- Key takeaways
- Building the technology transfer package: what it must contain
- How do you transfer analytical methods between sites?
- Engineering runs: the step most programs skip and later regret
- What are the most common QTA negotiation friction points?
- Change control and deviation management: the terms that matter most
- How should IP, data ownership, and liability be handled in the QTA?
Every CDMO partnership has two moments that determine whether it succeeds: the technology transfer, when the manufacturing process moves from the development organization to the CDMO's site, and the quality technical agreement, when responsibility for every quality function in the manufacturing process is negotiated and assigned in writing. Neither receives enough preparation from the sponsoring organization, and both are more consequential than the CDMO selection process that preceded them.
Key takeaways |
|
|
|
|
|
For the strategic context of CDMO selection and the full scope of the CDMO partnership lifecycle, see the related article on navigating GMP facilities and outsourcing. For the fill-finish-specific technology transfer considerations, see the related piece on sterile fill-finish operations in commercial bioproduction.
Building the technology transfer package: what it must contain
A technology transfer package is the complete set of documentation that enables a CDMO to understand, reproduce, and GMP-manufacture a biopharmaceutical process without relying on real-time knowledge from the originating organization. Best practices research on biopharmaceutical technology transfers confirms that a successful transfer is more than a document handoff: it encompasses the transfer of knowledge and experience to the commercial manufacturing unit to ensure consistent, safe, and high-quality product. The implication is that the documents must be accompanied by expert knowledge transfer, typically through a series of technical meetings and on-site demonstrations, to convey the process understanding that documents alone cannot capture.
The required components of a technology transfer package for a biologic drug substance include: the complete batch manufacturing record in a format that can be adapted to the CDMO's quality system; the process description and process flow diagram; all analytical methods with their validation data and qualification status; raw material specifications and qualified supplier lists; cell bank vials with characterization and release data; equipment and instrument qualification criteria for the key equipment the process uses; and the process development history document.
The process development history document is the most valuable and most frequently absent element of a technology transfer package. It explains why current process parameters were chosen, what alternatives were considered and rejected, and what failure modes the current parameter settings were designed to prevent. A CDMO that receives a batch manufacturing record without the process development history knows how to run the process but does not know how to troubleshoot it when it encounters a different equipment environment, a different raw material lot, or a process deviation that is not covered in the batch record instructions.
Package component | What it must contain | Common gap that delays transfer |
Batch manufacturing record (BMR) | Complete step-by-step manufacturing instructions, in-process controls, acceptance criteria, and deviation handling guidance for every unit operation | BMR written for internal use, not transfer: assumes equipment-specific knowledge, refers to instrument IDs rather than specifications, or omits rationale for critical steps |
Analytical method package | Full analytical method procedures, system suitability criteria, qualification/validation data, reference standard information, and instrument specifications for each QC release and in-process method | Methods validated on specific instruments not available at CDMO; reference standards not transferable without regulatory change; data files in proprietary formats requiring specific software versions |
Raw material specifications | Specifications and qualified supplier lists for all critical and non-critical raw materials, with characterization data supporting the specification requirements | Single-source raw materials without qualified alternatives; supplier qualifications that have not been updated; incomplete characterization data for critical raw materials |
Cell bank package | Master cell bank (MCB) and working cell bank (WCB) vials with full characterization, safety testing, and release data; vialing records and storage conditions | Cell bank safety testing incomplete for new regulatory requirements; inadequate vial inventory for CDMO qualification and extended commercial manufacturing; missing adventitious agent testing data |
Process development history | Summary of key process development decisions: which parameters were optimized, what ranges were studied, what failure modes informed current parameter choices, and what known process risks exist | Absent entirely or summarized at a level too high to convey practical process understanding; not updated to reflect last 12 to 24 months of development changes |
Scale-down model qualification | Evidence that the small-scale process model used in process development is representative of the commercial-scale process at the CDMO, including bioreactor hydrodynamics, mass transfer, and oxygen delivery equivalence | Scale-down model not developed or not qualified; equivalence not demonstrated between sponsor's scale-down model and CDMO's available equipment |
How do you transfer analytical methods between sites?
Analytical method transfer is the process of demonstrating that an analytical method performs equivalently at the receiving CDMO site as at the sending organization, using the CDMO's instruments, reagents, and laboratory personnel. The current ICH framework for analytical method development and transfer is guided by ICH Q14, which introduces a structured analytical quality by design (AQbD) framework for analytical method development, optimization, validation, and lifecycle management. Methods developed under AQbD principles, which characterize the method operating space around the target analytical procedure, have a better-defined design space and are therefore more transferable across sites than methods developed by trial-and-error optimization against a single set of instruments.
