Product Optimisation Report

At the completion of product optimisation, when the best product variant has been selected, it is a good idea to summarise the work conducted in a Product Optimisation Report. The report should reference the primary data from preformulation, product optimisation and stability studies, cross-referencing other investigational reports where necessary. It should clearly justify the recommendations for the quantitative formula and the excipient, component and product specifications. Such a document can be very useful to aid smooth technology transfer into production and for writing regulatory submissions. 

Experiments for optimisation of die production of L-phenylalanine are usually carried out at pH = 75, a temperature of 37 C, 50% DOT and 10 am dry weight biomass p>er litre medium. Maximum productivity is reported to vary between 3 and 6 g T h and product concentrations of between 11-28 g T have been reported. The time necessary for completion of the reaction is about 8 hours (see Figure 8.7). 

The major objective of the product optimisation stage is to ensure that the product selected for further development (the intended commercial product) is fully optimised and complies with the design specification and critical quality parameters described in the Product Design Report (refer to Chapter 5). The key outputs from this stage of development will be... 

A logical approach to packaging optimisation is, first of all, to define the packaging function, followed by selection of the materials, then testing the performance of the packaging to ensure that it will meet all the product design and functional requirements that were identified in the Product Design Report. 

On completion of the work programme, a Process Optimisation Report should be written. This will summarise the results of the activities specified in the protocol and provide a rationale to define the operating limits for the process and the critical parameters affecting product quality or performance. The report should also conclude that the specifications for the raw active, excipients, components, in-process and product can be met. 

The Development Report should be concise and structured. Clearly, it cannot be finalised until development is complete, but the preparation is much easier if summary reports have been compiled during development, such as the product and process optimisation reports. The Development Report needs to be available to the FDA prior to the inspection, ideally, to give the FDA inspection team confidence that the product has been developed satisfactorily, perhaps resulting in a shorter inspection. 

It has been reported that insect cells have a strong internal buffering capacity [85] however, there are indications that medium pH must be optimised for growth and production phases and should be kept under tight control in bioreactors, especially in high-density large scale cultivations [56]. 

Nevertheless, DO levels seem to be more important for product expression than for cell growth and the effect seems to be product specific. Cruz et al. [69] studying the influence of DO levels in HIV-VLPs production have analysed product titer and quality at DO levels of 10,25 and 50%, with quality defined as the percentage of high molecular weight particles in the final product they concluded that the best quality was obtained at a DO of 10%, but the best titer was obtained at 25% DO level. Conversely, Hu and Bentley studying IBVD [33] have obtained the best titer at 50 to 80% DO level, and a lower yield at 25% the same group has reported for the expression of epoxide hydrolase that the best DO level was 25% [89]. Thus, the operational DO conditions should be optimised for each product. 

However, as important as the Hu and Bentley Model is the stepwise approach to process optimisation that Hu and Bentley have reported [33]. The focus on quantitative analysis of protease degradation of the product over time, along with the similar approach followed by Cruz et al. [25], also indicate new directions to follow in mathematical modelling regarding product expression optimisation. 

Tober A J, Sanders, T G - Glycol System and Production Cooler Optimisation Study for the North Rankin A Platform, Worley Engineering Report No 25, December 1979. 

More recently, Priego-Capote et al. reported on the production of MIP nanoparticles with monoclonal behaviour by miniemulsion polymerisation [63]. In the synthetic method that they employed, they devised to use a polymerisable surfactant that was also able to act as a functional monomer by interacting with the template (Fig. 4). The crosslinker content was optimised at 81% mol/mol (higher or lower contents leading to unstable emulsions). In this way, the authors were able not only to produce rather small particles (80-120 nm in the dry state) but also to locate the imprinted sites on the outer particle surface. The resulting MIP nanobeads were very effective as pseudostationary phases in the analysis of (/ ,S)-propranolol by CEC. 

Conducting polymers have already been well documented in conjunction with the classical ionophore-based solvent polymeric ion-selective membrane as an ion-to-electron transducer. This approach has been applied to both macro- and microelectrodes. However, with careful control of the optimisation process (i.e. ionic/electronic transport properties of the polymer), the doping of the polymer matrix with anion-recognition sites will ultimately allow selective anion recognition and ion-to-electron transduction to occur within the same molecule. This is obviously ideal and would allow for the production of durable microsensors, as conducting polymer-based electrodes, and due to the nature of their manufacture these are suited to miniaturisation. There are various examples of anion-selective sensors formed using this technique reported in the literature, some of which are listed below. 

Dynamic optimisation of this type of periodic operation was first attempted and reported in the literature by Mayur et al. (1970), who considered the initial charge to the reboiler as a fresh feed stock mixed with the recycled off-cut material from the previous distillation task. Each batch cycle is then operated in two distillation tasks. During the Task 1, a quantity of overhead distillate meeting the light product specification is collected. The residue is further distilled off in Task 2 until it meets the bottom product specification. The overhead during Task 2 meets neither specifications (but the composition is usually kept close to the that of the initial charge for thermodynamic reasons) and is recycled as part of the charge for the next batch. As the batch cycle is repeated a quasi-steady state mode of operation is attained which is characterised by the identical amount and composition of the recycle (from the previous batch) and the off-cut (from the current batch). Luyben (1988) indicates that the quasi-steady state mode is achieved after three or four such cycles. 

Seiders et al. [92] have prepared enantiopure ruthenium-based catalysts 41a and 41b for use in asymmetric RCM reactions. Using catalyst 41b, the desym-metrisation of achiral triene 46 was not as selective as was reported by Hoveyda and Schrock [85] using 40a. Nonetheless, under optimised conditions (involving the formation of an iodo derivative of 41b in situ) a product (47) enantiomeric excess of 90% was possible (Scheme 15), thus showing that there is some potential for analogues of the more robust 4 to serve as efficient chiral catalysts. 

---