RTV.
• Insertion of the I50L substitution into HIV-1
  resulted in replication impaired viruses. The
  addition of the A71V change with I50L restores
  some viability.

12
ATV Clinical Resistance Analyses

• Mutations Associated with ATV-Resistance
• Phenotypic and genotypic analyses of evaluable
  clinical isolates from patients on ATV-containing
  regimens who experienced virologic failure or
  discontinued before suppression from studies 007,
  008, 034, 009 and 043
• Baseline Phenotype and Genotype Analysis
• Cross-Resistance

13
Evaluable Clinical Isolates from Patients on
ATV-Containing Regimens who Experienced Virologic
Failure or Discontinued before Suppression
14
Mutations Associated with ATV-Resistance in
Naïve-trials

• 14 ATV-resistant clinical isolates
• 11 (79) developed the I50L mutation
• Median 9-fold change in ATV resistance
• 7 of these 11 also developed the A71V mutation
• Development ranged from 2 to 80 weeks
• (mean 40 weeks)

15
Phenotype of ATV-Resistant Isolates that
Developed the I50L Mutation In Naïve Trials
Average fold-change from reference strain
16
Mutations Associated with ATV-Resistance in
Experienced trials

• 32 ATV-resistant clinical isolates
• ATV Treatment (21)
• 5 isolates developed the A71V or T mutation
• 2 isolates developed the I84V mutation (5-fold
  change from BL)
• 2 developed the N88S or D mutation (4-fold change
  from BL)

17
Mutations Associated with ATV-Resistance in
Experienced trials

• ATV/SQV Treatment (11)
• 5 isolates developed the I84V mutation (14-fold
  change from BL)
• 4 isolates developed the A71V or T mutation
• 2 isolates developed the L90M mutation
• 2 isolates developed the M46I mutation

18
Cross-Resistance of the Virologic Failure
Clinical Isolates that were ATV-Resistant from
Treatment- Experienced Patients in Trials 009
and 043
(N 32)
19
Baseline Analysis
20
74
56
20
21

• 24 of the Isolates from
• 009 and 043 showed ATV-Resistance at Baseline

22
100
62
59
Resistance
47
43
23
Response Based on Baseline Genotype
24
(No Transcript)
25
Cross-Resistance
26
Cross-resistance of HIV-1 Clinical Isolates by
Phenotype
27
Cross-Resistance of HIV-1 Clinical Isolates by
Genotype
28
ATV Resistance Against PI-Resistant Clinical
Isolates (n 551)
29
ATV Susceptibility Against PI-Resistant Clinical
Isolates (n 551)
30
ATV Resistance Summary

• I50L mutation is specific for ATV resistance and
  is the predominant mutation developing in
  antiretroviral therapy-naïve patients
• Viruses with the I50L mutation remain susceptible
  to other PIs
• Mutations L90M, I84V, N88S/D and A71V/T appear to
  confer ATV resistance and reduce the clinical
  response to ATV
• There is a clear trend toward ATV resistance as
  isolates become resistant to three or more PIs

31
Atazanavir Efficacy Results

• Thomas Hammerstrom, PhD
• Division of Antiviral Drug Products

32
Phase II and III Clinical Trials

• Pivotal Phase III Trial - 34
• Endpoint Percent Sustained lt 400 copies/mL to
  Week 48, Time Averaged Difference from Baseline
  (TAD)
• ART Naïve Population
• Control is Efavirenz
• Phase II Trials - 7 and 8
• Endpoint Percent lt 400, TAD
• ART Naïve Population
• Control is Nelfinavir

33
Phase II and III Clinical Trials

• Pivotal Phase III Trial - 43
• Endpoint TAD
• ART Experienced Population
• Control is Kaletra

34
Results in Trials with ART Naïve Subjects
35
Percent Sustained lt400 copies/mL to Week 48

• Trial Arm lt400 95 Interval ATV - Con
• 34 EFV 251 / 405 62
• ATV 271 / 405 67 -1.5, 11.6
• 7 NFV 62 / 103 60
• ATV 62 / 103 60 -13.8, 13.0
• 8 NFV 54 / 91 63
• ATV 121 / 181 69 -5.0, 19.4

