Pharmacokinetic-Pharmacodynamic Modeling & Simulation of the Virologic Response of Dolutegravir in HIV-Infected Patients with Integrase Inhibitor Resistant Virus Ivy Song, 1 Kimberly Adkison, 2 Mark Lovern 3, Joannellyn Chiu 3, Jenny Huang 1, Cindy Vavro 1, Mounir Ait-Khaled 1, Brian Wynne 1 and Sherene Min 1 1 GlaxoSmithKline; 2 GlaxoSmithKline (currently employed by PAREXEL International); 3 Quantitative Solutions UK/DTGP/37/16j Prepared March 17 Prescribing Information is available at the end of this presentation
Dolutegravir (TIVICAY ) HIV-1 integrase strand transfer inhibitor (INI) 5 mg QD: Treatment-naive or treatment-experienced INI-naive : INI-experienced with certain INI-resistant mutations Food increases exposure but may be taken without regard to meals Q148 mutation associated with lower susceptibility to DTG VIKING studies showed reduced efficacy in patients with Q148 + 2 or more additional INI-resistant mutations Objectives Develop exposure-response model and identify significant predictors of antiviral response Use model-based simulation to predict the antiviral response of various dosing scenarios which could increase or decrease DTG exposures in overall INI-resistant population and the subpopulation harboring Q148. Eron et al., (13) J Infect Dis 7:7-748; Castagna et al.,(14) J Infect Dis 21:354-362; Akil et al., (14) Antivir Ther. Doi: 1.3851/IMP2878. UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14.
Methods Logistic regression modeling of VIKING/VIKING-3/VIKING-4 Probability of being a responder (Snapshot/TLOVR HIV RNA < 5 c/ml at Week 24) as a function of plasma DTG exposure (linear and Emax) Both C min and C avg used as a measure of plasma exposure Various covariates tested Simulations Higher dose and co-administration with food Co-administration with moderate-strong enzyme inducers Co-administration with metal cation-containing vitamin supplements Modeling & simulations conducted using NONMEM UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14.
Summary of Subject Characteristics Characteristics (Units) VIKING (N=51) VIKING-3 (N=183) VIKING-4 (N=13) Total N=247 Age (years) 47 48 49 48 [19-68] [19-67] [19-66] [19-68] Baseline CD 4 (cell count /mm 3 ) 122 [19-729] 1 [19-11] 165 [19-525] 15 [19-11] Baseline Viral Load (HIV-1 RNA copies/ml) 4.31 [2.64-6.6] 4.38 [1.69-7.37] 4.41 [2-5.3] 4.38 [1.69-7.37] No Q148 27 (52.9) 126 (68.9) 5 (38.5) 158 (64) Baseline Mutation Q148+1 14 (27.5) 36 (19.7) 6 (46.2) 56 (22.7) Category [N(%)] Q148+ 2 1 (19.6) 21 (11.5) 2 (15.4) 33 (13.4) Use of Metal Cation- No 44 (86.3) 157 (85.8) 1 (76.9) 211 (85.4) Supplements [N(%)] Yes 7 (13.7) 26 (14.2) 3 (23.1) 36 (14.5) Background ART No inducer 19 (37.3) 29 (15.8) 2 (15.4) 55 (22.3) Includes Inducers Mild Inducer 27 (52.9) 144 (78.7) 9 (69.2) 175 (7.9) [N(%)] Mod-Strong 5 (9.8) 1 (5.46) 2 (15.4) 17 (6.9) Values are median [range] unless otherwise indicated No Q148: Includes Y143, N155H, T66, E92Q mutations, or historical evidence of resistance Q148 + 1: Q148H/K/R with one mutation of G1A/C/S. L741I, E138A/K/T Q148 + >=2: Q148H/K/R with two or more mutations of G1A/C/S, L741I, E138A/K/T UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
PKPD Model Logit =.698 1.8 (BVL 4.38) INIMC +.429 (BCD4 15) +.23 C min + η Int C min or C avg equally good predictors of Week 24 response Significant covariates Baseline HIV-1 RNA (BVL) Baseline mutation category (INIMC) INIMC= (No Q148); 1.45 (Q148 + 1); 2.65 (Q148+ 2) Baseline CD4 cell count (BCD4) Non-significant covariates Prior use of raltegravir/elvitegravir or duration of prior INI treatment Phenotypic Susceptibility Score (PSS) of optimized background therapy Genotypic Susceptibility Score (GSS) of optimized background therapy HIV Risk Factor CDC Category UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
