1. HIV that Is Resistant to Dolutegravir May not be Transmissible Mark A Wainberg McGill University AIDS Centre McGill University AIDS Centre Jewish General Hospital

2. Disclosures I have received honoraria from AbbVie, Bristol Myers Squib, Gilead, Janssen,Merck, ViiV

3. Palella F, et al. NEJM, 1998. 1994199519961997 Use of protease inhibitors Deaths 0 10 20 30 40 Deaths per 100 Person-Years 0 20 40 60 80 100 Therapy with a Protease Inhibitor (% of patient-days)

4. Patient Substitutions de codon associées à une pharmacorésistance Mutations - INTIMutations - INNTIMutations - IP 1 41L; 67N; 69N; 70R; 74V; 184V; 215F; 219Q 100I; 103N 10I; 36I; 54V; 63P; 71V; 73S; 82V; 90M 2 41L; 184V; 215Y103N, 179E 48V; 63P; 71V; 73S; 77I; 82A; 90M 3 Aucune103N 10I; 54V; 63P; 71V; 82T; 84V; 90M 4 184V; 215Y103N77I 5 184V108I20R, 77I 6 41L; 67N; 210W; 215YAucune63P; 71V; 73S; 90M 7 41L; 215Y101EAucune 8 41L; 215Y101EAucune Brenner, B. et coll., AIDS, 18(12), le 20 août 2004, p. 1653–1660. INTI : Inhibiteurs nucléosidiques de la transcriptase inverse INNTI : Inhibiteurs non nucléosidiques de la transcriptase inverse IP : Inhibiteurs de la protéase Exemples de transmission de virus multirésistants dans les cas d’une primo-infection par le VIH-1

5. Transmitted NNRTI resistance rising in the post-HAART era

6. Cohen C, et al. 6th IAS 2011; Abstract TULBPE032 ECHO and THRIVE double-blind study designs 48 weeks primary analysis 96 weeks final analysis N=690 patients RPV 25mg qd + TDF/FTC + EFV placebo qd (N=346) EFV 600mg qd + TDF/FTC + RPV placebo qd (N=344) ECHO (TMC278-C209) 1:1 *Investigator’s choice: TDF/FTC; AZT/3TC; ABC/3TC N=678 patients RPV 25mg qd + 2 N(t)RTIs* + EFV placebo qd (N=340) EFV 600mg qd + 2 N(t)RTIs* + RPV placebo qd (N=338) THRIVE (TMC278-C215) 1:1 RPV was non-inferior to EFV in confirmed response (viral load <50 copies/mL, ITT-TLOVR) at Week 48 (primary objective) 1 RPV had a more favourable safety profile but higher virologic failure rate than EFV 1 RPV is approved in the US as a single-agent tablet 2 and a qd fixed-dose, single-tablet regimen with TDF/FTC is under development 3 1 Cohen CJ, et al. XVIIIth IAC 2010. Abstract THLBB206 2 FDA label for EDURANT TM (rilpivirine) tablets. 2011 3 Mathias A, et al. XVIIIth IAC 2010. Abstract LBPE17 ITT = intent-to-treat; TLOVR = time-to-loss of virologic response

7. Rilpivirine RPV vs EFV

8. Cohen C, et al. 6th IAS 2011; Abstract TULBPE032 ≤100K Patients (%) RPV Responders Discontinued due to other reasons † VF eff Discontinued due to AE/death EFV 4.0 (–1.7, 9.7) § 8480 RPV N’=368 EFV N’=329 >100K Patients (%) 75 RPV N’=318 EFV N’=353 -5.2 (-12, 1.5) § 70 Non responders Pooled ECHO and THRIVE: ITT-TLOVR outcome at Week 96 by baseline VL Responses by baseline CD4 cell count were (≥200 cells/mm 3 ): RPV 82% vs EFV 79%, (≥50–<200 cells/mm 3 ): RPV 71% vs EFV 75% and (<50 cells/mm 3 ): RPV 56% vs EFV 69%

