Anticancer Agents for Treatment of Tumors in the Central Nervous System by Correspondent Substituent Substitution and Elucidation by Pattern Recognition Methods

Abstract

Within the United States, primary brain tumors account for 20 to 25 percent of all pediatric cancers. Chemotherapy utilizing a nitrosourea, notably semustine (MeCCNU) and carmustine (BCNU), has shown significant success in the treatment of tumors found in the central nervous system. In silico optimization of molecular properties by substituent substitution that is followed by pattern recognition analysis is utilized in this study to develop 14 novel anti-cancer drugs for the treatment of malignant cancers of the central nervous system. These 14 agents exhibit molecular properties that are suitable for penetration through the blood-brain barrier (BBB). All 14 agents are nitrosoureas having values of Log P ranging from 2.188 to 2.942, and having a constant total of 5 oxygens and nitrogens with zero violations of the Rule of 5 which indicates favorable bioavailability. Value of Log BB (Log [Cbrain/Cblood]) for these agents does not vary from - 0.441 (BB value of 0.362). The formula weight of the agents is highly correlated to molecular volume (r= 0.9848) and total number of atoms (r= 0.9948), but not correlated to number of rotatable bonds (r= 0.1814). Analysis of similarity (ANOSIM) indicated that all 14 new constructs are similar to the parent compound semustine. The Log P value for all 14 agents predicts favorable attributes for penetrating the BBB. Multiple regression analysis established that number of atoms, number of rotatable bonds, and molecular volume are strong prognosticators for molecular weight of this assemblage of pharmaceuticals. This study attests to the efficacy of in silico optimization of molecular substituents followed by pattern recognition analysis to develop new drug designs based on a successful nitrosourea framework for the treatment of malignant tumors of the brain.