Robert Shaw, Commercial director of the Stem Cell Initiative at Millipore discussed the state of manufacturing processes in the regenerative medicine industry today. In spite of the rapid progress in clinical trials, he began, the industry is moving slowly in terms of developments in manufacturing. He attributed this in part to a lack of standardization and tools for large scale processing.
The lack of standardization is partly a product of the diversity of cell types as well as batch sizes, making the manufacturing needs in this sphere highly variable. There is no such a thing as a generic stem cell therapy and therefore to what extent can there be a widely applicable manufacturing strategy and generic tools to accompany it? Millipore, Lonza and others in the space are favouring the use of fed-batch bioreactors for the growth and expansion of cells seeded onto microcarriers. The âMobius 3L bioreactor' made by Millipore for example has been used by the company to generate reproducible expansion of mesenchymal stem cells (MSCs) in 10-12 day protocols that generate up to 700 million MSCs per batch. This is contrasted to the 200-400 million cells generated in ten stack adherent culture plates within the same time frame. The comparability of cell growth in bioreactors is so far being assessed by gene arrays and differentiation assays, comparing these parameters in flat adherent and suspended microcarrier cultures. Although some differences are detected, particularly in the expression of proteins involved in cell attachment, so far there are judged to be no important difference that might impact on growth and/or differentiation. Shear stress does not appear to be a problem in stirred tank bioreactors effecting neither stress nor apoptotic signalling pathways, a finding echoed by David Smith at Lonza. The metabolic requirements of MSCs are small compared to those of traditional mammalian culture and thus aeration is not a particular problem either. MSCs have a tendency to from aggregates on microcarriers which cross link with neighbouring aggregates. These processes are currently being monitored using live cell imaging in the cell IQ system.
Overall, the use of bioreactors are projected to decrease processing costs at the expansion phase to a third of those seen in flat, adherent cell culture. However, the next big stage of development will be cell harvesting strategies; David Smith at Lonza later discussed how harvesting of cells from microcarriers was a major source of cell loss and a bottle neck in the manufacturing process.
Genomic integrity was highlighted as a key concern during long term expansion; karyotypic rearrangements occur at varying rates between cell types and cell lines. This parameter will therefore limit the expandability of cell lines. Millipore are comparing the relative genomic integrity over 30 population doublings in both suspension and adherent cultures using techniques with 15KB sensitivity.
Professsor David Williams from the University of Loughborough followed with a clear message about product characterisation and comparability; despite the old cell therapy paradigm âthe product is the process' it was argued that changes to the manufacturing process are inevitable during early development. Thus detailed product characterisation assays are essential from an early stage in development. âWithout comparability' Professor Williams stated, âthere can be no process changes, no multiple sites, no scalability, no roll out following commercial success and, accordingly, no effective manufacturing strategy'. The way that we define and characterise our product therefore is purported to be THE manufacturing issue of the moment. In addition he preached the importance of engaging clinicians early in order to generate the âpull' that will lead to product uptake and eventual reimbursement.
Professor Williams said that incremental improvements are being made and a ânew language' is seeping into the industry rhetoric. Companies are considering the whole system in manufacturing instead of cherry picking the difficult areas for trouble-shooting. In addition to this, unnecessary processing steps are being eliminated and there is a move towards âclosed' processing and an increased use of disposables. A key unanswered question is the degree of precision required by regulators, both in terms of quality control assays and overall product reproducibility. Professor Williams emphasised the power of companies and researchers to influence the interpretation and evolution of regulatory guidelines in this new sphere, but warned that only by presenting a united front can real change be affected.
Guest blog provided by Amelia Lane, UCL