In 2011 the FDA issued a Guidance for Industry on process validation. The guidance formally defined stages during the lifecyle of a commercial process design. These are as follows:
During stages 1 and 2b statistics is frequently implemented in the design of experiments. However, in particular for the development of a biological commercial process for the manufacturing of API/DS, further implementation of statistical tools is restricted due to the availability of material, sample sizes as well as, scale and facility considerations. At the start of stage 3, the FDA agency recommends a “continued monitoring and sampling of process parameters and quality attributes at the level established during the process qualification stage until sufficient data are available to generate significant variability estimates” Proper application of statistics requires unimodality of distribution curves, as well as, randomness and independence of data. Care must be taken to ensure that during the application of process monitoring tools when transitioning between Stage 2 to Stage 3 the above prerequisites are met. It is critical to have proper sample sizes, normality and a stable process (common cause variation only) prior to calculating process capabilities. This can be of particular importance in the development of biologics that rely significantly in outsourced services and use multiple contract manufacturing organizations (CMOs) to meet supply demands.