Challenges with Establishing a Control Strategy for Biosimilars FDA/PQRI Conference on Advancing Product Quality Bethesda, MD October 5 th Barbara Rellahan MS, PhD Director, Product Quality Amgen Inc
Challenges with Establishing a Control Strategy for Biosimilars
FDA/PQRI Conference on Advancing Product QualityBethesda, MDOctober 5th
Barbara Rellahan MS, PhDDirector, Product Quality Amgen Inc
Integrated Control Strategy
2
• The term “control strategy” refers to the combination of input, procedural and testing controls that ensure a process consistently delivers product meeting product quality attribute requirements
• The level of control for each individual quality attribute is determined on the basis of • the criticality level of the attribute • the capability of the process to consistently deliver product that meets
product quality expectations
• An integrated control strategy includes:
Procedural controls Raw material controls In‐process control (IPCs) testing Process monitoring and
product data monitoring Release specification testing
Stability testing Process validation Characterization testing Control of process variation Comparability testing
Control strategy needs to ensure process consistently meets biosimilar product quality expectations and can support commercial production demands
Clinical Development
Reference Product Characterization
Cell Line DS/DP ProcessDevelopment
ClinicalProduction
Process Validation
Stand Alone Product
Post‐Approval
Formalized integrated control
strategywith continued
process verification
Lifecycle Management of a Biosimilar
Analytical Similarity
Reference product profile defines the target; Design product/process to meet similarity expectations based on characterized reference product range
Development/Evaluation of an Integrated Control Strategy is an Iterative Process Performed Throughout a Product’s Lifecycle
4
Quality Target Product Profile (QTPP): The Quality Target Product Profile is defined as a prospective summary of the quality characteristics of a drug product that ideally will be achieved to ensure the desired quality, taking into account safety and efficacy of the drug product (ICH Q8 (R2))
Product Quality Attribute Assessment (PQAA): Scores attributes based on risk to safety (immunogenicity + toxicology) and efficacy (potency + PK). Highest S/E score = attribute severity score
Product Quality Risk Assessment (PQRA): Evaluates control based on attribute severity, occurrence (unit op impact), and detection (test strategy/capability)
QTPPDevelopment: RP is target
Post‐Approval: Biosimilar is target
Integrated Control Strategy
PQAA
PQRA
Biosimilar Integrated Control Strategy Challenges
5
Development of a control strategy for a biosimilar follows the same principles and ‘looks’ very much like that of an innovator product. Points to consider for a biosimilar product are:
• Condensed clinical development period, introduction of the ‘commercial’ process for first in human trial
• Establishment of attribute control ranges Translation of Tier 1 similarity assessment criteria into QTPP
target range Difficult to identify full allowable attribute range of reference
product (RP). Biosimilars may be required to have tighter acceptable ranges compared to RP
• Establishment of specifications
• Post‐approval comparability
Assessment Timing is Optimized to Support Control Strategy Development ‐ Innovator
Pre-clinical Phase 1 Phase 2 Phase 3 Filing Launch/Post-Launch
Commercial Process Development (CPD)
Commercial Process Development (CPD)
Process/Product Characterization
(P&PC)
Process/Product Characterization
(P&PC)Continued Process
VerificationContinued Process
Verification
Process Qualification
(PPQ)
Process Qualification
(PPQ)
PQRAssessment
PQ Assessment
PQRAssessment
PQ Assessment
PQRAssessment
PQ Assessment
Lifecycle Management
Prior KnowledgeProcess
UnderstandingProduct
Understanding
6
QTPP/PQ Assessment
Raw MaterialAssessment
Raw MaterialAssessment
Raw MaterialAssessment
Identify process improvement needs, commercial analytical requirements, and PC
strategy
Evaluate risk of commercial process
prior to PV and refine as controls as appropriate
Finalize commercial control
strategy (i.e. routine testing)
Reevaluate to maintain and, if
needed, improve control