The standard approach to method transfer is a comparative study in which both the sending and receiving laboratories analyze the same set of samples using the same method. Results are compared using pre-defined equivalence acceptance criteria: the receiving laboratory's results must fall within the defined equivalence range of the sending laboratory's results for each analytical parameter. For quantitative methods, equivalence ranges are typically set based on the method's historical precision data. For qualitative methods, agreement between classification outcomes is assessed.
Methods that fail the equivalence comparison during transfer typically do so for one of four reasons: instrument differences, where the CDMO's instrument model or configuration differs from the sending organization's in a way that affects performance; reagent or consumable differences, where lot-to-lot variability in reagents, columns, or other consumables affects results; reference standard differences, where the reference standard lot used during validation is not the same as the lot available at the CDMO; or analyst-related differences, where the method procedure has not been documented with enough detail to eliminate interpretation differences between operators. Each cause has a different remediation path.
The analytical method suite for a biopharmaceutical product typically includes 15 to 30 individual methods, ranging from standard wet chemistry assays such as pH and osmolality to complex biological assays such as cell-based potency assays. The complexity and transfer difficulty vary enormously across this suite. Standard physicochemical methods transfer in weeks. Complex cell-based potency assays, which require the receiving laboratory to establish and validate a cell line and assay procedure that produces equivalent results to the reference method, can take 6 to 12 months to transfer and qualify for GMP use.
Engineering runs: the step most programs skip and later regret
An engineering run is a non-GMP manufacturing run performed at the CDMO site before the first GMP batch, using the transferred process documentation and the CDMO's equipment, personnel, and raw materials. It is the practical test of whether the technology transfer package was complete, whether the CDMO's equipment produces the same performance as the originating site's equipment under the transferred conditions, and whether any process gaps exist that would otherwise be discovered for the first time during a GMP run.
Engineering runs are avoided by programs that are under timeline pressure, because they add cost and calendar time before the first GMP batch. The calculus of this decision consistently underestimates the cost of a GMP failure relative to the cost of an engineering run. A failed GMP batch requires batch rejection, investigation, remediation, and repeat manufacture, each of which takes longer and costs more than the engineering run that might have prevented the failure. The repeat GMP batch also consumes a manufacturing slot that may have been scheduled weeks or months in advance, adding further delay.
Scale-down model qualification, which demonstrates that the small-scale process model used in development is representative of the CDMO's commercial-scale equipment, is a related activity that should precede GMP manufacturing. Research from UCB Pharma on technology transfer and scale-down model development for biotherapeutics in mammalian cells confirmed that scale-down models must account for scale effects and be representative of the proposed commercial process to be reliable predictors of commercial-scale performance. A scale-down model developed at the originating organization on different equipment than the CDMO uses is not a qualified model for predicting CDMO commercial-scale behavior without qualification data at the CDMO.
What are the most common QTA negotiation friction points?
The quality technical agreement negotiation follows a predictable pattern across most CDMO partnerships. The operational provisions (which laboratory runs which tests, who reviews the batch record, and what the quality release process looks like) are negotiated relatively smoothly because both parties have aligned interests in establishing a functional quality system. The friction concentrates in the provisions that address failure: what happens when a batch fails, who investigates a deviation, and who pays when something goes wrong.
Batch rejection and repeat manufacture cost allocation is the highest-stakes QTA negotiation point. When a GMP batch fails to meet release specifications, the question of who bears the cost of the rejected batch and the cost of repeat manufacture is determined by the QTA in combination with the manufacturing agreement. If the failure is attributable to a CDMO process execution error, the CDMO typically bears the cost. If the failure is attributable to inadequate technology transfer documentation, the allocation may be disputed. If the failure is due to raw material variability from a sponsor-qualified supplier, the allocation depends on whether the QTA assigns material qualification responsibility to the sponsor or the CDMO.
Deviation investigation timeliness is a QTA provision that generates ongoing friction after the agreement is signed, because the timelines that work for the CDMO's internal quality system are not always aligned with the timelines the sponsor needs to meet its own regulatory and clinical obligations. A program that needs a deviation investigation closed within 30 days to maintain a clinical trial filing timeline will experience friction with a CDMO whose standard deviation investigation process runs 60 to 90 days. This timeline should be explicitly negotiated in the QTA rather than assumed from the CDMO's standard language.
