Quantitative structure-antitumor activity relationships of camptothecin analogues: Cluster analysis and genetic algorithm-based studies

Topoisomerase 1 (top1) inhibitors are proving useful against a range of ref ractory tumors, and there is considerable interest in the development of ad ditional top1 agents. Despite crystallographic studies, the binding site an d ligand properties that lead to activity are poorly understood. Here we re port a unique approach to quantitative structure-activity relationship (QSA R) analysis based on the National Cancer Institute's (NCI) drug databases. In 1990, the NCI established a drug discovery program in which compounds ar e tested for their ability to inhibit the growth of 60 different human canc er cell lines in culture. More than 70 000 compounds have been screened, an d patterns of activity against the 60 cell lines have been found to encode rich information on mechanisms of drug action and drug resistance. Here, we use hierarchical clustering to define antitumor activity patterns in a dat a set of 167 tested camptothecins (CPTs) in the NCI drug database. The aver age pairwise Pearson correlation coefficient between activity patterns for the CPT set was 0.70. Coherence between chemical structures and their activ ity patterns was observed. QSAR studies were carried out using the mean 50% growth inhibitory concentrations (GI(50)) for 60 cell lines as the depende nt variables. Different statistical methods, including stepwise linear regr ession, principal component regression (PCR), partial least-squares regress ion (PLS), and fully cross-validated genetic function approximation (GFA) w ere applied to construct quantitative structure-antitumor relationship mode ls. For our data set, the GFA method performed better in terms of correlati on coefficients and cross-validation analysis. A number of molecular descri ptors were identified as being correlated with antitumor activity. Included were partial atomic charges and three interatomic distances that define th e relative spatial dispositions of three significant atoms (the hydroxyl hy drogen of the E-ring, the lactone carbonyl oxygen of the E-ring, and the ca rbonyl oxygen of the D-ring). The cross-validated r(2) for the final GFA mo del was 0.783, indicating a predictive QSAR model.