Certara, Inc. · NASDAQ:CERT
Certara — Technical Presentation (October 2012)
56 slides from Certara's technical presentation, Oct 9, 2012
Slide 1: Modeling and Simulation in Pediatric Drug Development
Modeling and Simulation in Pediatric Drug Development Applications of “Top-Down” and “Bottom-Up” Approaches to Optimize Trial Design and Support Dosing Rationale JF Marier, PhD, FCP, Vice President and Lead Scientist Pharsight Consulting Services, part of Certara™ Trevor Johnson, PhD, Principal Scie
Slide 2: Today's Presenters
Today’s Presenters JF Marier, PhD, FCP Vice President and Lead Scientist, North America Pharsight Consulting Services, part of Certara Dr. Marier has directed PK/PD analyses, reporting, and modeling and simulation projects for hundreds of studies in early and late clinical research. His drug develop
Slide 3: Today's Presenters
Today’s Presenters Trevor Johnson, PhD Principal Scientist and Deputy Head of Systems Biology at Simcyp (now Certara) Dr. Johnson obtained his Pharmacy degree from the University of Bath, a Masters in Clinical Pharmacology from the University of Aberdeen and his PhD in Paediatric Clinical Pharmacolo
Slide 4: Agenda
Agenda 1- Drug Model: PK/PD Analyses and Top-Down Approaches – Leverage a priori knowledge in adults • From current drug development program (Phase II-III), same or different indication (e.g., line extension) • From literature – Semi-mechanistic population PK/PD models • PK (Intrinsic and extrinsic
Slide 5: 1- Drug Model: PK/PD Analyses and Top-Down Approaches
1- Drug Model: PK/PD Analyses and Top-Down Approaches © Copyright 2012 Certara, L.P. All rights reserved. 5
Slide 6: 1- Drug Model: Leverage a priori knowledge in adults - 1 of 9
1- Drug Model: Leverage a priori knowledge in adults - 1 of 9 Leverage a priori knowledge in adults (same of different indication) – Leverage PK/PD data collected in Phase II and III (line extension) – Understand sources of variability on PK/PD parameters – Size is the primary covariate for PK – Can
Slide 7: 1- Drug Model: Leverage a priori knowledge in adults - 2 of 9
1- Drug Model: Leverage a priori knowledge in adults - 2 of 9 Size and Maturation – Growth and development are two major aspects in pediatrics not seen in adults. – PK models based on adult data can be customized to account for organ maturation. PK Model Customization – Information about organ matur
Slide 8: Semi-mechanistic Population PK/PD models - Size vs. Maturation
Typical Maturation of GFR (% of adult 100 125 150 1- Drug Model: Semi-mechanistic TM50 = 47.7 w eeks Population PK/PD models - 3 of 9 Hill = 3.40 BSV=23% 2 years Size vs. Maturation 75 1 year – While size matters the most in pediatrics > 2 years of age 50 – … maturation remains the most important co
Slide 9: Semi-mechanistic Population PK/PD models - Size vs. Maturation
1- Drug Model: Semi-mechanistic Population PK/PD models - 4 of 9 Size vs. Maturation – Population PK Model may be customized accordingly CLTotal = CLNon-Renal + CLRenal Size 20 Original 15 Weight (kg) Pop PK 10 Model 5 (based on adult 0 data from 0.0 0.5 1.0 1.5 2.0 Phase II-III) Age (years) Customi
Slide 10: Implementation of a Custom PK/PD Model in Trial Simulator
1- Drug Model: Semi-mechanistic Population PK/PD models - 5 of 9 Implementation of a Custom PK/PD Model in Trial SimulatorTM 2-Compartment Model 2-Compartment Model with Maturation of CL/F IV IV Drug Drug Maturation van der Marel et al., (2003) Eur J Clin Pharmacol. 59(3):243-251. © Copyright 2012 C
Slide 11: Implementation of a PK/PD Model in Trial Simulator
1- Drug Model: Semi-mechanistic Population PK/PD models - 6 of 9 Implementation of a PK/PD Model in Trial Simulator® Drug Literature: PD Model Literature: Absorption model Prins SA et al. (2008) Paediatr Aneaesth, 18:582-592. Maturation Anderson et al.(2001) Eur J Clin Pharmacol. 57:559-569. van der
