Switch to DTG-containing regimen STRIIVING Study NEAT 022 Study

NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Design Randomisation stratified by country 1 : 1 Open label W48 W96 6 European countries HIV+ ≥ 50 years or ≥ 18 years with Framingham risk score > 10% at 10 years On PI/r + 2 NRTIs > 6 months HIV RNA < 50 c/mL ≥ 6 months No documented primary resistance mutations Switch to DTG + 2 NRTIs (unchanged) N = 205 N = 210 Continuation PI/r + 2 NRTIs (unchanged) Switch DTG + 2 NRTIs Primary endpoints Proportion of patients with virological success at W48 (no consecutive HIV-1 RNA > 50 c/mL and no treatment discontinuation): non-inferiority of DTG, by ITT, Kaplan-Meier analysis; lower limit of the 95% CI for the difference = - 10%, power 90% Mean percentage changes in fasting lipid total cholesterol at W48 (between treatment difference of 12%, 99% power) NEAT 022 Gatell JM. AIDS 2017;31:2503-14 ; Gatell JM. Clin Infect Dis 2019 ; 68 :597-606

Baseline characteristics and patient disposition NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Baseline characteristics and patient disposition DTG N = 205 PI/r N = 210 Female, % 11.7 10.0 Age > 50 years, % 87.3 87.6 Framingham score > 10 % at 10 years, % 75.6 71.9 CD4 cell count (/mm3), median 635 585 HIV RNA > 50 c/mL, % 3.4 0.5 HCV serology positive, % 13.4 11.6 Years of HIV RNA < 50 c/mL, median 4.9 5.3 Discontinuation before W48, N (%) For lack of efficacy For adverse event Death Lost to follow-up Withdrew consent Other 14 (6.8) 4 7 1 2 10 (4.8) 3 NEAT 022 Gatell JM. AIDS 2017;31:2503-14

CV risk factors and ARV therapy at screening NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk CV risk factors and ARV therapy at screening DTG N = 205 PI/r N = 210 Current smoker, % 38 37.8 Diabetes, % 5.5 6.3 Family history of cardiovascular disease, % 43.3 43.4 On lipid lowering agent, % 30.7 28.6 High blood pressure, % 35.3 37.6 Daily exercise, % 32.5 28.9 Fasting total cholesterol, mmol/L, median (IQR) 5.2 (4.5 - 5.8) 5.1 (4.5 - 5.6) Fasting triglycerides, mmol/L, median (IQR) 1.6 (1.2 - 2.3) 1.6 (1.2 - 2.2) NRTI, % TDF/FTC ABC/3TC Other 65.4 30.7 3.9 64.3 31.9 3.8 PI/r, % DRV/r ATV/r 51.5 37.7 10.7 51 35.2 13.8 NEAT 022 Gatell JM. AIDS 2017;31:2503-14

Difference (95% CI): - 2.1% (- 6.6 to 2.4) NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Outcome at W 48, ITT Treatment success Discontinuation for AE Lost to follow-up Other DTG PI/r 20 40 60 80 100 93.1 95.2 3.4 2.0 1.0 1.5 1.4 Virologic non-response 0.5 No virological data % Difference (95% CI): - 2.1% (- 6.6 to 2.4) Confirmed virological failure (HIV RNA > 50 c/mL): DTG, N = 4 vs PI/r, N = 1 ; genotype successful in 2/4 and 0/1 patients: no emergence of resistance mutations NEAT 022 Gatell JM. AIDS 2017;31:2503-14

