1. Pooled 24-week results of DUET-1 and DUET-2: TMC125 (etravirine; ETR) versus placebo in treatment-experienced HIV-1-infected patients P Cahn, R Haubrich, J Leider, G Pialoux, M Schechter, S Walmsley, J Vingerhoets, M Peeters, G De Smedt, MP de Béthune and B Woodfall on behalf of the DUET-1 and DUET-2 study groups

2. 24-week primary analysis DUET-1 and -2 differed only in geographical location; pooled analysis was pre-specified Plasma viral load >5,000 HIV-1 RNA copies/mL and stable therapy for ≥8 weeks ≥1 NNRTI RAM, at screening or in documented historical genotype ≥3 primary PI mutations at screening Patients recruited from Thailand, Australia, Europe and the Americas Screening 6 weeks 600 patients target per trial 48-week treatment period with optional 48-week extension *BR = darunavir/ritonavir with optimised NRTIs and optional enfuvirtide TMC125 + BR* Placebo + BR* Follow up 4 weeks DUET study design and major inclusion criteria BR = background regimen; RAM = resistance-associated mutation

3. Baseline characteristics and background ARVs *from extended NNRTI RAM list (Tambuyzer et al. Abstract 67 EHDRW 2007); § assessed by phenotypic sensitivity score (PSS); BR = background regimen; RAM = resistance-associated mutation Parameter, % or median (range) TMC125 + BR (n=599) Placebo + BR (n=604) Patient demographics Male (%) 9089 Caucasian (%)70 Disease characteristics Viral load (log 10 copies/mL)4.8 (2.7–6.8)4.8 (2.2–6.5) CD4 cells (cells/mL)99 (1.0–789)109 (0.0–912) CDC category C (%)5859 Prior ARV use 10–15 ARVs (%)6664 Darunavir/r (%)45 Detectable mutations ≥2 NNRTI RAMs* (%)66 ≥4 primary PI RAMs (%)6263 Background regimen Used enfuvirtide (total; %)4647 Used enfuvirtide de novo (%)2627 Active background agents § = 0 (%)1716 Active background agents § = 1 (%)3639

4. Patients with viral load <50 copies/mL at Week 24 (primary endpoint; ITT TLOVR) BR = background regimen; CI = confidence interval; ITT = intent-to-treat population; TLOVR = time to loss of virologic response imputation algorithm 59% 41% p<0.0001 100 90 80 70 60 50 40 30 20 10 0 Responders (%) ± 95% CI 04812162024 Time (weeks) TMC125 + BR (n=599) Placebo + BR (n=604)

5. Viral load reduction from baseline (ITT NC=F) BR = background regimen; SE = standard error; ITT = intent-to-treat population; NC=F = noncompleter equals failure imputation algorithm; changes below the detection limit (<50 copies/mL) were imputed as 49 copies/mL –1.7 –2.4 Mean change in viral load from baseline (log 10 copies/mL) ± SE 0.0 –0.5 –1.0 –1.5 –2.0 –2.5 –3.0 –3.5 Time (weeks) 04812162024 p<0.0001 TMC125 + BR (n=599) Placebo + BR (n=604)

6. Change in CD4 cell count from baseline (ITT NC=F) +86 +67 Mean change in CD4 cell count from baseline (cells/mm 3 ) ± SE Time (weeks) p<0.0001 BR = background regimen; SE = standard error; ITT = intent-to-treat population; NC=F = noncompleter equals failure imputation algorithm 100 75 50 25 0 04812162024 TMC125 + BR (n=599) Placebo + BR (n=604)

7. Response (<50 copies/mL) according to enfuvirtide use (primary analysis) Patients with viral load <50 copies/mL at Week 24 (%) p<0.0001 Using de novo ENF DRV FC category, % TMC125 + BR Placebo + BR <2 2027 2–104039 10–4026 >40158 Placebo + BR (n=604)TMC125 + BR (n=599) 67% Using de novo ENF p=0.427 34% 56% Unadjusted response rates 0 20 40 60 80 Re-using or not using ENF 62% DRV = darunavir FC = baseline fold change Using de novo ENF p<0.05 62% 73% Adjusted for differences in baseline DRV FC between groups

