Impact of filling processes on protein solutions

During production proteins are exposed to various stresses which can cause protein
denaturation and inactivation. The objective of the present study was to investigate
the effect of shear forces which can occur during filling operations of pharmaceutical
solutions with dosing equipment. Such shear forces possibly have a negative
influence on shear sensitive substances and may lower the quality and yield of the
final drug product.
In the scope of this work a peristaltic pump and different sizes of rotary piston pumps
(RPPs) were compared in respect to induced protein aggregation due to shear
damage caused by dosing equipment. The influence of various parameters such as
filling speed, dosing volume, friction surface and exposure to air-liquid interfaces and
on the intensity of the shear stress was examined. A characteristic rotary piston
pump parameter δ was developed and introduced as an indicator describing the
potential of a rotary piston pump to cause protein damage. Furthermore, excipients
were tested on their ability to protect the model protein against shear-induced
damage. 2 model proteins in solution, lactase (β-galactosidase) and rituximab, a
recombinant chimeric monoclonal antibody, were used and tested for their suitability
as model proteins. No activity loss was seen for the sheared lactase solution,
therefore finally rituximab was chosen as a model protein.
The level of protein aggregation in the unsheared and sheared solutions was
determined by Photon Correlation Spectroscopy (PCS) and SEC-HPLC. TEM was
used to visualise protein aggregation.
It was found that protein aggregation was induced by rotary piston pumps however
not by the peristaltic pump. The degree of protein damage was marginally low for
large rotary piston pumps such as RPP 3 and 4 and showed a considerable increase
with smaller sizes like RPP 1 and 2. A loss of protein monomers of 3.2 % ± 1.8% was
found after 3 hours of circulation with RPP 1 in the test system. For RPP 2 a loss of
0.4% ± 0.2% was found. No loss was seen for RPP 3. The different clearance
between the piston and the cylinder in the different sizes of pumps was suggested to be one reason for the large difference in exerted shear stress leading to protein
aggregation. Two more factors were suggested to have an influence on the exerted
stress caused by a RPP, which are the generated friction surface per dosed ml and
the dosage volume. Although an influence of the dosage volume could not be
confirmed by the conducted shear experiments. These 3 factors were respected for
the calculation of parameter δ. The evaluation of the filling speed showed minor
influence with a trend to fast filling speeds being more favourable, whereas the
exposure to the air-liquid interface did not show an influence. A slight trend was seen
that the combination of 5% trehalose dihydrate and 0.5% polyethylene glycol showed
the best protective effect out of the excipients examined.
An evaluation of the analytical methods used in this work revealed that PCS is an
extremely sensitive method to detect protein aggregates and was therefore very
suitable to monitor the changes in the protein solutions after circulation in the test
system. A significant lower sensitivity was observed for SEC-HPLC.
It can be concluded that for filling of shear sensitive pharmaceutical protein solutions,
it is critical to choose the appropriate equipment. Large sizes of RPPs such as RPP 3
and 4 or peristaltic pumps should be employed as dosing equipment. Furthermore
high speed gives better results than low speed, i.e. machine stops during production
should be avoided.