1	FDA Analyses of Maraviroc Safety Data Scott Proestel, M.D. Division of Antiviral Products Food and Drug Administration April 24, 2007
2	Presentation Outline Overall Summary of AEs Mortality Malignancies Hepatic AEs Infections Additional AEs Selected Laboratory Data
3	Summary of AEs During Double-Blind Phase Studies 1027 and 1028
4	Duration of Therapy During Double-Blind Period Studies 1027 and 1028 Median Treatment Days MVC QD 236 MVC BID 239 Placebo 145
5	Days to Individual AEs During Double-Blind Period Placebo Arms, Studies 1027 and 1028
6	Days to Subject Discontinuation From the Double-Blind Period Placebo Arms, Studies 1027 and 1028
7	Mortality Data
8
9	Mortality Data Numerical increase in mortality in MVC arms overall, however: Imbalance occurred in only 1 of 4 trials Degree of imbalance extremely small (for double-blind phase, p-value 0.60, Barnard's unconditional exact test) No clustering of death causes Types of deaths were consistent with the population studied Sick population at baseline (there were 11 additional deaths between screening and randomization)
10	Malignancies
11	All Lymphomas by Treatment Arm Studies 1027 and 1028
12	Non-Hematologic Malignancies During Double-Blind Phase* Studies 1027 and 1028
13	Hepatic Adverse Events
14	Hepatic AEs
15	Hy’s Law Summary Hy’s Law AST or ALT >= 3x ULN TB >= 2x ULN No marked increase in alkaline phosphatase No evidence of another cause Predicts a 10-50% likelihood of death or liver transplant No Hy’s Law cases were observed
16	Case History The subject is a 43 year-old woman with history of injection drug use, alcoholism, and hepatitis C who received MVC for 203 days. On enrollment, she was HBsAg negative, HBcAb negative, HCV RNA positive, and reported drinking 5 alcoholic beverages per week. Her baseline AST and ALT were elevated at 101 and 43 IU/L. There were no substantive increases from these baseline LFT values through Day 169. Her baseline total bilirubin was normal at 1.1 mg/dL, but peaked at 5.2 mg/dL on Day 15. The bilirubin subsequently decreased from the peak level and was normal at the last measurement on Day 169. Of note, her total bilirubin was 2.3 mg/dL 41 days prior to starting MVC (peak direct and indirect bilirubin levels were abnormal at 3.3 and 1.9 mg/dL, respectively). She started atazanavir 537 days prior to MVC, and stopped on the day that MVC was started. She had 3 reported liver AEs during her study participation (grade 3 increase in bilirubin on two occasions, and grade 1 hepatosplenomegaly once).
17	AST and/or ALT Elevation During Double-Blind Period Studies 1027 and 1028
18	Total Bilirubin During Double-Blind Period Studies 1027 and 1028
19	Infectious Adverse Events
20	Most Common Infection-Related AEs During Double-Blind Period (Studies 1027 and 1028)
21	Category C AEs During Double-Blind Period Studies 1027 and 1028
22	Herpes Virus Infections During Double-Blind Period Studies 1027 and 1028 65 Herpes AEs MVC QD 25 AEs in 18 (4.3%) subjects MVC BID 32 AEs in 29 (6.8%) subjects Placebo 8 AEs in 8 (3.8%) subjects 19 Category C Herpes-related AEs MVC QD 11 AEs in 10 (2.4%) subjects MVC BID 6 AEs in 6 (1.4%) subjects Placebo 2 AEs in 2 (1.0%) subjects
23	Herpes Virus Infections During Double-Blind Period* Studies 1027 and 1028
24	Candidiasis Infections During Double-Blind Period Studies 1027 and 1028 70 Candidiasis AEs MVC QD 41 AEs in 26 (6.2%) subjects MVC BID 18 AEs in 17 (4.0%) subjects Placebo 11 AEs in 11 (5.3%) subjects 19 Category C Candidiasis AEs MVC QD 14 AEs in 12 (2.9%) subjects MVC BID 3 AEs in 3 (0.7%) subjects Placebo 2 AEs in 2 (1.0%) subjects
25	Candidiasis Infections During Double-Blind Period* Studies 1027 and 1028
26	Additional Adverse Events
