Chemistry Reference
In-Depth Information
Table 2
''Fine tuning'' test formulations used to determine the relative effects on
T
0
g
values of NaCl and sucrose on buffered WIN solutions
Formulation
A
Component
B
C
D
WIN (mg)
20
20
20
20
Lactic acid (0.1 M, in ml)
0.5
0.5
0.5
0.5
NaOH (0.5 M, in ml)
12
12
12
12
Sucrose (mg)
100
50
40
30
NaCl (mg)
2.7
2.7
3.8
5.0
Water for injection (ml)
1
1
1
1
pH
4.0
4.0
4.0
4.0
T
0
g
(1C)
37
38
42.5
45.5
in a mechanically stronger cake structure. A preference for Formulation
B was decided upon, and test samples were freeze-dried (primary drying
temperature:
401C). The mean residual water content of optimally dried
vials was 1.4% (Karl Fischer coulometric analysis), with T
g
ΒΌ
421C,
giving a maximum ''safe'' storage temperature of ca. 371C. Dried samples
were subjected to extended stability trials at 30 and 401C. The product
was shown to be chemically stable; no changes in assay or pH were
observed after 6-month storage.
This discussion demonstrates how a logical refinement of acceptable
chemical formulation parameters against measured thermomechanical
properties of the dried preparations will aid in the development of stable
products. Thus, the often-repeated dictum,
13
even by those who should
know better, that acceptable products can only be obtained by trial-and-
error methods is outdated and should be discarded.
12.5
Inadvertent Failure: ''Unexpected Problems''
On occasions, even long-established processes may give rise to a rogue
failed batch of vials. Failure through collapse of the cake will be easily
observable, but that may not be the case when the failure resides in a
partial or complete loss of biological activity of the product without
observable changes in the ultrastructure of the cake. It then requires a
''forensic'' approach to investigate the causes of failure. This event
happened with one isolated freeze-drying run of Plasma Coagulating
Factor VIII, which had been freeze-dried on a large scale by a well-
established process cycle. Comparative tests on successful and failed
products revealed that failure was due to an unexpectedly low T
0
g
value
of
551C, which, according to specification, should have been
451C.
Search WWH ::
Custom Search