Systems Technology In Pharmaceutical And Biologics QbD . - Uml.edu

Systems Technology in Pharmaceuticaland Biologics QbD ImplementationRichard D. BraatzNovartis-MIT Center for Continuous ManufacturingCenter for Biomedical InnovationDepartment of Chemical Engineering

Background on QbD approachProcess development involves: Define the target product profile Identify the critical quality attributes (CQAs) Select an appropriate manufacturing strategy Implement a control strategyThis talk is on control systems technologyfor integrated pharmaceutical and biologicsmanufacturing2 Novartis/MIT, All Rights Reserved

Why Integrated Manufacturing?Reduce contact betweenbiology/chemistry & personnelContinuous operation hasthe potential to Increase product quality Increase yields Enable new drug productformulations (e.g., thin films) Reduce scale-up risks Reduce footprint3 Novartis/MIT, All Rights Reserved

Outline Control Systems for Integrated Continuous Operations Design Spaces vs. Feedback Control Application to Biologics Manufacturing4 Novartis/MIT, All Rights Reserved

Integrated Control Strategyfor Continuous ManufacturingCTCATs1 P1s9s2 P2As10s3 C2LTs15s14FTP8P7PU1W1s17s5 P5S2PU2s18s16LTS1D1LTP12DTs6 P6S3s7S1s21S1CTP9P11s Tightintegrationoperationscan results inS3M3 sR2 s propagating ctsare suppressedby an CT34LT Thestrategy must optimize the overall plant operationsES1s24P1725P20FTP21DTinsteadP21 of only isolated units (i.e., need plantwide control)SsCTs3715 Novartis/MIT, All Rights Reserved36s45PU3

Plantwide Control of Continuous Manufacturing ChallengesCTCATA Many connected unit operations Very fast to slow processesBS1S2S3S1S1P10s1 P1s9s2 P2s10s3 8P7PU1W1s15s14FTD1s17LTP12DTs6 P6CTs7 P9s21P11CTS3 P16 Multi-purpose plants with short development timeCFTs5 P5s26DC1S1PU2s18M3 s27s22P13s23P14s24P17s25P20s37P21R2s28M4 4D2LTFTP21 Alignment with regulatory requirements (e.g., design space)ES1S1DTCTs36s45s46 Approach adapted from the chemical industrys40M5s41s42EX1s38P22S5E1P24R. Lakerveld, B. Benyahia, R.D. Braatz, & P.I. Barton, Model-based design of a plant-wide controlstrategy for a continuous pharmaceutical plant, AIChE Journal, 59, 3671-3685, 2013 Novartis/MIT, All Rights ReservedS1ST Experimentally demonstrated on continuous pilot plant6S1s43TCS6 Employs systematic and modular design of plantwidecontrol strategies for continuous manufacturing facilitiesEX2s39PU3s44CSFTs47CTFP

A continuous pilot plant7 Novartis/MIT, All Rights Reserved

A continuous pilot plantCATspCATA B LCspDCFCS3S1S1S1M3spM4R2CCRCspC3LCS3FCI1 C I 2 PBPLCLCS4sp CCCFTFCDspC4LCW2LCspPU4spD2I 2 E APICTLCspDCES1S1 First-principlesdynamicmodels werebuilt for eachunit operation(UO) as theywere developed Models werevalidated andthen placed intoa plant-widesimulationPU3S1S1M5spFCTCEX1S5S6EX28 Novartis/MIT, All Rights ReservedE1CSFP Plant simulationused to designUO & plantwidecontrol strategy

Model-based Design of a Plant-wide Control Strategy2. Optimizing controlobjectives Control tasks classified intooptimizing and stabilizing Hierarchical decomposition Reduces complexity Exploits separation of time scales1. Stabilizing controlobjectives (local)9 Novartis/MIT, All Rights ReservedLevel 1 – Total I/OControl tasks(CQA, CPP)1Level 2 – IntermediatesControl tasks(CQA, CPP)2Level 3 – RecyclesControl tasks(CQA, CPP)3Level 4 – DetailedStructure of plant-widecontrol strategy