9. Cohen C, et al. 6th IAS 2011; Abstract TULBPE032 Pooled ECHO and THRIVE: summary of resistance findings 1 Tambuyzer L et al. Antivir Ther 2009;14:103–9 2 Johnson VA et al. Top HIV Med 2009;17:138–45 * One VF res occurred after Week 96 in RPV group (E138K, K219E, M184I); † At least one emergent NNRTI RAM (from the NNRTI RAM list) 1 or IAS-USA N(t)RTI 2 RAM RPV N=686 EFV N=682 Time of failure All* Up to Week 48 Week 48to 96* All Up to Week 48 Week 48 to 96 VF res with resistance data, n 866718422814 No emergent NNRTI 1 and N(t)RTI RAMs 2, n (%) 35 (41)24 (36)11 (61)19 (45)11 (39)8 (57) Any emergent NNRTI 1 or/and N(t)RTI RAMs 2, n (%) 51 (59)43 (64)7 (39)23 (55)17 (61)6 (43) Any emergent † NNRTI RAMs 2, n (%) 46 (53)39 (58)6 (33)20 (48)16 (57)4 (29) Most frequent NNRTI RAM, n (%) E138K 31 (36) E138K 27 (40) E138K 3 (17) K103N 14 (33) K103N 11 (39) K103N 3 (21) Any emergent † N(t)RTI RAMs 1, n (%) 48 (56)41 (61)6(33)11 (26)9 (32)2 (14) Most frequent N(t)RTI RAM, n (%) M184I 32 (37) M184I 27 (40) M184I 4 (22) M184V 6 (14) M184V 6 (21) M184I 2 (14)

10. Compensatory effect of E138K on the replication capacity (RC) of M184I/V. The RC of both E138K and M184I are each decreased by 3-fold compared to wild-type and decreased by 2-fold for M184V. There is no difference in RC of double mutants E138K/M184I or E138K/M184Vc compared to wild-type.

11. The N348I mutation in RT Prevents the Occurrence of the E138K NNRTI Mutation.

13. ‡ 13 STaR: CPA vs. ATR – Week 96 Study Design Multicenter, international, randomized, open-label, Phase 3b, 96- week study RPV/FTC/TDF STR EFV/FTC/TDF STR ARV-naive HIV-1 RNA ≥2500 c/mL Sensitivity to EFV, FTC, RPV, TDF (N=786) Stratified by HIV RNA (≤ or >100,000 c/mL) n=394 n=392 96 Weeks 48 Weeks Primary Endpoint Primary Endpoint:Efficacy of the 2 STRs by proportion with HIV ‑ 1 RNA <50 c/mL at Week 48 (FDA Snapshot analysis); non-inferiority margin of 12% Secondary Endpoints:Safety and efficacy of the 2 STRs by proportion with HIV ‑ 1 RNA <50 c/mL at Week 96 (FDA snapshot analysis) Change in CD4 cell count at Weeks 48 and 96 Genotype/phenotype resistance at time of virologic failure 13

14. ‡ 14 STaR: CPA vs. ATR – Week 96 EPA Maintains Non-Inferior Virologic Suppression Compared to ATR Through 96 Weeks 4.1-1.19.2 W48W96W48W96W48W96 -0.65.511.5 W96 W48 0 -12%12% Mean CD4 count change (cells/mm 3 ): Week 48: RPV/FTC/TDF +200 vs. EFV/FTC/TDF +191 (p=0.37) Week 96: RPV/FTC/TDF +278 vs. EFV/FTC/TDF +259 (p=0.17) Favours EFV/FTC/TDF Favours RPV/FTC/TDF

15. ‡ 15 STaR: CPA vs. ATR – Week 96 Virologic Failure Stratified by Baseline HIV-1 RNA * Post hoc analyses; analyses for non-inferiority only pre-specified for ≤100,000 c/mL and >100,000 c/mL Baseline HIV-1 RNA, c/mL Virologic failure definition: Week 48 or 96 HIV-1 RNA > 50 c/mL, or discontinued study drug due to lack of efficacy or other reasons and last available HIV-1 RNA >50 c/mL