36
Results in Trial with ART Experienced Subjects
37
Trial 43, ATV vs Kaletrafor 24 Weeks

• Endpoint Arm Estimate 95 Interval
• lt400 KAL 98 / 150 65
• ATV 70/150 47 -30, -7.9
• TAD KAL -1.65
• ATV -1.39 .078, .44

38
Is ATV Better than Placebo ART Experienced
Subjects?

• Two Meta-analysis Methods
• 1. Calculate Difference of ATV and Placebo from
  Trial 43 and Kaletra Trials
• 2. Compare Confidence Intervals for ATV 2
  NRTIs in Trial 43 with Confidence Intervals for
  2 NRTIs alone from other NDAs

39
ATV vs Placebo lt400

• Source Arm Estimate DIFF SEE
• Trial 43 ATV 70/150 47 KAL 98 / 150
  65 -19 5.73
• Trial 863 KAL 259 / 326 79
• NFV 233 / 327 71 8 3.36
• Trial 511 NFV 66/ 99 67
• PLA 7/ 101 7 60 5.37
• Imputed ATV
• PLAC 49 8.54

40
ATV vs Placebo lt400

• Source Arm Estimate DIFF SEE
• Trial 43 ATV 70/150 47 KAL 98 / 150
  65 -19 5.73
• Trial 888 KAL 84 / 148 57
• S. PI 46 / 140 33 24 5.69
• Imputed ATV
• Selected PI 5 8.07

41
ATV vs Control lt400

• Source Control Difference 95 Interval
• Trial 43 KAL -19 -30, -7.9
• Imputed Sel PI 5 -10.8, 21
• Discounted Sel PI 4.5 -12.3, 23
• Imputed PIacebo 49 32, 66

42
(No Transcript)
43
ATV vs PlaceboTAD, Week 24

• Source Arm Estimate DIFF SEE
• Trial 43 ATV -1.39 KAL -1.65 .26 .093
• Trial 863 KAL -1.798
• NFV -1.801 .003 .057
• Trial 511 NFV -1.77
• PLA -1.40 -.37 .083
• Imputed ATV
• PLAC -.107 .137

44
ATV vs PlaceboTAD, Week 24

• Source Arm Estimate DIFF SEE
• Trial 43 ATV -1.39 KAL -1.65 .26 .093
• Trial 888 KAL -.972
• S. PI -.867 -.104 .078
• Imputed ATV
• Selected PI .156 .121

45
ATV vs ControlTAD, Week 24

• Source Control Difference 95 Interval
• Trial 43 KAL .26 .078, .44
• Imputed Sel PI .156 -.081, .393
• Imputed Placebo -.107 -.376, .162

46
(No Transcript)
47
Efficacy Conclusions

• 1. Equal or Better than NFV or EFV on lt400 at
  week 48 in 3 Trials with Naïve Subjects
• 2. 95 Lower Limits on lt400 no more than 5
  worse than NFV or EFV in 2 out of 3 Trials

48
Efficacy Conclusions

• 3. Equal or Better than NFV or EFV on TAD at
  week 48 in 2 out of 3 Trials with Naïve Subjects
• 4. 95 Upper Limits on TAD no more than .28 log
  copies worse than NFV or EFV in all 3 Trials

49
Efficacy Conclusions

• 5. Statistically Significantly Worse than Kaletra
  on both lt400 and TAD at week 24 in 1 Trial with
  Experienced Subjects
• 6. Indirect Imputations Support for Efficacy on
  Primary Endpoint, Ambiguity on Secondary Endpoint

50
Efficacy Conclusions

• 6. Indirectly shown at least 33 better than
  placebo, no more than 10 worse than selected PI
  on lt400
• 7. 95 Confidence Limits on lt400 higher than
  limits seen on all 2 drug combinations in
  previous NDAs

51
Efficacy Conclusions

• 8. Indirectly shown no more than .16 log copies
  worse than Placebo on TAD at week 24
• 9. 95 Confidence Limits on TAD comparable to
  limits seen on several 2 drug combinations in
  previous NDAs