8. Response (<50 copies/mL) according to number of active background ARVs Patients with viral load <50 copies/mL at Week 24 (%) Number of fully active background ARVs (PSS) 45% 60% 8%a ≥ 30% 67% 74% Analysis excludes patients who discontinued except for virological failure; PSS = phenotypic sensitivity score; darunavir and enfuvirtide are counted as fully active if FC<10 or used de novo, respectively 7/91 40/88 63/211 120/199 171/257 191/258 Placebo + BR (n=559)TMC125 + BR (n=545)

9. Mutations associated with a decreased response to TMC125 13 baseline NNRTI RAMs were associated with a decreased response to TMC125 (TMC125 RAMs): V90I A98G L100I K101E/P V106IV179D/F Y181C/I/VG190A/S These RAMs seldom occurred in the absence of other NNRTI RAMs K103N is not associated with resistance to TMC125 A decreased response was defined as ≤75% of the response for patients with zero NNRTI RAMs at baseline from the extended NNRTI RAM list; RAM = resistance-associated mutation

10. Response (<50 copies/mL) according to number of TMC125 RAMS The greatest added benefit in the TMC125 versus placebo group was seen in patients with <3 TMC125 RAMs 86% of patients had <3 TMC125 RAMs Patients with viral load <50 copies/mL at Week 24 (%) Number of TMC125 RAMs present at baseline 75% 58% 60% 41% Analysis excludes patients who used de-novo enfuvirtide or discontinued except for virological failure BR = background regimen; RAM = resistance-associated mutation; ≥ 73/121 59/157 38% 25% 44% Placebo + BR (n=414)TMC125 + BR (n=406) 02040 6080 25% 17% 121/161 64/147 37/64 17/68 13/32 6/24 7/28 3/18

11. Proportion of patients with any new AIDS-defining illness or death Patients with any new AIDS- defining illness or death (%) 3.7% 6.8% p=0.4419 Overall population 22/59941/604 0 2 4 6 8 10 Placebo + BR (n=604)TMC125 + BR (n=599) BR = background regimen

12. Proportion of patients with any new AIDS-defining illness or death ENF = enfuvirtide; BR = background regimen Patients with any new AIDS- defining illness or death (%) 3.7% 6.8% p=0.4419 Placebo + BR (n=604)TMC125 + BR (n=599) Overall populationRe-using or not using ENF 3.8% 8.3% p=0.0051 22/59941/60417/44637/444 0 2 4 6 8 10

13. Overview of adverse events (AEs; regardless of causality) *no deaths in the TMC125 group were considered at least possibly related to trial medication; ‡ in >10% patients in either group, excluding injection site reactions; § p=0.0001 vs placebo There were no consistent or clinically relevant trends in laboratory, vital signs or ECG data The profile of laboratory abnormalities, including hepatic and lipid parameters, was generally similar between the TMC125 and placebo groups Parameter, % TMC125 + BR (n=599) Placebo + BR (n=604) Any AE (any cause) 93 Grade 3/4 AE 2527 Discontinuation due to AE 5.84.5 Serious AE 1319 Death (any cause), % (n) 1.3 (8)*2.5 (15) Most common AEs ‡ Rash (any type) 17 § 9 Diarrhea 1520 Nausea1411 Headache912 AEs of interest Nervous system disorders 1519 Psychiatric disorders 1315 Hepatic AEs 55

14. Summary of rash in the TMC125 group Overall incidence: 17% in TMC125 group versus 9% in placebo group (p=0.0001) Early onset: most frequent in 2 nd week of therapy; median onset Day 12 Short duration: median duration 11 days Usually mild to moderate severity: 1.3% grade 3 and no grade 4 events – no rashes with mucosal involvement Infrequently lead to discontinuation – 2.2% of patients permanently discontinued – most self-limiting with continued treatment Higher incidence in women, but no clear difference in severity or treatment discontinuations according to gender No association with baseline CD4 cell count No increased risk in patients with a history of NNRTI-related rash