20. 14th European AIDS Conference; October 16-19, 2013; Brussels, Belgium FLAMINGO (ING114915) Study Design Clotet et al. EACS 2013; Brussels, Belgium. Abstract LBPS4/6. Primary endpoint: proportion with HIV-1 RNA <50 c/mL at Week 48, FDA Snapshot analysis, -12% non-inferiority (NI) margin Secondary endpoints: antiviral activity, safety, tolerability, health outcomes and viral resistance HIV+ ART-naive VL ≥1,000 c/mL Stratified by screening plasma HIV-1 RNA (≤ vs >100,000 c/mL) and background dual NRTI (ABC/3TC or TDF/FTC*) Week 96 analysis Randomization Week 48 analysis DRV/r 800 mg/100 mg QD + 2 NRTIs DTG + ART Open-label randomized phase DTG 50 mg QD + 2 NRTIs DTG 50 mg QD + 2 NRTIs Extension phase *Investigator selected backbone of choice

21. 14th European AIDS Conference; October 16-19, 2013; Brussels, Belgium Results were confirmed in per protocol analysis: 91% DTG versus 84% DRV/r, ∆ (CI): 7.4 (1.4 - 13.3) Proportion (95% CI) of Individuals With HIV-1 RNA <50 c/mL Over Time – Snapshot Clotet et al. EACS 2013; Brussels, Belgium. Abstract LBPS4/6. DTG: 90% DRV/r: 83% Week BL481216362448 Proportion (%) 95% CI for difference a Favours DRV/r Favours DTG -20%020% 7.10.913.2 -12% Test for superiority: P=0.025

22. Resistance to INSTIs in clinical trials in treatment-naïve patients Treatment Major resistance mutations detected by genotyping in treatment-naïve patients failing therapy Minor resistance mutations Raltegravir Y143 N155H Q148 Multiple Elvitegravir T66I E92Q N155H Q148 Multiple DolutegravirNONE RALTEGRAVIR Cooper et al., NEJM, 2008 Sichtig et al, JAC, 2009 Canducci et al, AIDS, 2009 Hatano et al, JAIDS, 2010 ELVITEGRAVIR Sax et al, Lancet, 2012 DeJesus et al, Lancet, 2012 DOLUTEGRAVIR vanLunzen et al., Lancet Infect. Dis., 2012

23. Major resistance pathways against INSTIs (clinical and tissue culture data) Resistance pathways Fold resistance RALEVGDTG Y143 pathway Y143C<10<2 Y143R<50<2 T97A/Y143C>100<2 T97A/Y143R>100<2 L74M/T97A/Y143G<50ND<2 L74M/T97A/E138A/Y143C<20ND<2 N155 pathway N155H<50 <2 E92Q/N155H<100>100<10 L74M/N155H<50 <2 Q148 pathway Q148H<20<10<2 Q148K<100 <2 Q148R<50<100<2 E138K/Q148H<10<20<2 E138K/Q148K>100 <20 E138K/Q148R>100 <10 G140S/Q148H>100 <20 G140S/Q148K<10<100<2 G140S/Q148R>100 <10 E138A/G140S/Y143H/Q148H>100ND<50 Quashie et al., Curr. Opin. Infect. Diseases, in press

24. Secondary INSTI-resistance mutations often restore HIV replication capacity Mbisa et al., Infect. and drug resistance, 2011--Canducci et al., JAC, 2010--Reigadas et al., Plos One, 2010--Delelis et al., AAC, 2009 Secondary Mutations (pathway) Effect on viral fitness in the presence of primary resistance mutations Y143 pathway- (often)  L74M, T97A+ N155H pathway-  Q95K, T97Q, G163R/K + Q148 pathway-  G140A/S/C, E138K/A+