Treatment success at W48 (ITT, per-protocol and sub-groups) NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Treatment success at W48 (ITT, per-protocol and sub-groups) 2.1% 3.0% 2.9% 0.4% 1.1% 6.2% 0.2% 0.1% - 9 - 6 -3 3 6 9 12 0.44 1.00 95.2 97.5 97.2 92.9 90.1 100 96.9 97.8 93.1 94.5 94.3 92.5 89.0 93.8 96.7 97.7 97.3 53.0 47.0 39.5 23.9 14.9 21.7 ITT analysis Per-protocol analysis < 15% > 15% United Kingdom Spain Germany Belgium-France-Italy Framingham 10-years CV risk Country PI/r DTG % of patients Difference, % (95 % CI) p for interaction Success (%) DTG better PI/r better NEAT 022 Gatell JM. AIDS 2017;31:2503-14

Fasting plasma lipids (mmol/L): mean percentage change at W48 NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Fasting plasma lipids (mmol/L): mean percentage change at W48 5 10 - 5 - 10 - 15 - 20 - 25 DTG PI/r Total cholesterol Non-HDL cholesterol Triglycerides LDL cholesterol HDL cholesterol Total cholesterol: HDL-chol. ratio p < 0.001 p = 0.286 - 8.7 0.7 - 11.3 0.5 -18.4 4.2 - 7.7 2 1.1 2.5 - 7.0 0.4 % No changes in the utilization of lipid lowering agents (around 30% in each arm, both at baseline and W48) NEAT 022 Gatell JM. AIDS 2017;31:2503-14

NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Adverse events, % DTG + 2 NRTI N = 205 PI/r + 2 NRTI N = 210 Grade 3-4 adverse event 5.9 9.1 Serious adverse event 7.7 Adverse event related to treatment 12.8 41 AE in 26 patients including 15/41 mood, sleep or CNS disorders 7.2 21 AE in 15 patients including 6/21 mood, sleep or CNS disorders Discontinuation for adverse event 3.4 (N = 7 *) 1.4 (N = 3 **) Death 0.5 (N = 1, accidental fall) * Acute HCV infection (N = 1), mood and/or sleep disorders (N = 6) ** Acute HCV infection (N = 1), dyspepsia (N = 1), worsening of renal function (N = 1) Median change in eGFR (CKD-EPI) at W48 (p < 0.001) DTG: ≈ - 8 mL/min PI/r: 0 mL/min NEAT 022 Gatell JM. AIDS 2017;31:2503-14

NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk Conclusions Over 48 weeks, in virologically suppressed patients with high cardiovascular risk (older than 50 years and/or with a Framingham score > 10%) and receiving triple therapy with PI/r + 2 NRTIs Switching to a DTG regimen was non-inferior. Sensitivity and subgroup analysis support this conclusion Improved total cholesterol and other lipid fractions in the overall population and in several subgroups Very few episodes of confirmed virological failure and no resistance mutations selected Overall tolerance was good and similar in both arms NEAT 022 Gatell JM. AIDS 2017;31:2503-14

NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk 87 Outcome at W96, ITT Treatment success Adverse event Lost to follow-up Other DTG, immediate switch DTG, deferred switch 20 40 60 80 100 92.2 87.0 3.9 6.7 2.0 0.5 1.0 1.9 Virologic non-response 2.4 No virologic data % Virologic failure DTG, immediate switch DTG, deferred switch Confirmed virologic failure 5 Genotype results available 2 1 Emergence of resistance mutation NEAT 022 Gatell JM, CID 2018, 14 June (Epub ahead of print)

NEAT 022 Study: Switch to DTG vs continuation of PI/r in patients with high cardiovascular risk 88 Discontinuation for adverse event W48-W96 Deferred switch, N = 9 (mood and/or sleep disorders, N = 7) Immediate switch: none Change in fasting lipids Deferred switch: improvement of lipid parameters at W96 (total cholesterol, non-HDL cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides), all statistically significant and of same level as at W48 in immediate switch group Immediate switch: stabilisation of improvement between W48 and W96 Change in % of patients receiving or requiring lipid lowering agents according to NCEP ATP-III guidelines DTG, immediate switch DTG, deferred switch 30 35 W0 40 45 W48 W96 NEAT 022 Gatell JM. Clin Infect Dis 2019 ; 68 :597-606