The tumor cell growth inhibitory activities (log 1/GI(50)) of 166 anticancer agents studied at the National Cancer Institute (NCI) in vitro anticancer screening program have allowed us to analyze the relative importance of physicochemical parameters in influencing the inhibitory activities. Increased molecular weight, as measured by the logarithm of molecular weight (log MW), is found to be an important contributor to the tumor cell growth inhibitory activities. The tumor cell growth inhibitory activities in different subpanels of the tumor cells are highly inter-correlated with each other. A simple binary quantitative structure-activity relationship (QSAR) model was derived from the 166 anticancer drugs, based on the tumor cell growth inhibitory activities transformed into a binary (active or inactive) data format. The model obtained can be tested with additional new data, and may be useful to identify active compounds from a large compound library to be included in high throughput screening.