52
Safety Issues

• Hyperbilirubinemia

53
HyperbilirubinemiaToxicity Grading Scale

• Total bilirubin toxicity grading scale
• Grade 1 1.1 1.5 x ULN
• Grade 2 1.6 2.5 x ULN
• Grade 3 2.6 5.0 x ULN
• Grade 4 gt 5.0 x ULN
• Upper limit of normal for total bilirubin ? 1.0 -
  1.5 mg/dL
• Upper limit of normal for direct bilirubin ? 0.2
  - 0.5 mg/dL

54
Studies 007 008HyperbilirubinemiaDose
Dependence
55
Percentage of Subjects with HyperbilirubinemiaAta
zanavir - 400mg

56
Incidence of Jaundice and Scleral Icterus Phase
2 and 3 Clinical Studies

• Atazanavir - 400 mg

Three patients in 034 (1) and 2 patients in 043
(1) discontinued for jaundice or scleral icterus
without grade 4 hyperbilirubinemia.
57
Grade 4 Hyperbilirubinemia and Dose Reduction

• Atazanavir - 400 mg

58
Study 034 Mean Total and Direct Bilirubin
Atazanavir - 400 mg
59
Total Bilirubin gt 10 mg/dL

• Occurred in 10 patients across clinical trials
• 4/10 transient, predominantly unconjugated (DB ?
  0.3 mg/dl)
• 1/10 diagnosed w/SHL, DB - 0.6 mg/dl
• 5/10 also with other LFT abnormalities
• 4 with viral hepatitis (A, B, C)
• One had a total bilirubin gt 10 mg/dl prior to
  randomization which resolved prior to treatment
  and then worsened temporarily on study.

60
LFT Abnormalities - ATV versus NFV Background
ddI/d4T
61
LFT Abnormalities ATV versus EFV Background
AZT/3TC
62
Discontinuations Due to Hepatotoxicity/Abnormal
LFTsAll Studies

• Atazanavir - 15/1596 (1)
• Ten had chronic hepatitis B or C
• One had acute hepatitis B
• One had history of hepatic steatosis
• Three subjects with no apparent risk factors
• Comparators - 8/892 (1)
• Five had chronic hepatitis B or C
• One had acute hepatitis B
• One was hep B core Ab positive but surface Ab and
  antigen negative
• One on RTV/SQV w/ no apparent risk factors

63
HyperbilirubinemiaConclusions

• Inhibition of UGT 1A1
• Predominantly unconjugated
• Reversible upon discontinuation of atazanavir
• Jaundice/scleral icterus are likely to be common
  adverse events in clinical practice resulting in
  more frequent discontinuations than seen in
  clinical trials.
• Risk for hepatotoxicity similar to other marketed
  ARV

64
Lipid Profiles
65
Study 034 Lipid Profiles at Week 48Percent
Change from Baseline
Percent Change from Baseline
66
Study 034 Lipid Profiles at Baseline and Week 48
TCgt240
LDLgt160
Percentage of Subjects
TGgt400
67
Study 043Lipid Profiles at Week 24Percent
Change from Baseline
Percent Change from Baseline
68
Study 043 Lipid Profiles at Baseline and 24 Weeks
TCgt240
TGgt400
mg/dL
LDLgt160
69
(No Transcript)
70
LipodystrophyTreatment-Naïve Studies
Includes events of lipoatrophy,
lipohypertrophy, and lipodystrophy.
71
CV Events -Myocardial Infarction

• Three MIs in atazanavir-treated patients and
  three MIs in patients receiving comparators
• One subject receiving RTV/SQV underwent three
  vessel bypass surgery

72
Lipid ProfilesConclusions

• Lipid effects of atazanavir appeared to persist
  through 108 weeks of treatment, although data
  from phase 2 trials is limited by study design.
• Benefits for treatment-experienced patients less
  well defined as factors other than current
  protease inhibitor use appear to contribute at
  least to hypertriglyceridemia.
• Lipid effects do not appear to be associated with
  a reduced incidence of lipodystrophy.
• Cardiovascular benefit is unknown at this time.

73
Evaluation of the QT Interval
74
In Vitro Evaluation of Potential Cardiac Effects

• Modest inhibition of IKr (HERG) - 15 at 30 µM
• Moderate inhibition of Ca channels - IC50 of 10
  µM
• Weak inhibition of Na channels - IC50 gt 30 µM
• In Purkinje fiber studies, a dose-dependent
  increase in mean action potential duration was
  observed.

