Wj. Welsh et al., PHARMACEUTICAL FINGERPRINTING - EVALUATION OF NEURAL NETWORKS AND CHEMOMETRIC TECHNIQUES FOR DISTINGUISHING AMONG SAME PRODUCT MANUFACTURERS, Analytical chemistry, 68(19), 1996, pp. 3473-3482
The present study was undertaken to evaluate several computer-based cl
assifiers as potential tools for pharmaceutical fingerprinting by util
izing normalized data obtained from HPLC trace organic impurity patter
ns, To assess the utility of this approach, samples of L-tryptophan (L
T) drug substance were analyzed from commercial production lots of six
different manufacturers, The performance of several artificial neural
network (ANN) architectures was compared with that of two standard ch
emometric methods, K-nearest neighbors (KNN) and soft independent mode
ling of class analogy (SIMCA), as well as with a panel of human expert
s, The architecture of all three computer-based classifiers was varied
with respect to the number of input variables, The ANNs were also opt
imized with respect to the number of nodes per hidden layer and to the
number of hidden layers, A novel preprocessing scheme known as the Wi
ndow method was devised for converting the output of 899 data entries
extracted from each chromatogram into an appropriate input file for th
e classifiers, Analysis of the test set data revealed that an ANN with
46 inputs (i.e., ANN-46) was superior to all other classifiers evalua
ted, with 93% of the chromatograms correctly classified. Among the cla
ssifiers studied in detail, the order of performance was ANN-46 (93%)
> SIMCA-46 (87%) > KNN-46 (85%) = ANN-899 (85%) > ''human experts'' (8
3%) > SIMCA-899 (78%) greater than or equal to ANN-22 (77%)= KNN-22 (7
7%) greater than or equal to KNN-899 (76%) > SIMCA-22 (73%), These res
ults confirm that ANNs, particularly when used in conjunction with the
Window preprocessing scheme, can provide a fast, accurate, and consis
tent methodology applicable to pharmaceutical fingerprinting. Particul
ar attention was paid to variations in the HPLC patterns of same-manuf
acturer samples due to differences in LT production lots, HPLC columns
, and even run-days to quantify how these factors might hinder correct
classifications. The results from these classification studies indica
te that the chromatograms evidenced variations across LT manufacturers
, across the three HPLC columns and, for one manufacturer, across lots
, The extent of column-to-column variations is particularly noteworthy
in that all three columns had identical specifications with respect t
o their stationary-phase characteristics and two of the three columns
were from the same vendor.