4.1. Considerations When Scaling up CDM Manufacturing
The scaling of CDM production from laboratory to manufacturing has multiple considerations, as seen in Figure 3: batch variability, manufacturing cost, and quality control.
Considerations when scaling up CDM manufacturing. Created with BioRender.com
on 1 October 2021.
Batch variability refers to the consistency of the CDM composition. Any change to the intrinsic and extrinsic factors during manufacturing will result in the deviation of the CDM product. The first consideration in scaling up production is to fix and secure all the various factors: cell sources, medium components, bioreactors, chemicals, and processing methods.
Cell sources such as primary cells and cell lines play a role in the batch variability of CDM. Primary cells are variable by nature; thus, appropriate pre-characterisation and cell-banking strategies must be implemented to maintain the required standard. Without genetic modification, only stem cells derived from primary sources have the capability to undergo sufficient cell division for CDM manufacturing at scale. For example, embryonic stem cells are often considered immortal in culture 
, mesenchymal stem cells can have between 13 and 25 population doublings 
, and epidermal stem cells can have up to 115 population doublings 
before undergoing senescence. For industrial manufacturing, maintaining a stem-cell source is a tremendous task, considering the logistics of obtaining donor cells, isolation, ‘stemness’ maintenance, banking and quality assurance.
Cell line, or genetically engineered cells, is a much better option for industrial CDM production due to their capability of being maintained for a very long period. However, cell lines may still lose their special characteristics and need regular quality control and characterisation 
. Numerous mammalian cell lines have been developed for the industrial production of biotherapeutic proteins, notably the Chinese hamster ovary (CHO) expression systems 
. However, more development is needed to express larger ECM proteins using these systems.
Cell culture conditions (such as temperature, oxygen, additives, and substrate scaffolds) directly affect cell growth and ECM deposition and should be kept under strict control 
. Finally, the decellularisation/extraction process should also be standardised or automated, as any alteration in methods, chemicals, or perfusion rate will result in changes to the CDM product 
Manufacturing cost includes the costs of research/development and of production. The cost of developing chemically defined media needs to be considered. Optimising culture media is a costly and time-consuming endeavour due to the complexity of potential formulations. However, a well-defined medium may be invaluable when it comes to reducing the usage of expensive growth factors and serum 
. Another factor is the choice of scale-up cell culture platforms, the costs of which depend on the medium consumption over CDM yield and whether the technology requires extra downstream processing steps (such as the removal of microcarriers). Finally, the potential for automation should also be considered. Automation will reduce manual handling and hence labour cost. Plus, automation will mitigate human errors, resulting in more consistent quality and improved manufacturing performance 
Quality control ensures the final product consistently meets the designed specifications, which is the hallmark of any industrial process. This can be achieved by identifying and monitoring critical quality attributes (CQAs) and critical process parameters (CPPs). CQAs is defined as “a physical, chemical, biological characteristic that should be within an appropriate limit, range, or distribution to ensure the desired product quality”. CPPs refer to the parameters in the production process that impact the CQAs, which in turn affect the quality of the final product. Hence, by establishing the CQAs and CPPs, manufacturers would ensure consistency in the final products 
4.2. Challenges of Existing Scale-up Technologies for CDM Production
Existing scale-up technologies address the issue of nutrition transfer with a high surface area-to-volume ratio. However, they do not address other factors, such as the effect of dynamic culture and additional steps for downstream processing.
The dynamic nature of the platforms (e.g., perfusion rate for the hollow fibre bioreactors, sparging rate for the microcarrier bioreactor vessels, revolutions per minute for the roller bottles, rocking cycles per minute for the rocker bioreactor) will influence the composition of the final CDM products. Since the effect of dynamic culture on CDM production is not well-established, it is important to carefully characterise the obtained CDM with each new scale-up technology prior to its adoption for large-scale production.
Originally, most scale-up platforms were developed for the extraction of cells or biologics (such as soluble proteins), not the whole ECM with abundant insoluble macromolecules. As such, CDM manufacturers would need to develop extra downstream processes in order to harvest CDM from the existing systems (e.g., removal of CDM from the hollow fibre channels, microcarriers, or vessel surface). This extra processing must be carefully designed to avoid altering the desired CDM composition or contaminating the products with unwanted residual chemicals.