QTA friction point | Sponsor concern | CDMO position | Resolution approach |
Batch rejection cost allocation | When a batch fails, the sponsor needs clear allocation of re-manufacturing cost that does not leave them bearing CDMO execution errors | CDMO accepts responsibility for execution errors but resists liability for failures attributable to sponsor's process design, documentation, or materials | Define categories of failure cause explicitly and map each category to responsibility allocation; require root cause investigation before final cost allocation; cap CDMO liability to batch cost |
Deviation investigation timeline | Sponsor needs investigation closed within defined timeline to meet regulatory submission deadlines; standard CDMO timelines may be 60 to 90 days | CDMO needs sufficient time for thorough investigation; premature closure creates regulatory risk; resource constraints limit simultaneous investigation capacity | Negotiate tiered timelines by severity: critical deviations within 15 to 30 days, major within 45 to 60 days, minor within 90 days; define what constitutes each tier explicitly |
Change control notification threshold | Sponsor needs prior notification and approval for any change affecting product quality or regulatory commitments; broad notification protects against supply disruption | CDMO needs operational flexibility to make routine improvements without sponsoring approval overhead on every minor operational decision | Define categories of change by risk level; require prior approval for equipment changes, supplier changes, and process parameter changes outside ranges; allow CDMO discretion for like-for-like consumable substitution within qualified specifications |
Audit and inspection rights | Sponsor needs right to conduct facility audits and be present for regulatory inspections; advance notice should not be required for cause audits | CDMO needs to manage audit scheduling to avoid operational disruption; confidentiality of other client information limits sponsor access during some activities | Negotiate annual routine audit with 30-day notice; cause audit with 5-business-day notice; inspection presence rights with notification on receipt of inspection notice |
Change control and deviation management: the terms that matter most
Change control and deviation management are the operational quality processes that run continuously throughout the CDMO relationship after the QTA is signed, and they are where the practical implications of the QTA language become visible. A QTA that was negotiated as an agreement document without operational input from the people who will manage changes and deviations day-to-day will generate friction at every change and every deviation, because the language will not accurately reflect how the processes should work in practice.
The change control provisions in the QTA should define at minimum: the categories of change that require prior sponsor notification versus notification after the fact; the categories of change that require prior sponsor approval before implementation; the process for regulatory impact assessment when a change affects regulatory commitments; and who is responsible for preparing and submitting regulatory notifications when the change requires one. The last point is frequently left ambiguous in CDMO QTAs, which state that both parties will cooperate on regulatory submissions without specifying who drafts the submission and who reviews and approves it.
Deviation management provisions should address: the timeframes within which deviations must be reported to the sponsor; the level of information required in the initial deviation notification versus the final investigation report; the sponsor's rights to participate in deviation investigation and root cause analysis; the process for reaching agreement on the impact assessment and CAPA when the sponsor and CDMO have different views; and how deviations that implicate both the CDMO's manufacturing process and the sponsor's process design are jointly investigated.
How should IP, data ownership, and liability be handled in the QTA?
Intellectual property provisions in the QTA define what belongs to the sponsor, what belongs to the CDMO, and what is jointly owned from the work performed under the manufacturing agreement. The fundamental principle is that process improvements and innovations developed by the CDMO using the sponsor's proprietary process knowledge belong to the sponsor, while general manufacturing improvements developed by the CDMO that are not specific to the sponsor's product belong to the CDMO. The practical application of this principle is frequently contested, because the boundary between product-specific improvements and generally applicable manufacturing innovations is not always clear.
Data ownership and confidentiality provisions should specify: who owns the manufacturing data, batch records, deviation records, and stability data generated under the agreement; what data the CDMO may retain after the agreement ends; whether and under what conditions the CDMO may use the sponsor's product as a reference case in their commercial literature; and how electronic data is secured, backed up, and transferred if the manufacturing agreement ends. The increasing digitization of GMP manufacturing makes data ownership more significant, because the data generated during commercial manufacturing can have substantial value for future regulatory filings and post-approval process improvements.
Liability limitation clauses typically cap the CDMO's total liability to the sponsor under the agreement at some multiple of the annual fees paid, or at the cost of the specific batch that gave rise to the claim. These caps protect the CDMO from unlimited exposure for consequential damages that could dwarf the manufacturing fees if a quality failure delays a product launch. Sponsors should evaluate whether the cap is sufficient to cover the realistic downside scenarios for their program, and negotiate carve-outs from the cap for willful misconduct, gross negligence, and third-party claims arising from the CDMO's failure to comply with its quality obligations.
For the strategic view of how technology transfer and QTA negotiations fit into the broader CDMO selection and partnership management process, see the hub article on the CDMO playbook.
This article was produced under Drug Discovery News' AI Editorial Guidelines.
