Numerous studies have documented that various naturally derived ligands or synthetic non-thiazolidinediones (TZD) as peroxisome proliferator-activated receptor gamma (PPARgamma) agonists have shown moderate or potent antitumor activities, which is PPARgamma independent or partially dependent. However, the PPARgamma agonistic or glucose-lowering activity is ranked first more often than antitumor activity to determine promising novel PPARgamma agonists for potential clinical use. In this study, we hypothesized that there might exist some compounds with less PPARgamma agonistic activity but potent antitumor activity. Thereafter, we evaluated the PPARgamma agonistic and antitumor activity of a novel series of alpha-aryloxy-alpha-methylhydrocinnamic acid derivatives synthesized with the initial aim of developing novel PPARgamma agonists as hypoglycemic agents. MTT assay results revealed that several compounds were able to inhibit cell proliferation in a dose-dependent manner with IC(50) 12.7-29.7 microM, better than that of rosiglitazone (45.9-141 microM), although the PPARgamma agonistic activity of most compounds is much lower than rosiglitazone. Some compounds induced cell cycle arrest and apoptosis tested by Flow Cytometry. Oral administration of DH9 (100 mg/kg/d) for 21 days to BALB/c nude mice bearing xenografts including MGC-803, NCI-H460, HT-29 and OS-RC-2 cells significantly retarded tumor growth. DG8 and DJ5 showed benefits in some of the above four xenografts. Our findings demonstrate that these compounds have potent antitumor activity in vitro and in vivo and pyrimidinyl-arylpropionic acid derivatives might be viable resources in the development of new antineoplastic agents.