Parametric Sensitivities Used to EvaluateRelationships Between CPPs and CQAs Use sensitivities (Si,j) to identify yiSi , j p jcausal relations CPPs-CQAs: Direction and order of magnitude Guide selection of automated control loops Determined from process simulation (could use DOE)dx (t ) f x (t ), u(t ), p, t ,dty (t ) g x (t ), u(t ), p, t t t0 , t f , x t t0 x0 ,h x (t ), u(t ), p, t 0,x (t )- State variablesu (t )- Input variablesy (t )- Output variables (CQAs)p- Parameters (PPs)t 1d uMV (t ) K p (t ) ( )d D , for all feedback control loops dtI 0 (t ) ySP yCV (t )10 Novartis/MIT, All Rights Reserved

Example Sensitivity Results:Level 1: Total Inputs & Outputs11 Novartis/MIT, All Rights Reserved

Also use sensitivities to evaluate dynamic I/O relationships,to assess controllability and disturbance propagationModel-based sensitivity of two final product quality variableswith respect to feed flow rate of reactantNormalized Sensitivity10.80.60.4API dosage0.2Impurity content0012 Novartis/MIT, All Rights Reserved50Time / h100150

A continuous pilot plantCATspCATA B LCspDCFCS3S1S1S1M3spM4R2CCRCspC3LCS3FCI1 C I 2 PBPLCLCS4sp CCCFTFCDspC4LCW2LCspPU4spD2I 2 E APICTLCspDCES1S1PU3S1S1M5spFCTCEX1S5S6EX213 Novartis/MIT, All Rights ReservedE1CSFPMet allpurityspecs inSummer2012

Outline Control Systems for Integrated Continuous Operations Design Spaces vs. Feedback Control Application to Biologics Manufacturing14 Novartis/MIT, All Rights Reserved

Design Space vs. Feedback Control (both areconsistent with quality-by-design principles) Design-space methods: Control strategy based on operationwithin a fixed parameter space Applicable to each continuousprocess unit operation More complicated to apply to anentire continuous pharmaceuticalmanufacturing plant15 Novartis/MIT, All Rights Reserved Feedback methods: Control strategy based onfeedback to a “parameter space” Easier to scale up Design space does not need tobe exhaustively validated a priori Necessary for continuousmanufacturing

Outline Control Systems for Integrated Continuous Operations Design Spaces vs. Feedback Control Application to Biologics Manufacturing16 Novartis/MIT, All Rights Reserved

Manufacturing biologic drugs todayProduct QC: Haverhill, UKProduct manufacturing:Allston, MA USAFill/finish: Waterford, IECerezyme patients distributed worldwide17

Towards Biomanufacturing on Demand (BioMOD)Design Requirements PatientBioMOD capabilities Enable flexible methodologies for genetic engineering/modification of microbialstrains to synthesize multiple and wide-ranging protein-based therapeutics Develop flexible & portable device platforms for manufacturing multiplebiologics with high purity, efficacy, and potency, at the point-of-care,in short timeframes ( 24 hours), when specific needs arise Include end-to-end manufacturing chain (including downstream processing)within a microfluidics-based platform Focus on currently approved therapeutics by FDA (i.e. no drug discovery)18

Integrated and Scalable Cyto-Technology(InSCyT) biomanufacturing platformUpstreamAnalyticsMultivariateModelpH, DO, TReal-TimeProcessDataSterility Safety OK forReleaseNot OKSterileMediaPotencyYeastInoculumPurityOn-line Reactor ControlPerfusionCrude Product Hold TanksConcentrationPolishing Membrane#1 g Membrane#2 Tanks1PolishingMembrane #1Fill2Purified Productfor Quality TestingPolishingMembrane teWasteUF/DFMembrane19

Rationale for Pichia pastorisas microbial host for biosimilar productsAdvantages from a regulatory perspective Many products on market or in late-stage development(including one Phase I target) Reduced risk for viral contamination in InSCyT process Human-like post-translational modifications (folding, glycosylation, etc.)Technical benefits Genetically stable organism High density cultivation (culture volume 70% biomass) High yields of secreted proteins (up to 15 g/L) Limited host cell protein (HCP) profile (eases burden on downstream) Amenable to lyophilization20

Integrated and Scalable Cyto-Technology(InSCyT) biomanufacturing ualityanalyticspH, DO, TReal-TimeProcessDataSterility Safety OK forReleaseNot OKSterileMediaPotencyYeastInoculumPurityOn-line Reactor ControlPerfusionCrude Product Hold TanksConcentrationPolishing Membrane#1 g Membrane#2 Tanks1PolishingMembrane #1Fill2Purified Productfor Quality TestingPolishingMembrane teWasteUF/DFMembrane21

