FDA Perspective on Continuous Manufacturingccpmj.org/downloads/1_Dr.Lee_rev(en).pdf · – ICH Q8,...
Transcript of FDA Perspective on Continuous Manufacturingccpmj.org/downloads/1_Dr.Lee_rev(en).pdf · – ICH Q8,...
FDA Perspective on Continuous Manufacturing
Sau (Larry) Lee, Ph.D.Director & Emerging Technology Team Chair
Office of Testing and ResearchOffice of Pharmaceutical Quality
US FDA Center for Drug Evaluation and ResearchConsortium on Continuous Pharmaceutical Manufacturing
December 13, 2018Tokyo, Japan
2
The Desired State
The Vision“A maximally efficient, agile, flexible pharmaceutical manufacturing sector that reliably produces high quality drugs without extensive regulatory oversight.”
2
3
Office of Biotechnology Products
Office of Pharmaceutical Quality
4
1. COLLABORATE: Strengthen OPQ’s collaborative organization– Leverage a collaborative culture, an engaged and
empowered workforce, streamlined processes, and effective teaming to ensure an efficient, high‐performing, innovative, and results‐oriented organization
2. INNOVATE: Promote availability of better medicines– Minimize barriers to encourage innovation within FDA and in
the manufacturing sector through sensible oversight, research, risk‐based decision‐making, and continuous process improvement
3. COMMUNICATE: Elevate awareness and commitment to the importance of pharmaceutical quality– Effectively communicate the importance of quality and that
the American public can trust their drugs
4. ENGAGE: Strengthen partnerships and engage stakeholders– Build productive relationships with business partners within
and outside FDA and jointly foster effective stakeholder engagement to meet the needs of the American public
4
OPQ Strategic Priorities: 2018‐2022
5
Emerging Technologies Key to Addressing Pharmaceutical Manufacturing Challenges• Address the underlying causes of product recalls and drug shortages– Two thirds of drug shortages resulted from product‐specific quality
failures or general manufacturing facility issues– Product recalls has surged over the past couple of years
• Facilitate new clinical development – precision medicines– Enable a wider range of novel dosage forms, a wider range of
doses without extensive alterations of the process, and convenient fixed‐combination dosage forms
• Improve manufacturing efficiency – Increase process robustness– Lower manufacturing costs for pharmaceutical products– Increase supply chain flexibility
6
Emerging Technology ProgramEncourage and support the adoption of innovative technology to modernize pharmaceutical development and manufacturing through close collaboration with industry and other relevant stakeholders
Mission
ETTOPF
OLDP
ONDPOPPQ OBP
OC
ORA
OTR
A small cross‐functional Emerging Technology Team (ETT) with representation from all relevant FDA quality review and inspection programs (OPQ/CDER & ORA)
7
Program Objectives1. To serve as a centralized location for external inquiries on novel
technologies2. To provide a forum for firms to engage in early dialog with FDA to support
innovation3. To ensure consistency, continuity, and predictability in review and
inspection4. To identify and evaluate potential roadblocks relating to existing guidance,
policy, or practice5. To help establish scientific standards and policy, as needed6. To facilitate knowledge transfer to relevant CDER and ORA review and
inspection programs7. To engage international regulatory agencies to share learnings and
approaches
Contact us: CDER‐[email protected]
8
Why CM?• FDA has identified CM as an emerging technology• FDA recognizes that CM has the potential to increase
the efficiency, flexibility, agility, and robustness of pharmaceutical manufacturing– Integrated processing with fewer steps
• No manual handling, increased safety• Shorter processing times
– Smaller equipment and facilities• More flexible operation• Lower capital costs, less work‐in‐progress materials• Reduced environmental foot print• Feasible to manufacture small batch sizes
– On‐line monitoring and control for increased product quality assurance in real‐time• Amenable to Real Time Release Testing approaches
• Benefits to both patients, and industry
9
CDER Progress• Over 30 requests accepted to the Emerging Technology Program since the launch of the program in late 2014 Over 60 ETT‐industry interactions (including both t‐con and face‐to‐face meetings) and ~50% of these interactions related to CM https://www.fda.gov/AboutFDA/CentersOffices/OfficeofMedicalProductsandTobacco/CDER/ucm523228.htm
• FDA approvals of applications utilizing continuous manufacturing (CM) Vertex ORKAMBI
(lumacaftor/ivacaftor) Janssen Prezista
(darunavir) Eli Lilly Verzenio
(abemaciclib) Vertex Symdeko
(tezacaftor/ivacaftor and ivacaftor) Pfizer Daurismo
(glasdegib)
10
CDER Current Experience• CM processes for drug product
Direct compression, dry and wet granulation; CM models for solid orals; modular CM processing system
• CM processes for drug substance Continuous drug synthesis (flow reaction)
plus batch or continuous crystallization
• End‐to‐end CM processes Integrated synthesis, purification, and final