25. Dolutegravir activity on RAL-resistant clinical isolates (n=39) (median IC 50 for wild-type=1.07 nM) GenotypeMedian fold change N155H1.37 Y143R/T97A1.05 Q148H/G140S3.75 Q148R/G140S13.3 Underwood et al., JAIDS, 2012

26. Subtype-specific mutations selected in vitro with dolutegravir HIV-1 subtype Most common mutations selected with dolutegravir BR263K, H51Y CG118R, H51Y Quashie, Mesplède et al., Journal of Virology, 2012

27. The R263K mutation confers low-level resistance to dolutegravir in cell culture GenotypeIC 50 fold change* R263K2.5 to 6 *Methodological differences (EC 50 for wild-type ≈1-6nM) Quashie, Mesplède et al., Journal of Virology, 2012

28. SAILING CROI 2013. A study in which Dolutegravir was shown to be superior to Raltegravir in treatment-experienced integrase inhibitor- naïve subjects. The R263K mutation was present in two individuals who either rebounded or did not achieve virologic suppression to <50 c/ml.

30. The R263K mutation decreases integrase activity in cell-free assays Quashie, Mesplède et al., Journal of Virology, 2012

31. The R263K mutation decreases dolutegravir residency time in an integrase-vDNA complex

32. The addition of H51Y to R263K further decreases IN strand transfer activity A B

33. The combination of H51Y and R263K negatively impacts viral fitness

34. Dolutegravir resistance associates with a decrease in viral replication capacity GenotypeResistance Effect on viral fitness R263K+- H51YNone H51Y/R263K++- G118R+- H51Y/G118R++-

35. Virus Weeks 8-15 WTM184V H51YM184I R263KM184I H51Y/R263KNone G118RNone H51Y/G118RNone

39. The R263K Mutation Confers a Higher Level of Drug Resistance against DTG than INSTI Mutations Associated with RAL and EVG Mutations at positions Fold resistance to RALEVGDTG E92Q2-3.515-20<2 Y1433-161.5-2<1.5 Q14810-5010-100<1.2 N155H5-1525-50<1.2 R263K<12-44-5 This explains why R263K is selected preferentially by DTG and why the R263K virus is then unable to proceed along any of the alternative INSTI resistance pathways that are associated with high level resistance against all members of the INSTI family of drugs

40. Replication Capacity of HIV Containing Various Combinations of INSTI Resistance Mutations Mutation(s)% fitness E92Q≈ 75% Y143≈ 72% Q148≈ 75% N155≈ 75% R263K≈ 70% G140/Q148≈ 95% R263K/H51Y≈ 25% R263K/E138K≈ 25% R263K/Q148R0 R263K/Y143C0 R263K/E92Q0

42. Conclusions Resistance mutations selected in vitro with dolutegravir are: R263K or G118R plus H51Y R263K and G118R confer low-level resistance against dolutegravir, e.g. 2.5-6 fold The addition of H51Y to either R263K or G118R increases resistance against DTG but also further decreases viral fitness These findings help to explain why resistance against dolutegravir in INSTI-naïve patients has not been observed

43. No compensatory mutations in regard to DTG resistance and viral fitness have developed over more than two years in culture.

44. Bluma Brenner Hongtao Xu Dimitri Coutsinos Jerry Zaharatos Maureen Oliveira Thibault Mesplède Peter Quashie Acknowledgements

45. MERCI

46. Long-Term Safety and Efficacy of Raltegravir-Based Versus Efavirenz-Based Combination Therapy in Treatment-Naïve HIV-1 Infected Patients: Final 5-Year Double-Blind Results From STARTMRK AIDS 2012 Poster #LBPE19

47. Proportion (%) of Patients Achieving HIV RNA <50 copies/mL (95% CI) Over Time Non-Completer = Failure Approach

48. Change From Baseline in CD4- Cell Count (95% CI) Over Time

49. Thanks to CIHR and CANFAR