75
Study 076

• 3-treatment, 3-period crossover study
• 72 subjects received multiple, once-daily doses
  of atazanavir
• Subjects assigned atazanavir 400 mg, 800 mg, and
  placebo in six different sequences
• Washout period of ? 14 days

76
QTc Changes by Correction Formula

• ? QTcB at Tmax from baseline to 800 mg dose is
  7.9 msec
• (95 CI 2.8, 12.9)
• ? QTcF at Tmax from baseline to 800 mg dose is
  -1.6 msec
• (95 CI -4.2, 1.1)

77
Studies 034 and 043Prolonged QTc Intervals

• Study 034
• Incidence of prolonged QTc intervals similar
  between atazanavir and efavirenz regimens (2).
• One subject receiving efavirenz had a QTc
  interval gt 500 msec.
• Study 043
• Nine subjects (ATV, 2 subjects LPV/RTV, 7
  subjects) experienced a post-baseline QTc
  prolongation.

78
Study 034 and 043 CV Events Potentially Related
to Arrhythmia

• Sudden death or torsades de pointes
• CV events leading to treatment discontinuation
• CV events coded as SAEs
• Grade 3 - 4 CV events
• All CV events
• Events reviewed - No events of sudden death,
  torsades de pointes, events suspicious for TdP,
  or an imbalance between treatment arms in events
  potentially attributable to TdP was observed.

79
Effect of Atazanavir on the QT IntervalConclusion
s

• Data from placebo-controlled study 076 limited by
  lack of positive control (e.g. moxifloxacin)
• Current data indicates that atazanavir has little
  or no effect on the QT interval however, the
  overall risk is unknown
• No signal for increased risk relative to
  comparators was identified in clinical trials

80
Evaluation of the PR Interval
81
Causes of PR Interval Prolongation and AV Block

• Medications - e.g. antihypertensives, digoxin
• Fibrosis of the conduction system
• Ischemic heart disease
• Valvular or congenital heart disease
• Cardiomyopathy
• Myocarditis

82
First Degree AV BlockClinical Significance

• ACC/AHA/NASPE 2002 Guidelines for Implantation of
  Cardiac Pacemakers and Antiarrythmic Devices
• Class II B recommendation
• First degree AV block greater than 300 msec in
  patients with LV dysfunction and symptoms of
  congestive heart failure in whom a shorter AV
  interval results in hemodynamic improvement

83
Study 076 (Healthy Volunteers)Plot of Mean PR
Versus Time Since Dosing
84
Study 034 PR Interval

• Time Mean PR Interval
• Post-dose (min, max)
• msec
• EFV 2-3 hours 153 (101, 251)
• ATV 2-3 hours 160 (110, 287)
• Maximum PR intervals 265 - 307

85
Study 034 First Degree AV Block

• Treatment Incidence of First Degree AV Block
  ()
• Males Females Total
• Efavirenz 3
  0 2
• Atazanavir 4 6
  4.5

86
Study 041 PR Interval

• Time Mean PR Interval
• Post-dose (min, max)
• msec
• NFV 2-3 hours 158 (97, 203)
• ATV 2-3 hours 164 (120, 243)
• Maximum PR intervals 234 - 250

87
Studies 041 and 044First Degree AV Block

• Study Dose N 1st Degree
• AV Block ()
• 041 ATV 400 148 5
• NFV 47 9
• 044 ATV 400 172 9
• ATV 600 127 14

88
Study 043 PR Interval

• Time Mean PR Interval
• Post-dose (min, max)
• msec
• LPV/r 2-3 hours 157 (114, 250)
• ATV 2-3 hours 157 (114, 209)
• Maximum PR Intervals 194 - 218

89
Study 043 First Degree AV Block

• Treatment Incidence of First Degree AV Block
  ()
• Males Females Total
• Lopinavir/r 6 4
  6
• Atazanavir 7 3
  6

