The employment of biochemical concepts to select drugs for use in the treatment of colorectal cancer is advocated. It is stressed that new drug mixtures, even those employing agents inactive alone, with unique mechanisms of action may be fashioned through biochemical design. Following delineation of activity in animal model systems, these drug combinations then become candidates for use in man. It is emphasized that 6-thioguanine is a particularly attractive agent for consideration as a component of new combinations of drugs to be used in patients with cancer. The desirability of this agent derives from (a) the known, albeit weak, activity of 6-thioguanine against colorectal neoplasms of man, thereby providing some inherent activity to be joined by additional materials; (b) a knowledge of several major biochemical and pharmacologic determinants of tissue susceptibility to its cytotoxic action, possibly allowing the ultimate selection of patients with a high probability of response; and (c) the availability of four different agents or classes of agents which synergistically interact with 6-thioguanine to inhibit the growth of malignant cells by diverse biochemical mechanisms.