Slide 12: A few words about monoclonal antibodies
1- Drug Model: Semi-mechanistic Population PK/PD models - 7 of 9 A few words about monoclonal antibodies – Size – Target (maturation and baseline severity) 1- Population PK/PD Modeling and Sources of Variability to Support Dosing Rationale IV Dose Ksyn On mAb-Target mAb + Target Off Complex CL Kdeg
Slide 13: 1- Drug Model: External Validation - 8 of 9
1- Drug Model: External Validation - 8 of 9 Last Step: External Validation to Evaluate Predictive Performance of PK Model : Animal Data Predictive Performance of Allometric Model: Adult Monkeys <25% Male Adult Monkeys <50% Female Adult Monkeys © Copyright 2012 Certara, L.P. All rights reserved. 13
Slide 14: External Validation: Off-Label Information in Pediatrics
1- Drug Model: External Validation - 9 of 9 External Validation: If available, leverage “Off-Label” information in pediatrics – Off-label use (compassionate use). – Troughs samples collected for safety monitoring (hematology measurements) – Bioanalytical assay of drug concentration – Manually entere
Slide 15: 2- Patient Model
2- Patient Model © Copyright 2012 Certara, L.P. All rights reserved. 15
Slide 16: PK/PD Simulations: Targeted Patient Population
2- Patient Model - 1 of 3 PK/PD Simulations: Targeted Patient Population (age, weight, disease status…) – Importance of Using Realistic Covariates in Trial Simulations – Relationship Between Body Weight and Age © Copyright 2012 Certara, L.P. All rights reserved. 16
Slide 17: Using Realistic Covariates in Trial Simulations
2- Patient Model - 2 of 3 Using Realistic Covariates in Trial Simulations – Create an “In Silico” Patient Population – Allometric Function Fed with Realistic Body Weight Distribution of the Targeted Patient Population – Goal: Obtain age-weight-sex matched data for the intent-to-treat population – Co
Slide 18: Using Realistic Covariates in Trial Simulations Example
2- Patient Model - 3 of 3 Using Realistic Covariates in Trial Simulations – Example of patient population that doesn’t follow the typical CDC growth curve Lower Body Weight Higher Body Weight e.g., malnutrition, short-bowel syndrome… e.g., obese pediatrics © Copyright 2012 Certara, L.P. All rights r
Slide 19: 3- Trial Model to Optimize Design
3- Trial Model to Optimize Design 3.1- Determine Optimal Dose to Ensure Adequate Safety and Efficacy 3.2- Technical Aspect: Optimize Study Design © Copyright 2012 Certara, L.P. All rights reserved. 19
Slide 20: Simulations: A Well Recognized Tool for Optimizing Pediatric Designs
3- Trial Model: Determine Optimal Dose to Ensure Adequate Safety and Efficacy 1- Simulations: A Well Recognized Tool for Optimizing Pediatric Designs – FDA: Series of Case Studies & Design by simulations 100% pediatric protocols – European Medicine Agency © Copyright 2012 Certara, L.P. All rights re
Slide 21: Trial Model: Determine Optimal Dose
3- Trial Model: Determine Optimal Dose to Ensure Adequate Safety and Efficacy Next Step – Drug Model – Patient Model – Trial Model: Optimize Dosing • Safety and Efficacy 100% Clinical Response 80% • Trade-Off (Risk/Benefit) Efficacy 60% 40% 20% 0% Adverse Events 40% Tolerability 20% Safety 0% 0
Slide 22: FDA's Pediatric Decision Tree
3- Trial Model: Determine Optimal Dose to Ensure Adequate Safety and Efficacy Determine Candidate Dose(s) that will Optimize Safety and Efficacy in Pediatrics – FDA’s Pediatric Decision Tree available from Exposure-Response Guidance e.g., anti-inflammatory, 1 immunomodulatory products e.g., anti-hyp
Slide 23: Safety and Efficacy Exposure Targets
3- Trial Model: Determine Optimal Dose to Ensure Adequate Safety and Efficacy A targeted range of exposure associated to adequate safety must be identified. • Safety First ! (Upper Range of Exposure): – Phase II-III: The exposure (AUC, Cmax, Cmin) associated acceptable safety profile in adult patien