dosage formation
• Pharmacy on Demand Miniaturized, flexible manufacturing
platforms using CM technologies
• Continuous bioprocesses Continuous purification platform for
monoclonal antibody (e.g., continuous chromatography and viral filtration)
• Control Strategy Increasing use of active process control, PAT
tools, RTRT, process models for material traceability, non‐conforming material diversion, and blend uniformity
11
CM Elements Discussed Between CDER and Industry
Control Strategy
Facility and Process
Validation
Batch and Lot definition
Bridging Batch to Continuous Manufacturing
• System dynamics, RTD and transient disturbances
• Raw material control• Sampling strategy• Material traceability• Start up and shut down• PATs, models and RTRT• Product collection and
material diversion• Process monitoring and
control
• System dynamics, RTD and transient disturbances
• Raw material control• Sampling strategy• Material traceability• Start up and shut down• PATs, models and RTRT• Product collection and
material diversion• Process monitoring and
control
• Model maintenance and update
• System integration, data processing and management
• Equipment qualification, maintenance and cleaning
• Process performance qualification
• Continued process verification
• Model maintenance and update
• System integration, data processing and management
• Equipment qualification, maintenance and cleaning
• Process performance qualification
• Continued process verification
• Batch size – flow rate and run time
• Scale up – run time increase
• Primary stability batch size
• Mass balance or yield
• Batch size – flow rate and run time
• Scale up – run time increase
• Primary stability batch size
• Mass balance or yield
• Minor formulation changes
• Comparability and bioequivalence
• Stability data package
• Minor formulation changes
• Comparability and bioequivalence
• Stability data package
Examples
12
Quality Assessment Focus• Evaluation of the proposed attributes and specifications of raw
materials– Impact of variations in material properties on the performance of CM and
product quality• Characterization of process dynamics for critical steps and integrated
system – Understanding of the feeder dynamics, residence time distribution (RTD), and
system response to transient disturbances • Process monitoring and control strategy
– Monitor and detect transient disturbances and process deviation– PAT and active process controls
• Material collection and diversion– Start up and shutdown– Strategy to identify, isolate and
divert non‐conforming materials• Real‐time release testing
– PAT tools for assay and content uniformity– Models for product release (e.g., dissolution)
13
Process Understanding
Use the understanding of the impact of process parameters and material attributes on product quality to:• Establish design space based on the
design of the experiments
• Build predictive models and simulation tools (ICH Q8)
• Inform alarm and action limits and an approach to manage process deviations (e.g., adjustments)
• Establish criteria for incoming and in process materials
BoukouvalaFet.al.Comput.Chem.Eng. 2012;42;30‐47
14
Process Dynamics
AdityaandMuzzio.PowderTechnology2011;208,26‐36
• Evaluate response to set point changes (e.g., change in line rates)
• Assess the impact of Startup and Shutdown on material quality
• Evaluate back‐mixing of the system over time to predict the propagation of disturbances and materials through the system
• Obtain an understanding of process dynamics by characterizing the Residence Time Distribution (RTD)
• Identify typical failure modes or deviations (long term vs. short term) (e.g., feeder variability)
15
• For CM, this can include integration of processparameter limits (set points and alarms), in‐process monitoring (including PAT), process controls (feedback and feed forward), material diversion, and Real Time Release Testing (RTRT)
• Many continuous manufacturing systems promote the adoption of higher level controls, although a hybrid approach combining the different levels of control is viable for some continuous manufacturing process designs
Control Strategy – State of Control• A control strategy should:
‒ Be appropriate for each individual process and product based on the risks to product quality
‒ Consistently provide assurance of process performance and quality
‒ Be designed to mitigate product quality risks in response to potential variations over time for CM
Level 1Active or real time automatic control
Level 2Appropriate end product testing + Material attributes and process parameters within the
established design space
Level 3End product testing + tightly controlled material
attributes and process parameters
S.L. Lee et al., J Pharm Innov. 2015; DOI 10.1007/s12247‐015‐9215‐8
16
Raw Material Control Considerations• Use of multiple raw material lots in a
batch– Establish traceability of different lots to
finished products
• Characterization of input materials– Evaluate raw material attributes (e.g., particle
size distribution and density) affecting the formulation flow behavior, segregation potential, etc.