15. Conclusions In treatment-experienced patients, including those with NNRTI resistant virus, TMC125 consistently demonstrated superiority over placebo – 59% of patients achieved confirmed undetectable VL (<50 copies/mL) with TMC125 plus BR at Week 24 Even in the absence of any other fully active background agents, with TMC125, 45% of patients achieved undetectable (<50 copies/mL) viral load – response rates increased as more active agents were used in the background regimen 13 TMC125 resistance-associated mutations (TMC125 RAMs) were identified – the greatest added benefit in the TMC125 versus placebo group was seen in patients with <3 TMC125 RAMs – 86% patients had <3 TMC125 RAMs Except for rash, incidence and severity of AEs with TMC125 were similar to placebo TMC125 has the ability to extend and enhance the NNRTI class and provide a new treatment option for patients with resistance to other NNRTIs

16. Acknowledgements Argentina: H Ariza, J Benetucci, L Calanni, L Cassetti, J Corral, D David, A Krolewiecki, M Losso, P Patterson, R Teijeiro; Brazil: C A da Cunha, B Grinsztejn, E Kallas, JV Madruga, E Netto, J H Pilotto, J Suleiman, A Timerman; Chile: J Ballesteros, R Northland; Costa Rica: A Alvilés Montoya, G Herrera Martinez, A Solano Chinchilla; France: M Dupon, C Katlama, J M Livrozet, P Morlat, C Piketty, I Poizot-Martin; Mexico: J Andrade-Villanueva, G Reyes-Terán, J Sierra-Madero; Panama: A Canton, A Rodriguez, N Sosa; Puerto Rico: J O Morales Ramirez, J L Santana Bagur, R Soto-Malave; Thailand: T Anekthananon, P Mootsikapun, K Ruxrungtham; USA: M Albrecht, N Bellos, R Bolan, P Brachman, C Brinson, F Cruickshank, R Elion, W J Fessel, T Hawkins, S Hodder, T Jefferson, H Katner, C Kinder, M Kozal, J Lalezari, D McDonough, A Mills, K Mounzer, D Norris, W O’Brien, G Pierone, K Raben, B Rashbaum, M Rawlings, B Rodwick, P Ruane, J Sampson, S Schrader, A Scribner, M Sension, D Sweet, B Wade, D Wheeler, A Wilkin, T Wilkin, T Wills, M Wohlfeiler, K Workowski. Australia: J Chuah, D Cooper, B Eu, J Hoy, C Workman; Belgium: N Clumeck, R Colebunders, M Moutschen; Canada: J Gill, K Gough, P Junod, D Kilby, J Montaner, A Rachlis, B Trottier, C M Tsoukas; France: C Arvieux, L Cotte, J F Delfraissy, P M Girard, C Katlama, B Marchou, J M Molina, D Vittecoq, Y Yazdanpanah, P Yeni; Germany: K Arasteh, S Esser, G Fätkenheuer, H Gellermann, K Göbels, F D Goebel, H Jäger, A Moll, J K Rockstroh, D Schuster, S Staszewski, A Stoehr; Italy: A Antinori, G Carosi, G Di Perri, R Esposito, A Lazzarin, F Mazzotta, G Pagano, E Raise, S Rusconi, L Sighinolfi, F Suter; Netherlands: P H J Frissen, J M Prins, B J A Rijnders; Poland: A Horban; Portugal: F Antunes, M Miranda, J Vera; Spain: B Clotet, P Domingo, G Garcia, J M Gatell, J González-Lahoz, J López- Aldeguer, D Podzamczer; UK: P Easterbrook, M Fisher, M Johnson, C Orkin, E Wilkins; USA: B Barnett, J Baxter, G Beatty, D Berger, C Borkert, T Campbell, C Cohen, M Conant, J Ernst, C Farthing, T File, M Frank, J E Gallant, A E Greenberg, C Hicks, D T Jayaweera, S Kerkar, N Markowitz, C Martorell, C McDonald, D McMahon, M Mogyoros, R A Myers Jr, G Richmond, K Sathasivam, S Schneider, H Schrager, P Shalit, F P Siegal, L Sloan, K Smith, S Smith, P Tebas, L S Tkatch, W Towner. We express our gratitude to the patients that participated in the study, as well as the study centre staff, DSMB, clinical event adjudication panel, Virco, Tibotec personnel and the following principal investigators: DUET-1 DUET-2