Slide 24: PK Driven Optimization
3- Trial Model: Determine Optimal Dose to Ensure Adequate Safety and Efficacy PK Driven – Optimization According to Pre-Identified Therapeutic Window Therapeutic window © Copyright 2012 Certara, L.P. All rights reserved. 24
Slide 25: PK-PD Driven Optimization
3- Trial Model: Determine Optimal Dose to Ensure Adequate Safety and Efficacy PK-PD Driven – Optimization According to PK/PD Model PK/PD Model –Implementation in Trial Simulator® 7.5 mg/kg q6h 10 mg/kg q6h 10 th percentile 10 th percentile PK Population Model 9 Median 9 Median Drug effect Drug Effec
Slide 26: 3- Trial Model: Technical Aspect
3- Trial Model: Technical Aspect Next Step: – Drug Model – Patient Model – Trial Model: Dose(s) that will Optimize Safety and Efficacy – Trial Model: Optimize Technical Aspect of Pediatric Trials using a 3-Step Approach • Develop Sparse Sampling Strategy (optimizing Fisher Information Matrix)
Slide 27: Validate Robustness of Model
3- Trial Model: Technical Aspect Next Step: Validate Robustness of Model – Simulate 1000 Trials – Fit Data from 1000 Trials (Drug Model, Patient Model, Trial Model, including BLQ, drop-out, missed samples) – Derive Robustness Parameters (e.g., RSE <20% or 95% CI within 0.6-1.4)* – Determine sample s
Slide 28: Case Study: Fixed-Based Dosing of mAb in Pediatric Patients with Rare Disease
Case Study 1 Fixed-Based Dosing of mAb in Pediatrics Patients with Rare Disease © Copyright 2012 Certara, L.P. All rights reserved. 28
Slide 29: Fixed Based Dosing of mAb: Background and Goal
Fixed Based Dosing of mAb in Pediatrics Patients with Rare Disease 1- Background – Develop a dosing strategy from infants to adults (15-fold range of body weight) – mAb commercialized in adults (PK/PD Data in a total of ~200 adult patients) – Leverage information from adults and apply to another ped
Slide 30: Fixed Based Dosing of mAb: Safety and Efficacy Window
Fixed Based Dosing of mAb in Pediatrics Patients with Rare Disease 1- Safety – mAb well tolerated – Concentration higher than 500 µg/mL have never been explored – Upper exposure capped to 500 µg/mL Ksyn Ksyn 2- Efficacy – Concentration of Soluble Target: 200 µg/mL Target Target – 2:1 Stoichiometric
Slide 31: Fixed Based Dosing of mAb: Optimized Dosing Strategy
Fixed Based Dosing of mAb in Pediatrics Patients with Rare Disease 1- The following dosing strategy optimized the distribution of the mAb concentrations within the targeted window Body Weight Cohort Dosing ≥40 kg 800 mg Q2 weeks 20 - <40 kg 400 mg Q2 weeks 10 - <20 kg 200 mg Q2 weeks 5 - <10 kg 100
Slide 32: PK/PD Modeling: Concentrations Across Age Groups
Fixed Based Dosing of mAb in Pediatrics Patients with Rare Disease PK/PD Modeling: All Concentrations were >90% of Emax Across All Age Groups Safety: Maximum Concentrations (Target: <500 µg/mL) Targeted Window Geometric Mean (CV%) Population Efficacy Safety First Dose Steady State ≥40 kg 223.6 (28.2
Slide 33: Conclusion
Conclusion Modeling and Simulations: Top Down Applications – Leverage PK/PD knowledge gathered in Phase II and III for top down applications (line extension) – Perform PK/PD analysis and customize model • Size • Maturation • Literature, in-house, or structurally-related products – Perform Simulation
Slide 34: IVIVE and PBPK in paediatric drug development
IVIVE and PBPK in paediatric drug development Bottom up, Top down and Middle out Dr Trevor Johnson, Principal Scientist.
Slide 35: Overview
Overview • PBPK models • Paediatric Model Development • Some Applications • ‘Top down’ vs ‘Bottom up’ • ‘Middle out’ • Regulatory view • Consultancy applications • Buccal midazolam example © Copyright 2012 Certara, L.P. All rights reserved.
Slide 36: How does the PBPK M&S approach differ from conventional PK?