• Appropriate material specifications– Impact of drug substance or excipient lot‐to‐lot
variations on feeding– If legacy product, appropriateness of the
existing drug substance specifications for CM– Appropriateness of the compendial
specification for excipients
Measured properties of individual materials
(X)[m x n]
Blend Ratios (R)
[k x n]
Measured properties of blends
(Y)[k x l]
materials
prop
ertie
s
blen
ds
properties
Operating Conditions
(Z)[k x j]
conditions
S.G.Munozetal.(2014)ChemometricsandIntelligentLaboratorySystems.133,49‐62
17
Process Monitoring and ControlSpecify the role of PAT and Models• Provide process understanding during development; process monitoring during production;
process control; and/or real‐time release testing (RTRT) method
Consider instrument aspects• Interference due to flow; time of acquisition vs. flow rate; probes – number, location, probe
failure, probe maintenance, etc.
Feeding: a critical operation for CM• Demonstrate that acceptable quality material is manufactured near the upper and lower limits to
support feeding limits• Evaluate impact of operational variations (e.g., switching from gravimetric to volumetric flow
during feeder refill)• Assess impact of feeding variations of excipients on product performance (e.g., dissolution)
Engisch W. and Muzzio F.. J Pharm Innov. 2015; DOI 10.1007/s12247‐015‐9238‐1
Feeder Monitoring
18
Diversion of Non‐Conforming Material• The ability to isolate and reject non‐
conforming material can be one of the key aspects of a CM control strategy– Planned process start‐ups and shutdowns– Temporary process disturbances or upsets
• The evaluation and understanding of propagation of a disturbance in the system are important to justify the amount of material at risk
• Models of process dynamics are being assessed as part of the control strategy to detect and track non‐conforming material due to upstream disturbances
19
Scientific Considerations for Model Based Material Diversion• Develop models using scientifically‐sound principles and conditions
that reflect routine commercial production• Validate performance for high‐impact models
– Capability of the model to trace the identified non‐conforming material segment through the system to the rejection point
– ICH Q8, Q9, & Q10 Questions and Answers ‐‐ Appendix: Q&As from Training Sessions (Q8, Q9, & Q10 Points to Consider)
• Understand model assumption and risks to validity of model predictions– Model parameter uncertainty – Expected variations in process parameters and material attributes (e.g., line rate)– Product quality risks resulting from potential transient disturbances – Process failure modes that may not be identified by or included in the model
• Include model maintenance approaches within the quality system as part of a lifecycle approach– Routine monitoring to verify performance– Model updates
20
RTRT Considerations• Establish a valid combination of assessed material attributes and associated
process controls in relation to the final product quality• Evaluate ability of the sampling scheme(s) to detect non‐conforming
materials or products– Assess quality of a batch (i.e., % confidence, % coverage, and target range) – Monitor or assess system dynamics (i.e., disturbances) during the continuous operation – Determine whether the process is in a state of control during start‐up, shut‐down, and after
restarts
• If the on‐line PAT methods are submitted as routine methods (without alternatives), describe what actions will be taken when analyzer is not available
21
Batch Definition• 21 CFR 210.3 defines a batch as “a specific quantity of a drug or other
material that is intended to have uniform character and quality, within specified limits and is produced according to a single manufacturing order during the same cycle of manufacture”.
• Additionally, a lot is defined as “a batch, or a specific identified portion of a batch, that has uniform character and quality within specified limits; or, in the case of a drug product produced by continuous process, it is a specific identified amount produced in a unit of time or quantity in a manner that assures its having uniform character and quality within specified limits.”
Definitions for both “batch” and “lot” are applicable to continuous processes
22
Batch Definition Considerations• Regulatory expectation that:
− Product has “uniform character and quality within specified limits” and is therefore closely linked to the control strategy that is designed to ensure the process under a state of control
• Potential batch definitions based on:– Production time period; amount of material processed; production variation (e.g.
different lots of feedstock); amount of product produced; and others– Established prior to initiation of manufacturing, not after the fact
• Other considerations– Ensure material traceability to verify a complete history of the manufacture,
processing, packing, holding, and distribution of a batch/lot of the product and other materials (excipients);• Especially in cases of OOS/OOT investigations, consumer complaints, product recalls, or
any other situations that may have public health impact– Define procedures for start‐up/shutdown, and establishing a priori acceptance criteria
for determining when product collection starts– Material reconciliation including handling of non‐conforming materials
23
Facility Considerations• Adjustments to existing facility pharmaceutical quality system (PQS)
– Updates to Quality and production procedures– Quality oversight of automated controls, process
data, RTRT, and electronic batch records
• Quality evaluation when material is diverted and quarantined – Level of investigation, root cause analysis, corrective and preventive action,
understanding diversion event (common vs. unexpected) for continuous improvement, etc.