Validation: Model Adequately Reflects Observed Data at Week 24 Baseline INI Mutation Category Overall (n=2) No Q148 (n=143) Q148 + 1 (n=51) Q148 + 2 (n=26) Observed Mean Predicted Response Probability (%) Response Rate (%) C min Model C avg Model 68.6 67. 67. 8.4 77.7 77.7 56.9 57.9 57.8 26.9 25.7 25.8 Response is defined as the proportion of subjects with HIV-1 RNA < 5 copies/ml as defined by the TLOVR (VIKING) or Snapshot (VIKING-3, VIKING-4) algorithms UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
C min -Week 24 Response by Mutation Category UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
C min -Week 24 Response by Mutation Category Unknown Prandial State VIKING studies dosed without regard to food so actual data a mix of fasted/fed exposures Majority on PI/r so likely took dose with food (Dose-prop. & Food Effect) Best Case Scenario Assume VIKING subjects were fasted and dosing with moderate fat meal could C mn by additional 45% UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
Simulations: Effect of Higher Dose and Food Predicted Week 24 Response Rate (%) 1 9 8 7 6 5 3 1 Unknown Prandial State (fed/fasted mix) * ** No Q148 Assumes additional 45% Cmin with food * ** 1 9 8 7 6 5 3 1 Unknown Prandial State (fed/fasted mix) Q148 + 1 * ** Dosing Scenario Assumes additional 45% Cmin with food * ** 1 9 8 7 6 5 3 1 Unknown Prandial State (fed/fasted mix) * Q148 + >2 ** Assumes additional 45% Cmin with food * ** *Assumes less than dose-proportional PK; i.e., 1.75-fold higher C min for 1mg BID vs. 5mg BID, based on prior clinical pharm study. **Assumes dose-proportional PK; i.e., 2-fold higher C min for 1mg BID vs 5mg BID. Simulation results of replicates of each cohort (N=1) for each mutation category and regimen Mean and 5 th, 95 th confidence interval presented UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
Simulations: Effect of Moderate-Strong Enzyme Inducers Predicted Week 24 Response Rate (%) 1 9 8 7 6 5 3 1 No Q148 + Inducer + Inducer 1 9 8 7 6 5 3 1 Q148 + 1 Q148 + >2 UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14 + Inducer + Inducer Dose Regimen 1 9 8 7 6 5 3 1 + Inducer + Inducer Simulation results of replicates of each cohort (N=1) for each mutation category and regimen Mean and 5 th, 95 th confidence interval presented
Simulations: Effect of Metal Cation-Containing Supplements Predicted Week 24 Response Rate (%) 1 9 8 7 6 5 3 1 No Q148 + Metal Cation + Metal Cation 1 9 8 7 6 5 3 1 Q148 + 1 Q148 + >2 UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract # 14 + Metal Cation + Metal Cation Dose Regimen 1 9 8 7 6 5 3 1 + Metal Cation + Metal Cation Simulation results of replicates of each cohort (N=1) for each mutation category and regimen Mean and 5 th, 95 th confidence interval presented
Conclusions DTG concentration (C min or C avg ) is a predictor of long-term response. Other predictors include baseline mutation category, baseline HIV-1 RNA, and baseline CD4. Modeling and simulations support the current recommendation of in the majority of the INI-resistant patient population. Simulations confirmed the current recommendations for dosing with moderate-to-strong enzyme inducing drugs or metal cationcontaining vitamin supplements. UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
