Issue link: https://www.e-digitaleditions.com/i/1499694
www.biopharminternational.com Partnerships for Outsourcing eBook May 2023 BioPharm International ® 17 Process De veloPment begin earlier so that the batch readiness activities (in- cluding sample plans, batch records, and execution training) could be started. The answer lay in developing a dual-purpose road- map. The timing for each study was plotted, includ- ing available analy tical results and draf t reports, enabling visibility of when the first data points were available, while also building a critical-path schedule to drive accountability for both internal and exter- nal deliverables. Timings of the project were laid out against personnel and equipment resources to iden- tify challenges and opportunities. Risks to go forward without each individual study's outcome were also assessed. To mitigate risks, at- scale data, including normal operating ranges and release results from past batches were leveraged. For example, data on filter throughputs could de-risk filtration studies, and the age of prepared growth medium for cell culture during previous batches could de-risk aged media studies. In the longer term, although some aspects of this strategy could be lev- eraged, not all of the f urther investigations could be informed by past data. Process characterization pushed parameter ranges beyond any normal oper- ating ranges, and this information gap represented a risk that always requires careful assessment. While this roadmap allowed some studies to be moved of f t he cr it ica l pat h, cer ta in cr ucia l ones would still not have been ready within the program's time constraints. Each st udy had been itemized, noting those that would likely impact the schedule, and it became clear that by promoting the analytical program for one upstream study further up the time- line, the others would be feasible within the deadline. Similarly, moving a single downstream study would improve the overall readiness, and although others would be conversely delayed, the impact would be less, as these were identified as not being on the crit- ical path. Although this changing of priorities allowed the star t of the upstream execution readiness activi- ties, no batch readiness documentation (e.g., batch records) could be f ina lized w it hout t he required final reference documents, which covered the full upstream and downstream process. To overcome this issue, two versions of the control strategy were produced. The first, including seed train, production, and clarification unit operations, was finalized by all parties as all upstream studies had met the necessary control strategy deadline. In turn, the protocol, sam- ple plans, and batch production records could prog- ress. Later, when downstream processing was com- pleted, a second version could be published to capture the full manufacturing process. There were, however, some disadvantages to this approach. Two f ur ther revisions of the upstream content were required before downstream process information was added; and maintaining separate document portions added complexity. In conclusion, although fragmenting the control strategy might be appropriate in the future when time constraints are present, it should not necessarily be considered as a default approach. Another efficiency involved routing downstream batch records in parallel to PPQ cell line expansion and production. The PPQ batch records had been piloted in a batch performed at scale prior to PPQ. Therefore, starting the PPQ without the full set of approved batch records was, in part, risk-mitigated by that pilot batch. The pilot was performed against the FMEA with the presumption that the potential risks were confirmed by process characterization and carrying in the presumptive CPPs. To act conserva- tively, operating ranges had been tightened and re- inforced by procedural or engineering controls (e.g., alarm set points) for presumptive CPPs until proven acceptable ranges were defined. Later, the control of the batch records for PPQ was reduced when process characterization showed which operating ranges had no impact on the CQAs. Future refinement Un d e r s t a n d i n g pr o c e s s e s i n t h e d e v e l o pm e n t a nd m a nu f a c t u re of biolog ic a l d r ug s ubs t a nce is cr ucia l to successf u l ly nav igat i ng t he cl i n ica l phases toward commercia l launch, a ll within ev- er-t ighten ing t ime const ra ints a nd st r ict reg ula- tory frameworks. This timeline reduction was seen in the response to the COVID-19 pandemic, where deadlines were challenged and developers sought t i me-sav i ng mea su res where possible. However, lessons had been learned, in that if a process is not completely locked or optimization is not fully com- pleted, there is a risk that rework will be required, which can have a detrimental effect on a program. For i n st a nce, dead l i nes ca n be overcome by per- for m i ng met hod opt i m i zat ion a nd va l idat ion i n parallel to process characterization; however, that may necessitate analysis on two separate methods in parallel, duplicating work and increasing costs. L e ve r a g i n g s c a le -i nde p e nde nce a nd k e e pi n g up w it h i ndu s t r y g u ide l i nes a re k e y. Usi ng t he ever-g row i ng wea lt h a nd breadt h of ex per ience i n per for m i ng t h is process a l lows ref i nement of t he FMEA met hodolog y, and using impact factor ranges can narrow the scope of studies and reduce the total amount of work needed, gaining overall ef ficiency. Reference 1. F DA . P ro cess Validat ion : G e ne ral P r inc iples and P ractices. Guidance for Industr y, Januar y 2011. ■