Quantitative structure-activity relationships (QSAR) study has been performed for two sets of the antitumor drugs against human breast cancer MCF-7 cell lines, alpha-tocopherol and cholesterol derivatives. Constitutional, geometrical, physico-chemical and electronic descriptors (using the density functional theory, B3LYP/6-31G (d,p) basis set) were computed and analyzed. The most relevant of these descriptors were grouped and multiple linear regressions have been carried out. Optimal QSAR models with three and four variables, R(2)>0.95 and cross-validation parameter q(pre)(2)>0.88, were selected. Based on the QSAR study, novel vitamin-E derivatives (compounds D-1 and D-2) were designed and their antiproliferative activities were evaluated using the proposed regression models. Calculated antiproliferative activities of the designed compounds, IC(50) (D-1): 3.09 microM and IC(50) (D-2): 3.54 microM, were significantly stronger than anticancer effect of the other analyzed compounds IC(50): 4-1461 microM.