How does the PBPK M&S approach differ from ‘conventional’ PK? It is predictive, based on many pieces of prior data and their variability, whereas classical PK analysis is more descriptive being driven by the actual (usually plasma drug concentration) data, and seeks to find covariates post hoc. “TOP
Slide 37: The Complexity of PBPK Covariates – correlated Monte Carlo
The Complexity of PBPK Covariates – correlated Monte Carlo Genotypes (Distribution in Population) Renal Function Plasma Body Ethnicity Disease Proteins Fat Serum & Creatinine Haematocrit Sex Age (Distribution in Population) (Distribution in Population) Height Brain Body Volume Heart Surface Volume A
Slide 38: Separating Systems & Drug Information
Separating Systems & Drug Information Trial Design Input data: - Compound file Systems - Population library Drug Data - Project (workspace) Data Integrated Models: Mechanistic IVIVE - Simulation environment & PBPK/PD - Multi-dimensional models/algorithms Output data: Population PK (/PD) - Raw output
Slide 39: Paediatric PBPK – Data, data and more data
Paediatric PBPK – Data, data and more data Drug Elimination (IVIVE+) Drug Distribution (PBPK) Lung Lung Population: Population: N NSubject Subjectof Known Age/Sex of Known Age/Sex Population Population Adipose Adipose Liver: Liver: Specific Specific weight/blood flow weight/blood flow Paediatric Inp
Slide 40: Some underlying model changes with Age
Some underlying model changes with Age 1 2.5 Fraction of Adult Value Simcyp 0.8 2 In vivo 1.5 Adult 0.6 LV (L) CYP3A4 1 0.4 CYP1A2 0.5 0.2 CYP2C9 0 0 0 5 10 15 20 25 30 0 5 10 15 20 Age (y) Age (y) 80 1.4 70 Brain 1.2 60 1 BBF (L.h) AAG (g/L) 50 0.8 40 0.6 30 0.4 20 10 0.2 0 0 0 2 4 6 8 10 12 14 0.1
Slide 41: CL Prediction
CL Prediction Child Adolescent 2.5 7 Midazolam (IV) 6 Midazolam (Oral) 2 5 Adult 1.5 4 3 1 CL (L.kg.h) 2 0.5 1 0 0 0 20 40 60 80 100 0 20 40 60 80 100 0.60 1.4 Caffeine (Oral) 1.2 Diclofenac (IV) 0.50 1 Infant 0.40 0.30 0.8 0.6 Neonate 0.20 0.4 0.10 0.2 0.00 0 0 20 40 60 80 100 0 20 40 60 80 100 4.5
Slide 42: Predicting PK Profiles
Predicting PK Profiles 0.20 30 MDZ IBU 0.15 20 0.10 10 0.05 0.00 0 0 2 4 6 0.4 0 2 4 6 0.2 Zol ITR Conc (mg/L) 0.15 0.3 0.1 0.2 0.05 0.1 0 0 0 3 6 9 12 0 5 10 15 20 25 12 10 THEO CAFF 8 5 4 0 0 0 2 4 6 8 0 3 6 9 12 Time (h) © Copyright 2012 Certara, L.P. All rights reserved.
Slide 43: Paediatric DDI
Paediatric DDI 1 Fraction of adult values 0.8 0.6 0.4 CYP1A2 Lack of paediatric clinical data 0.2 CYP2C9 0 0.001 0.01 0.1 1 10 Age (y) COMP 1 COMP 2 1 Median Fraction of adult values 0.8 CYP1A2 CYP2C9 CYP3A4 CYP2D6 Median Age AUC fm % fm % fm % fm % AUC Ratio Ratio 0.6 0.4 CYP3A4 year 20 93 7 1.05 7
Slide 44: Bottom-Up Meets Top-Down
Bottom-Up Meets Top-Down Clin Pharmacol Ther 2008 1 1 CYP3A4 activity (DX/3HM CYP2D6 activity (DM/DX (A) (B) ratio) relative to adult ratio relative to adult 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0 0 0 4 8 12 0 4 8 12 Age (Months) Age (Months) Changes in CYP2D6 (a) and CYP3A4 (b) activity relative to adul
Slide 45: Pop-PK and Covariate Effects
Pop-PK and Covariate Effects BJCP, 2007, 64(6) 772- 784 CL = Typical parameter estimate x (Body weight/13) 3/4 x [1 - 0.0542 x (Cholesterol - 5.4)] x [1 - 0.00732 x (Haematocrit - 31)] x [1 + 0.000214 x (Serum creatinine - 524)] The typical values refer to a patient with a body weight of 13 kg, chol
Slide 46: Bottom-Up Meets Top-Down
Bottom-Up Meets Top-Down Clpo fu/[(CRBC/Cp)*HC + (1- HC)] fu = 0.037 and CRBC/Cp = 1.8 fuB = 0 0.0296 (at HC = 0.31) [1 - 0.00732 x 100*(HC - 0.31)] [[0.037/[1.8*HC + (1- HC)] ]/ 0.0296] 1.4 Relative Change in CL 1.2 1 0.8 0.6 CL Multiplier (Top-Down) 0.4 CL Multiplier (Bottom-Up) 0.2 0 0 0.1 0.2
Slide 47: Just Made It! ASCPT 2012
Just Made It! ASCPT 2012 YES NO © Copyright 2012 Certara, L.P. All rights reserved.