90
Case Narratives - Other Conduction Abnormalities
Potentially Related to Atazanavir

• On day 1053 of therapy in study 007/041, a 43
  year old male intentionally ingested a large
  number of ATV/3TC/d4T pills
• Received activated charcoal
• Severely prolonged PR interval with bifascicular
  block was observed on ECG
• Patient was monitored for 5 days until ECG
  normalized

91
Case Narratives - Other Conduction Abnormalities
Potentially Related to Atazanavir

• A 50 year old male with HTN was hospitalized on
  day 11 of ATV/ 3TC/DLV/TDF for angina and SOB
• On verapamil SR for HTN
• An ECG showed junctional rhythm with retrograde
  atrial activation. ARV medications held
• One day following admission an ECG showed
  persistence of junctional rhythm
• Two days following admission the patient was
  found unresponsive with idioventricular rhythm
• Autopsy showed 90 LAD without infarct

92
Effects on PR Interval Conclusions

• Dose dependent prolongation of the PR interval
• First degree AV block most common abnormality
  observed
• Incidence of first degree block appears to be
  similar to that observed with selected PIs
• Severe prolongation (gt 300 msec) or more serious
  events appear to be rare

93
Pediatric ProtocolPACTG 1020-A

• Atazanavir 2 NRTIs
• 48 enrolled 29 continuing on study
• Adverse event profile in these patients appears
  similar to adults
• Due to wide variability of pK data in all age
  cohorts, a dose has not yet been defined for any
  group

94
Drug-Drug Interactions
95
Drug-Drug Interactions

• Potential interaction with ATV
• CYP3A inhibitor, inducer, or substrate
• drugs that increase pH
• drugs that cause PR prolongation
• 2C9 (e.g. warfarin) or 1A2 (e.g. theophyline)
  -not studied

96
Drug Interactions - Diltiazem

• Diltiazem
• CYP 3A substrate inhibitor
• PR prolongation
• ATV ? diltiazem Cmax and AUC 100

97
Drug Interactions - Oral Contraceptives

• Co-administration of atazanavir and ethinyl
  estradiol/norethindrone (Ortho-Novum 7/7/7?) was
  evaluated
• Cmax AUC
• Ethinyl estradiol 1.15 1.48
• Norethindrone 1.67 2.10

98
AtazanavirOverall Conclusions

• Antiviral activity similar to efavirenz or
  nelfinavir in treatment-naive patients
• Inferior to lopinavir/r in treatment-experienced
  patients, but multiple analyses indicate activity
  in this population
• Low pill burden may enhance compliance in
  selected patients
• Unique resistance pathway in treatment-naïve
  subjects

99
AtazanavirOverall Conclusions

• Hyperbilirubinemia appears to be due to
  inhibition of UGT 1A1 and reversible with
  treatment discontinuation
• Risk for hepatotoxicity appears to fall within
  the range of that seen with other ARV medications

100
AtazanavirOverall Conclusions

• Dose-dependent prolongation of the PR interval
• Incidence of first degree AV block appears to be
  similar to that observed in lopinavir/ritonavir
  and nelfinavir treated patients
• Clinically significant events due to prolongation
  of the PR interval appear to be rare
• Effects of atazanavir on the QT interval appear
  to be minimal

101
AtazanavirOverall Conclusions

• Favorable lipid profile as compared to selected
  protease inhibitors and efavirenz
• Impact on cardiovascular events unknown
• Does not appear to reverse triglyceride
  elevations seen in treatment-experienced patients
• Does not appear to result in a decreased
  incidence of lipodystrophy

102
Questions
103
Question 1

• Do the efficacy and safety of atazanavir support
  its approval for the treatment of HIV infection?
  As part of your discussion please comment on
• Treatment effects seen in na?ve and experienced
  patients
• Hyperbilirubinemia observed in clinical trials
• The effect of atazanavir on the PR and QT
  intervals

104
Question 2

• If atazanavir is recommended for approval, does
  its safety profile warrant additional clinical or
  laboratory monitoring?

105
Question 3

• Does the effect of atazanavir on lipid parameters
  offer patients a clinically significant advantage
  over other treatment options?

106
Question 4

• Based on the resistance data, what
  recommendations would you have regarding its use
  in naïve and experienced patients?

107
Question 5

• Please provide recommendations for any Phase 4
  studies of atazanavir?