Recall Quality by Design approachProcess development involves: Define the targetproduct profile Identify thecritical qualityattributes (CQAs) Select anappropriatemanufacturingstrategy Implement acontrol strategy22

Plant-wide control approachUpstreamMultivariateModelAnalyticspH, DO, T Characteristics of InSCyT––––Sterility SafetySterileMediaMany connected unit operationsMany discrete operationsMulti-product plantAlignment with regulatoryrequirements (e.g., design space) Real-TimeProcessDataOK forReleaseNot OKPotencyYeastInoculumPurityOn-line Reactor rude Product Hold TanksPolishing Membrane ing Membrane #2Tanks1PolishingMembrane #1Fill2Purified Product forQuality TestingPolishingMembrane #2WasteEluteWasteWasteWasteUF/DFMembraneDownstream QbD approach adapted from chemical industry– Employing systematic and modular design of plantwidecontrol strategies for production-scale manufacturing facilities– Using numerical algorithms that can handle discrete operationsand multiple productsR. Lakerveld, B. Benyahia, R.D. Braatz, & P.I. Barton, Model-based design of a plant-wide controlstrategy for a continuous pharmaceutical plant, AIChE Journal, 59, 3671-3685, 201323

Application to biologic drug production Build first-principlesdynamic models foreach unit operation (UO) Design control systemfor each UO to meet“local” material attributes Evaluate performance insimulations and proposedesign modifications as needed Implement and verify the control system for each UO Design and verify plantwide control system to ensurethat the CQAs are met24

Application to biologic drug production Build first-principlesdynamic models foreach unit operation (UO) Design control systemfor each UO to meet“local” material attributes Evaluate performance insimulations and proposedesign modifications as needed Implement and verify the control system for each UO Design and verify plantwide control system to ensurethat the CQAs are met25

Integrated and Scalable Cyto-Technology(InSCyT) biomanufacturing platformUpstreamAnalyticsMultivariateModelpH, DO, TReal-TimeProcessDataSterility Safety OK forReleaseNot OKSterileMediaPotencyYeastInoculumPurityOn-line Reactor ControlPerfusionCrude Product Hold TanksConcentrationPolishing Membrane#1 g Membrane#2 Tanks1PolishingMembrane #1Fill2Purified Productfor Quality TestingPolishingMembrane teWasteUF/DFMembrane26

Local “UO” control for bioreactor unit operations27

Local “UO” Control: Microscale controlled cell culture28Kevin S. Lee et al., Lab on a Chip, 11, 1730-1739 (2011)

Reproducible microbioreactor cultivationsOD600 (Cell Concentration)POD 0POD 1POD 2POD 3Time, hOutgrowth TransitionProduction29

Local “UO” control: Microscale controlled cell cultureKevin S. Lee et al., Lab on a Chip, 11, 1730-1739 (2011)30

Application to biologic drug production Build first-principlesdynamic models foreach unit operation (UO) Design control systemfor each UO to meet“local” material attributes Evaluate performance insimulations and proposedesign modifications as needed Implement and verify the control system for each UO Design and verify plantwide control system to ensurethat the CQAs are met31

Acknowledgments (Novartis-MIT)Paul I. BartonBernhardt L. TroutAllan T. MyersonKlavs F. JensenTimothy F. JamisonRichard Lakerveld (Delft)Brahim Benyahia (Loughborough)32

Acknowledgments (Bioman)Philippe MouriereNate CosperChris LoveRajeev RamAnthony SinskeyTimothy LuMichael StranoJongyoon HanJames LeungGK RajuStacy SpringsPaul BaroneKerry LoveLisa BradburyRalf KuriyelRussell NewtonSusan DexterSteve CramerPankaj KarandeWilliam Hancock33

strategy for a continuous pharmaceutical plant, AIChE Journal, 59, 3671-3685, 2013 Column Tanks 1 2 Downstream Affinity Chromatography Polishing Sterile Media Yeast Inoculum Crude Product Hold Tanks Upstream Multivariate 1 2 Polishing Membrane #1 Tanks Polishing Membrane #1 Membrane #2 Waste Waste Waste Polishing Membrane #2 UF/DF Membrane Waste