We investigated the use of ornithine alpha-ketoglutarate in treatment of rats bearing Morris hepatoma 7777. Rats received diets containing either ornithine alpha-ketoglutarate, which has been used in other catabolic states (i.e. injury, sepsis), or an isonitrogenous, isocaloric diet containing glycine. Untreated tumors grew to a mass of 11 g/100 g body weight over the 3-wk period after implantation and induced progressive anorexia, negative nitrogen balance, and body and tissue wasting. Compared with glycine, ornithine alpha-ketoglutarate had no effect on tumor growth, but also did not alter the catabolic effects of the tumor on its host. We hypothesized that capture of amino acids by the tumor limited the efficacy of supplemental nutrition here and in published reports in which tumor burden comprised 4-30% of body weight. This is supported by our observation that a 3-wk of implantation the rate of protein deposition plus amino acid oxidation by the tumor was equivalent to approximately 70% of the host's daily protein intake. To parallel the clinical situation in which tumor burden is small at diagnosis and initiation of treatment, the same diets were tested in rats treated by excision of the tumor at a limited stage of the disease. Rats received 3 d preoperative nutrition with ornithine alpha-ketoglutarate or glycine, and continued on the same diets for 3 or 6 d postoperatively. Compared with glycine-fed rats, ornithine alpha-ketoglutarate-fed rats showed a more positive nitrogen balance, higher concentrations of glutamine and branched-chain amino acids in muscle, and accelerated protein deposition in small intestine (P < 0.05). Our results explain the lack of success of nutritional support in untreated cancer and underline the need for clinically relevant animal models for further studies.