Biosimilars Have a Condensed Clinical Development Period
7
Identify process improvement needs, commercial analytical requirements and PC strategy
Evaluate risk of commercial process prior to PV and refine as controls as
appropriate
Reevaluate to maintain and, if needed, improve
control
Pre‐clinical Ph1 Phase 3 Filing Launch/Post‐Launch
PPQPPQ
PQRAssessment
PQRAssessment
PQ Assessment
Lifecycle Management
QTPP/PQAssessment Raw Material
AssessmentRaw MaterialAssessment
Continued Process Verification
Continued Process Verification
RP characterization
Biosimilarproduct/process understanding
CPDP&PC
Finalize commercial control
strategy (i.e. routine testing)
Challenges of a Condensed Clinical Development Period
8
• Commercial process implemented at first‐in‐human study initiation
• PPQ preparation, including design of the commercial control strategy, may need to be initiated prior to completion of all process/product characterization work
• Analytical similarity assessment is a repetitive and iterative operation conducted throughout process development. Target reference product (RP) attribute ranges may therefore change during development as experience with the RP increases but needs to be ‘locked’ down as product nears marketing application submission
• Team needs to be flexible and respond quickly to new information
Challenges with Establishment of Attribute Acceptable Ranges – 1
Yi Tsong, FDA, DIA/FDA statistics Forum 2015
• FDA expectations for what constitutes a Tier 1 attribute are unclear. Creates uncertainly around process capability expectations for select CQAs
• Tier 1 attributes are subject to very tight assessment criteria, these may translate into very tight control limits but,
• Tier 1 assessment uses equivalence margins that are difficult to translate directly into a QTPP target range
1) Translation of Tier 1 analytical similarity expectations into a QTPP target range
Challenges with Establishment of Attribute Acceptable Ranges ‐ 1 Tier 1 equivalence margins are difficult to translate into QTPP
target ranges
10
‐1.5 SD +1.5 SD
90% CIOn Mean Difference
Mean ‐ X SD Mean + X SD
Tier 1 equivalence margin is a relative measure of the difference in means
Tier 2 assessment criteria provide a range based on an estimate of the RP standard
deviation
VS
Individual values
Challenges with Establishment of Attribute Acceptable Ranges ‐ 2
*Schiestl, M. et al. 2011. Acceptable changes in quality attributes of glycosylated biopharmaceuticals. Nat Biotech 29:310‐312.
*2) QTPP attribute target ranges may change over the course of development. • RP range available on the
market during the development of a biosimilar may not be reflective of the allowable attribute range for the reference product
• Biosimilar manufacturers may be restricted to a much tighter attribute range compared to that of the innovator
Areas of Uncertainty for the Biosimilar Integrated Control Strategy
12
Establishment of Specifications‐ Regulatory expectations for commercial specifications are not completely clear
EU guidanceThe selection of tests to be included in the specifications (or control strategy) for both drug substance and drug product is product specific and should be defined as described in ICH Q6B
US guidanceNo specific reference to expectations for development of a biosimilar controls strategy in guidance documents
Areas of Uncertainty for the BiosimilarIntegrated Control Strategy
13
Establishment of Specifications• An enhanced, quality by design approach is being used for establishment of
the biosimilar specifications. Incorporate extensive product and process knowledge to optimize control strategy and reduce redundant testing
• Specification acceptance criteria will be based on biosimilar process and product knowledge, and historical data
DS specification exampleParameter Test Method Acceptance Criteria
Identity ELISA Pass
Purity SE-HPLC 1.5% HMW
Adventitious agentsBacterial endotoxins 0.2 EU/mg
Bioburden 10 CFU/10 mL
Potency Bioassay 80% to 120% relative potency
Process impurity clearance validation
Optimized IPC testing Drug product specification
testing Stability testing
Areas of Uncertainty for the BiosimilarIntegrated Control StrategyEstablishment of Specification Acceptance CriteriaBase on biosimilar historical data or incorporate RP information into range?