Slide 48: Regulatory – PBPK Pediatric
Regulatory – PBPK Pediatric CPT 2012 • Summarises FDA experience in review of paediatric submissions containing PBPK – 4 drugs • Identifies challenges in using PBPK to predict paediatric PK by relating research experience with ‘model drugs’ • Reviews workflow of incorporating PBPK into paediatric dr
Slide 49: Linking Top down and bottom up – a learn and confirm approach
Linking Top down and bottom up – a learn and confirm approach BOTTOM UP BOTTOM UP Predict PK in Predict PK in paediatrics, Compound X adults and Use adult PK data – optimise model retrograde model Use PK data to further refine PBPK model Confirm and understand system. Use generated Learn knowledge t
Slide 50: Application of Simcyp Paediatric in Drug Development
Application of Simcyp Paediatric in Drug Development Dose linearity at different ages. Different route of administration – buccal administration with different % swallowed. Exposure following second dose Drug interaction potential at different ages and with different routes of administration
Slide 51: Buccal midazolam example
Buccal midazolam example • Buccal midazolam for treatment statue epilepticus in children • Wide age range 3 months to adults • Second dose may be needed after minimum of 10 minutes a. Dose Linearity 60 14 50 R² = 0.999 12 R² = 1 AUC (mg/L/h) 40 10 R² = 1 Cmax (mg/L) 30 8 R² = 0.9999 6 20 4 10 2 0 0
Slide 52: Buccal midazolam example: Dose swallowing
Buccal midazolam example b. Dose swallowing 3 month 2 years Adult 0.25 0.25 0.25 Systemic Concentration (mg/L) Systemic Concentration (mg/L) Systemic Concentration (mg/L) 0.20 0.20 0.20 0.15 0.15 0.15 0.10 0.10 0.10 0.05 0.05 0.05 0.00 0.00 0.00 0 8 16 24 32 40 48 0 8 16 24 32 40 48 0 8 16 24 32 40
Slide 53: Buccal midazolam example: Effects of second dose
Buccal midazolam example c. Effects of second dose 3 month age group Second dose after 30 minutes Second dose after 60 minutes Second dose after 10 minutes 0.40 0.35 Systemic Concentration (mg/L) 0.45 Systemic Concentration (mg/L) Systemic Concentration (mg/L) 0.40 0.35 0.30 0.35 0.30 0.25 0.30 0.25
Slide 54: Summary
Summary • Paediatric PBPK becoming more established in drug development. • Increasing understanding through more application – evolving models. • Marriage of Bottom up and Top down approaches. • Can answer ‘what if’ questions and explore scenarios where clinical PK studies would be difficult to perf
Slide 55: Thanks to all members of Simcyp team
Thanks to all members of Simcyp team who have been involved in the development of Simcyp Paediatric © Copyright 2012 Certara, L.P. All rights reserved. 55
Slide 56: Contacts & Follow-Up: Thank you!
Contacts & Follow-Up: Thank you! For more information regarding today’s presentation, please contact: JF Marier, PhD, FCP Vice President and Lead Scientist, North America Pharsight Consulting Services, part of Certara™ [email protected] Trevor Johnson, PhD Principal Scientist Simcyp Limited, par