Issue link: https://www.e-digitaleditions.com/i/1494960
32 Pharmaceutical Technology ® Quality and Regulatory Sourcebook March eBook 2023 PharmTech.com K nowledge ManageMent Most qua lit y quar tets are accompanied by par- allel counter parts, given that CQAs are normally supported by more than a single CPP, at the same or different stages of a process, along with their related CAs and CDEs. Recommended specification and visu- alization is via a one-page x four-box data sheet (or a digital equivalent) for each quality quartet, whereby context is transparent and both process understand- ing and knowledge gaps are immediately apparent. Ideal qualit y quartets are analogous to pyramids, with CQAs at the base and CDEs at the apex, with each layer protected by progressively tighter tolerances (i.e., increased precision) of the adjoining upper layers. From an engineering perspective, they are self-reg- ulating, and equivalent to shock absorbers operating in support of C&Q/QRM optimization. For standard platforms in particular, most quality quartets and their contents are predictable, reusable, and amenable to codification based on bidirectional "pairing rules" between their parts. Quality and risk are viewed as two sides of a plus-minus coin, best managed as parallel rather than sequential activities. For example, if accuracy is a quality requirement, in- accuracy is the related risk. Rather than being docu- mented independently, ICH Q8 control strategy (CS) is interpreted as the sum of quality quartets for a pro- cess and is a direct beneficiary of the paradigm. The ICH Q12 requirement for established conditions (ECs) to be documented and reported is also satisfied. Quality quartets can be symbolized as strings, with each qualified part assigned a validity status in addi- tion to its dataset, within the string. Working from right to left in conjunction with the lifecycle, each qualified part contributes 0.25 to an overall score of one for the "string quartet." This can also be done for parallel and subsequent quartets, and at a higher level, for entire manufacturing systems and process trains. This monograph defines quality quartets as "knowl- edge management outputs that provide a continuous and testable expression of the understanding that a firm possesses about the relationship between CQAs and CPPs and their associated CAs and CDEs. Figu- ratively speaking, quality quartets have a valency of four." They are introduced here as the missing KM link within C&Q and QRM, and the paradigm is in- tended to be used and shared by progressive manu- facturers, regulators, and other stakeholders. Biopharma specifics Biopharmaceutical manufacturing systems fall into these five main categories: facilities and equipment; lab- oratory controls; raw materials; packaging and labeling; and production. Depending on project scope, product transportation is considered an additional manufactur- ing operation and system. The manufacturing process is divided into drug substance (DS) and drug product (DP) operations, and these can occur at shared or separate buildings or geographical locations. DS operations include upstream and downstream processing (USP and DSP), supported by ancillary oper- ations such as media and buffer preparation, purified water generation/distribution, and so on. USP oper- ations include cell culture and clarification, and DSP involves product capture and purification. The overall process is a batch process, with some individual oper- ations performed on an internally continuous basis. Material sampling, and transfers between operations are largely manual, while the process operations them- selves are primarily automated. Processes typically employ a combination of permanent and single-use equipment and disposable bags (all of which are stan- dard), with cleaning requirements eliminated in the case of single-use equipment. Biopharmaceutical facilities can operate in ded- icated or multi-purpose/campaign modes. Several are owned by contract manufacturing organizations (CMOs) engaged in the delivery of outsourced products. The majority are designed for commercial operation, with process development activities performed within pilot plant facilities on the same or at separate loca- tions. There is a global drive to digitalization, as man- ifested by the utilization of electronic batch records, paperless standard operating procedures (SOPs), train- ing records, and qualification protocols. Regulations Requirements in relation to validation exist in all jurisdictions, with FDA considered the historical leader in this space. FDA's revised validation guide outlines a three-stage lifecycle approach, as follows: process design, process qualification, and continued process verification (2). C&Q activity is positioned as the initiator of stage 2, on the proviso that CQAs and CPPs are known and specified as part of the comple- tion of stage 1. In Europe, the European Medicines Agency's (EMA) Annex 15 follows a similar approach to FDA, using slightly different terminology in cer- tain cases (3). Within the validation guides, there is a transition from a fragmented and one-off activity to an integrated and risk-based lifecycle, enabled by a combination of process understanding and engineer- ing/supplier know-how. This equates to a directive to stakeholders to define workf lows and datasets, and deliver evidence, that "follow the molecule." Both FDA and EMA were proactive in the provision of regulatory relief to manufacturers in the course of the COVID pandemic. EMA in particular posted a for- mal notice to stakeholders on its website to this effect (4). In addition to FDA/EMA, the Pharmaceutical In- spection Co-operation Scheme (PIC/S) is a non-binding, informal co-operative arrangement between regula- tory authorities in the field of good manufacturing