• Specification acceptance criteria should primarily be based on the biosimilar’shistorical data since that is reflective of pre‐clinical and clinical experience, with consideration for process capability
• Minor differences in clinically inactive components may be present between the RP and biosimilar that could impact specification acceptance criteria (eg, C‐terminal lysine variant by CEX‐HPLC)
• Per ICH Q8 (R2) ‘The degree of regulatory flexibility is predicated on the level of relevant scientific knowledge provided…’ Use of RP knowledge should be allowed if justified by process and product understanding
Specification Acceptance CriteriaHypothetical example
Biosimilar process understanding
2.53
3.54
4.55
5
7
9
11
60
80
100
120
Defect
Attribute A
Biological Activity
Biosimilar product understanding
50
60
70
80
90
100
110
120
0 1 2
BS ActualBS Predicted
RP range
RP range
TIUL
TI LL
Based on demonstrated process control, and process and product understanding (including that of the reference product), it may be acceptable to use RP characterization data to justify acceptable ranges for some attributes0.5 1.0 2.0
Expectations for Post‐Approval Comparability is Another Area of Uncertainty for the Biosimilars
16
Definitions
ComparabilityComparison of pre‐change and post‐change products from the same manufacturer: Knowledge of product clinical experience, process history, and control strategy
SimilarityComparison of a biosimilar product to a reference product from a different manufacturer: Gaps in prior knowledge (animal, safety, and critical studies)
Areas of Uncertainty for the Biosimilar Integrated Control Strategy
17
Post‐approval comparabilityEU Guidance:It is acknowledged that the biosimilar will have its own lifecycle. When changes to the manufacturing process (active substance and/or finished product) are introduced during development, a comparability assessment (as described in ICH Q5E) should be performed…. there is no regulatory requirement for re‐demonstration of biosimilarityonce the Marketing Authorisation is granted.
Health Canada Guidance:Once a Notice of Compliance (NOC) is issue for a Subsequent Entry Biologic it is considered to be a new biologic drug (i.e. a "stand‐alone" product) and regulated accordingly. After a manufacturing change to the Subsequent Entry Biologic or reference biologic drug, a Subsequent Entry Biologic sponsor is not required to re‐establish its similarity to its reference biologic drug.
US Guidance:No specific reference to biosimilar post‐approval comparability expectations in guidance documents
Areas of Uncertainty for the Biosimilar Integrated Control Strategy
18
Post‐approval comparabilityAfter approval, biosimilar products should be considered stand alone products and utilize comparability to support process changes
Biosimilar manufacturer’s can not be locked into the process being used at the time of MA approval. They need to have the ability to make process changes post‐approval
Once approved as a biosimilar, the biosimilar manufacturer can not be tied to the attribute profile of the reference product that is currently on the market since there may be changes in the RP profile compared to the original profile the biosimilar met to gain regulatory approval
Comparability assessments have a long history of providing a robust assessment of the impact of process changes on product quality and are sufficient to control/limit biosimilar attribute shifts after post‐approval changes
Post‐approval continued process verification and product monitoring are sufficient to control/limit process drift
Summary• Biosimilars have a condensed clinical development period
• Biosimilar’s have a unique lifecycle in that during development the characterized reference product range is the target for process development, whereas after approval the biosimilar should become a stand alone product
• Biosimilar specifications should primarily be established based on the biosimilar’s historical knowledge and experience though the characterized RP range may be considered if justified appropriately
• Comparability should be used to support post‐approval manufacturing changes as this will ensure the quality attribute ranges of the biosimilar’s remain consistent with those of the biosimilar’s pre‐clinical and clinical experience including the analytical similarity assessment that supported biosimilar approval
• Rick Burdick
• Barry Cherney
• Gino Grampp
• Simon Hotchin
• Margaret Karow
• Tony Mire-Sluis
Acknowledgements
20