27	Cardiovascular AEs During Double-Blind Period Studies 1027 and 1028
28	Cardiovascular Disorders at Baseline 95 patients with any cardiovascular disorder MVC QD 40 (10%) MVC BID 40 (9%) Placebo 15 (7%) 45 patients with any ischemic cardiac disorder MVC QD 21 (5%) MVC BID 17 (4%) Placebo 7 (3%)
29	Thrombotic AEs During Double-Blind Period* Studies 1027 and 1028
30	Postural Hypotension During Double-Blind Period Studies 1027 and 1028 Dose-limiting AE during phase 1 at doses of 600-1200 mg MedDRA terms: Hypotension, orthostatic hypotension, dizziness, syncope 109 Patients: MVC QD 51 (12%) MVC BID 39 (9%) Placebo 19 (9%)
31	AEs Potentially Associated with QT Prolongation During Double-Blind Period (Studies 1027 and 1028) Preclinical testing revealed potential for QT prolongation A previous CCR5 antagonist (“Schering C”) found to cause QT prolongation Thorough QT study for MVC determined to be inadequate No significant prolongation with MVC, but control arm was not interpretable Attempts to resolve this issue are underway MedDRA terms: Arrhythmia, Palpitations, Syncope, Tachycardia 14 patients with potential ventricular events: MVC QD 4 (1.0%) MVC BID 7 (1.6%) Placebo 3 (1.4%)
32	CPK Elevation AEs During Double-Blind Period Studies 1027 and 1028 MedDRA terms: Blood creatine phosphokinase increased, myositis, rhabdomyolysis 19 Subjects with AEs MVC QD 7 (1.7%) MVC BID 11 (2.6%) Placebo 1 (0.5%)
33	CPK Elevation by Treatment Group During Double-Blind Period (Studies 1027 and 1028)
34	Fasting Total Cholesterol During Double-Blind Period Studies 1027 and 1028
35	Fasting Total Cholesterol at Week 24 (+/- 2 Weeks) Studies 1027 and 1028
36	Fasting Total Cholesterol During Double-Blind Period Studies 1027 and 1028
37	LDL Cholesterol During Double-Blind Phase Studies 1027 and 1028
38	LDL Cholesterol During Double-Blind Phase Studies 1027 and 1028
39	Summary No increase in mortality or malignancy observed No clear evidence of hepatotoxicity, but this was a complicated patient population with a high rate of liver abnormalities at baseline. Infections No evidence of increase overall Possible increase in Candida, Herpes, and influenza infections Possible increase in cardiac ischemic events Possible increase in myositis Mild increase in total cholesterol and LDL levels
40	FDA Analyses of Maraviroc Efficacy Data
41
42	Primary Efficacy Endpoint (Week 24) Sensitivity Analyses Analysis 1: Using both on- and off- treatment HIV-1 VL data MVC-QD -Placebo: -0.50 (-0.90,-0.10) MVC-BID-Placebo: -0.49 (-0.90,-0.09) Analysis 2: Completers through Week 24 MVC-QD-Placebo: -0.48 (-0.93,-0.02) MVC-BID-Placebo: -0.54 (-0.99,-0.09) Analysis 3: Imputing a single value (-0.4 to 0.3 by 0.1) to missing Results are similar to the sponsor’s sensitivity analyses The imputed value is the median for the placebo MVC-QD-Placebo: -0.79~0.97 MVC-BID-Placebo: -0.88~1.06 All 99.95% CIs support the superiority of MVC
43	Efficacy at Week 24 by Baseline Demographics Studies 1027 and 1028
44	Efficacy at Week 24 by Disease Characteristics Studies 1027 and 1028
45	Change in CD4+ Cell Counts (Studies 1027 and 1028) Completers through Week 24 (n=672) n Mean (se) Treatment Effect (99.95% CI) MVC QD 283 137.8 (7.7) 41.1 (-5.0,87.3) MVC BID 296 123.6 (6.2) 26.9 (-18.9,66.3) Placebo 93 96.7 (10.8) LOCF through Week 24 (n=1033) n Mean (se) Treatment Effect (99.95% CI) MVC QD 408 113.5 (6.1) 58.8 (26.5,91.1) MVC BID 418 105.9 (5.1) 51.2 (21.0,81.4) Placebo 207 54.7 (7.0) Treatment effects are robust
46	Conclusions Evidence of Superiority Over Placebo Convincing Efficacy reduced by the most conservative analyses At least of 0.5 log₁₀ copies/mL Increase in CD4+ cell count observed Patients with OSS >= 3 without apparent benefit