• Process Validation, readiness for commercial manufacturing, and knowledge management– Demonstration of robustness, process monitoring, and broader control strategy– Assessment of change controls for total impact
• Integrated equipment train– Knowledge gained from equipment qualification to support the proposed batch
size or run time– Cleaning validation, maintenance, and performance monitoring to support
commercial lifecycle and multiproduct manufacturing• Additional controls for incoming raw materials
24
ParadigmShiftforPQS
‐ Discrete unit operations‐ In‐process materials are collected/tested at the end of each unit operation‐ Hold time studies allow for investigation
‐ Integrated unit operations‐Material is constantly generated and moving‐ Data‐rich manufacturing environment‐ QC oversight must be built into process decision‐making
S. Lee et al., J. Pharm. Innov., 2015US FDA Guidance to Industry: Process Validation: General Principles and Practices , ICH Q10
ContinuousProcess
Verification
QualityControlUnitneedstoadaptitsfunctiontoCMrequirements
Knowledge&QualityRiskManagement
25
Process Validation• Consult process validation guidance and verify performance of
the process using the intended control strategy Demonstrate robustness of the process including the ability to remain in a state
of control and make quality decisions in real‐time
• Process Qualification To examine a run time or manufacturing period that should be representative of
the intended commercial run time for the initial product launch• Continuous process verification
Use in‐line, on‐line, or at‐line monitoring or controls to verify process performance on an on‐going basis
Evaluate trending for further process understanding and improvement Provide the advantage of enhanced assurance of intra‐batch uniformity,
fundamental to the objectives of process validation
• Expect retrospective data/trending analysis as part of the process validation guidance and lifecycle management
• Comparability protocol to increase the batch size post approval
26
Quality Control Unit and Automation• Quality decision making must be programed
– Interlocks, communication checks, recipe integrity, data collection frequency– Monitoring , alert & alarm limits, segregation points, automatic stops– Start up sequence, restart, material collection criteria, and shut‐down processes
• Some oversight is delegated to the automation but the QCU is ultimately responsible for the product
• Design, validation, and qualification of automation with equipment is critical
• Maintenance of Automation Control System– Robust change control process for code improvements– Monitoring of ACS performance– Version control, back‐ups, and security
27
Knowledge Management• Documentation: development reports, technology transfer
activities, process validation protocols and reports, batch records• Establish programs to collect, analyze, and maintain data related to
product quality– Design a monitoring plan: attributes/metrics, frequency of
trending/analysis, statistical approach(es)– Intra‐batch and inter‐batch comparisons– Long term assessment of impact of raw material attributes on system
dynamics and process – Equipment performance indicators over time– PAT Model maintenance activities
• Monitor and improve process capability
28
Future Topics for Collaborations• CM and performance‐based approach
– Promote the use of PATs for high frequency measurements of critical quality attributes
– Promote active controls and process improvement– Can this performance based approach be applied to the entire integrated
continuous manufacturing line?• CM and parametric‐based approach
– Require comprehensive understanding of quantitative, multivariate relationships among material attributes, process conditions, and critical quality attributes
– Big data set to develop and validate multivariate models– Complexity increases dramatically with increasing number of parameters– Monitor the “health” of a process to make quality decision
• Combination of both performance‐ and parametric approaches
Performance‐based approach Parametric‐based approach
Process
Materials
29
Future Topics for Collaborations• Model development, validation, maintenance and
update– Regulatory standards for model validation currently lacking– Mostly considered as a “high impact” (e.g., PAT chemometric and dissolution models)
– What does industry want?• Consistency among regulatory agencies for reporting requirements for model maintenance and update
• Most updates or changes managed under the company’s PQS (e.g., post approval change management plan)
• Continuous bioprocessing– Emerging hot topic that should be one of the main focus in the near future
– Lag behind small‐molecule CM in many aspects– Need some breakthrough in PAT areas
30
Future Topics for Collaborations• Data management and storage
– Large size of process and quality data due to high‐frequency measurements
– All raw data be strictly managed according to CFR 211.180?– Technological advance needed for data storage– Data transformation or reconstruction– Areas related to data transformation needed to be addressed
• Loss of information and hinder investigation in the event of a product recall or adverse event
• Loss of information for trending analysis and continuous improvement• Data integrity
– Cloud computing and AI
31
Concluding Remarks• No regulatory hurdles for implementing CM
– Both the Agency and industry are gaining experience
• Recommend early and frequent discussion with the Agency during CM development– Emerging Technology Program should be utilized for early FDA‐Industry
interactions even before the drug molecule is identified
• Process understanding is key to identifying product quality risks and developing a robust control strategy
• A robust control strategy for a CM process can include a combination of different scientific approaches
• FDA supports the implementation of CM technologies using science and risk‐based approaches
Thank You!