1. Comparison of NRTI combinations  ZDV/3TC vs TDF + FTC –Study 934  ABC/3TC vs TDF/FTC –HEAT Study –ACTG A5202 Study –ASSERT Study  FTC/TDF vs FTC/TAF –Studies GS-US-292-0104 and GS-US-292-0111

2. HEAT Study: ABC/3TC vs TDF/FTC  Design Smith KY. AIDS 2009;23:1547-56 ABC/3TC + TDF/FTC placeboQD LPV/r 800/200 mgQD  Objective –Non inferiority of the 2 fixed dose NRTI combinations at W48: % HIV RNA < 50 c/mL, ITT-exposed, missing = failure [ITT-E, M = F] (lower margin of the 95% CI for the difference = - 12%, 90% power) –Primary safety endpoint: incidence of adverse events at W96 Randomisation* 1 : 1 Double-blind placebo-matched 694 ARV-naïve patients > 18 years HIV RNA > 1,000 c/mL Any CD4 cell count No HLA-B*5701 screening W48W96 TDF/FTC + ABC/3TC placeboQD LPV/r 800/200 mgQD *Randomisation was stratified on HIV RNA 100,000 c/mL HEAT N = 345 N = 343

3. HEAT Study: ABC/3TC vs TDF/FTC Randomized ABC/3TC N = 347 TDF/FTC N = 347 Treated eligible patients, N343345 Median age, years38 Female16%20% White/Black/Other52% / 36% / 13%51% / 36% / 13% HIV RNA (log 10 c/mL), median4.94.8 HIV RNA > 100,000 c/mL45%41% CD4 cell count (/mm 3 ), median214193 CD4 < 200/mm 3 47%52% CD4 < 50/mm 3 18%20% HBV positive/HCV positive6% / 8%3% / 7% Discontinuation by W96N = 109 (32%)N = 124 (36%) For virologic failure/adverse event, N 8 / 206 / 21 Lost to follow-up/subject decision/non compliance, N 45 / 13 / 1052 / 23 / 11 Note: change of NRTI (to NRTI other than ABC or TDF) allowed if intolerance; change of LPV/r QD to BID allowed if gastrointestinal intolerance, or to other PI if LPV/r-limiting intolerance. LPV/r was administered as soft-gel capsules (6/d) to week 48 then as tablets (4/d) Patient disposition and baseline characteristics HEAT Smith KY. AIDS 2009;23:1547-56

4. Proportion of patients with HIV RNA < 50 c/mL at week 48 Median CD4 increase at W96: 250/mm 3 (ABC/3TC) vs 247/mm 3 (TDF/FTC) HEAT Study: ABC/3TC vs TDF/FTC HEAT ITT-E, M/D = FTLOVR % ABC/3TCTDF/FTC 95% CI for the difference = - 6.6; 7.4 Primary efficacy endpoint ITT, M/D = F Observed analysis, ITT-E ITT-E, M/D = F stratified by baseline HIV RNA (c/mL) < 100 000 > 100 000 68 63 64 84 71 63 67 61 62 87 69 65 0 20 40 60 80 100 343N =345343345343345188205155140 Smith KY. AIDS 2009;23:1547-56 ITT-E, M = F: ITT-exposed, missing/discontinuation = failure

5. HEAT Study: ABC/3TC vs TDF/FTC Safety and tolerability (median exposure = 96 weeks) * Including suspected ABC HSR (N = 14), immune reconstitution syndrome (N = 2), hepatotoxicity (N = 1) ** Including suspected ABC HSR (N = 3), renal failure (N = 2), decreased creatinine renal clearance (N = 1) ABC/3TC N = 343 TDF/FTC N = 345 Grade 2-4 drug related adverse events Any50% 46% Diarrhoea 19% Nausea8% 6% Increased triglycerides6% Increased cholesterol7% 4% Decreased Glomerular Filtration Rate 5% Suspected Hypersensitivity Reaction to ABC 3%<1% Any drug-related serious adverse events N = 18 * (5%)N = 10 ** (3%) Discontinuation for adverse event6% HEAT Smith KY. AIDS 2009;23:1547-56

6. HEAT Study: ABC/3TC vs TDF/FTC Change in laboratory parameters (lipids, renal, biomarkers) Median change from baseline at W96ABC/3TCTDF/FTC Total cholesterol (mg/dL)+ 36+ 28 HDL-cholesterol (mg/dL)+ 10+ 12 Total cholesterol: HDL-cholesterol ratio- 0.27- 0.44 LDL-cholesterol (mg/dL)+ 9+ 8 Triglycerides (mg/dL)+ 54+ 42 GFR, MDRD equation (mL/min/1.73 m 2 )00 Proximal renal tubular dysfunction occurrence N = 0N = 5 (1%) Biomarkers (% change from baseline)W48W96W48W96 sVCAM-1- 49- 51- 48- 50 IL-6- 26- 19- 23- 25 hs-CRP- 12- 5- 20- 17 HEAT Smith KY. AIDS 2009;23:1547-56

7. HEAT Study: ABC/3TC vs TDF/FTC  Conclusions –As initial antiretroviral regimens, ABC/3TC and TDF/FTC, each in combination with LPV/r QD, have the same efficacy rate –HIV RNA responses by baseline HIV RNA strata ( 100,000 c/mL) were similar between groups at W48 and W96 –Rate of virologic failure was similar in both groups (14%) –CD4 response at W96 was similar in the 2 groups –Both treatments were well tolerated More gastrointestinal intolerance with TDF/FTC More lipid abnormalities with ABC/3TC –Of note, rate of discontinuation was high (34% at W96) Smith KY. AIDS 2009;23:1547-56