Acknowledgments Investigators, staff & patients of VIKING, VIKING-3 & VIKING-4 UK/DTGP/37/16j
Backup Slide UK/DTGP/37/16j
PKPD Model Parameter Estimates (Cmin) Parameter (units) Point Estimate θ 1: Intercept.698 55.6 NONMEM Estimates Implementation of %RSE Covariate Effect θ 2: DTG C min Effect.23 43.9 θ 1 + θ 2 x Cmin θ 3: Baseline HIV-1 RNA Effect -1.8 25.4 θ 1 + θ 3 x (BVL-4.38) θ 4: Q148 + 1 Effect -1.45 29.4 θ 1 + θ 4 x INIMC1 θ 5: Q148 + 2 Effect -2.65 18.2 θ 1 + θ 5 x INIMC2 θ 6: Baseline CD 4 Effect.429 36.4 θ 1 + θ 6 x (BCD 4-15) Non-significant covariates tested: Prior use of raltegravir/elvitegravir at screening, duration of prior INI treatment, Phenotypic Overall Susceptibility Score (PSS), or Genotypic Susceptibility Score (GSS) of optimized background therapy, HIV Risk Factor, CDC Category UK/DTGP/37/16j Song et al. 16th International Workshop on Clinical Pharmacology of HIV and Hepatitis Therapy 15; Washington, DC. Abstract #14
Tivicay dolutegravir 5mg tablets Prescribing Information See Summary of Product Characteristics before prescribing Indication: HIV in >12 years and >kg as part of combination therapy. Dosing: 5mg once daily with or without food if no proven/ suspected integrase resistance. 5mg twice daily with efavirenz, nevirapine, tipranavir/ritonavir, etravirine (without boosted PI), carbamazepine, oxcarbazepine, phenytoin, phenobarbital, St John s Wort or rifampicin. Adults with proven/ suspected integrase resistance: 5mg twice daily preferably with food. Elderly: Limited data in 65+ yrs. Caution in severe hepatic impairment. Contraindications: Hypersensitivity to any ingredient. Co-administration with dofetilide. Warnings/precautions: Risk of hypersensitivity reactions. Discontinue dolutegravir and other suspect agents immediately if suspected. Risks of osteonecrosis, immune reactivation syndrome. Monitor LFTs in Hepatitis B/C co-infection and ensure effective Hepatitis B therapy. Caution with metformin: monitor renal function and consider metformin dose adjustment. Use with etravirine requires boosted PI or increased dose of dolutegravir. Use with Mg/Alcontaining antacids, calcium, multivitamins or iron requires dosage separation. Pregnancy/ lactation: Not recommended. Avoid breastfeeding. Side effects: See SPC for full details. Headache, GI disturbance, insomnia, abnormal dreams, depression, dizziness, rash, pruritus, fatigue, elevations of ALT, AST and CPK, arthralgia, myalgia, hypersensitivity, suicidal ideation or suicide attempt. Basic NHS costs: 3 tablets 498.75 EU/1/13/892/1. MA holder: ViiV Healthcare UK Ltd, 98 Great West Road, Brentford, Middlesex TW8 9GS. Further information available from Customer Contact Centre, GlaxoSmithKline UK Ltd, Stockley Park West, Uxbridge, Middlesex UB11 1BT. POM S1A Tivicay is a registered trademark of the ViiV Healthcare Group of Companies Date of approval: January 17 Zinc code: UK/DLG/55/13(9) Adverse events should be reported. For the UK, reporting forms and information can be found at www.mhra.gov.uk/yellowcard. Adverse events should also be reported to GlaxoSmithKline on 8 221 441. Adverse events should be reported. For Ireland, adverse events should be reported directly to the HPRA; Freepost, Pharmacovigilance Section, Health Products Regulatory Authority, Earlsfort Terrace, Dublin 2, Tel: +353 1 676 4971, medsafety@hpra.ie. Adverse events should also be reported to GlaxoSmithKline on 18 244 255.