Folinic acid (dlFA) is increasingly used in clinical oncology. The active isomer lFA is intensively metabolized into l5-methyltetrahydrofolate (l5MTHF), the relative proportions of lFA, dFA and l5MTHF in blood varying considerably between oral and i.v. FA administration. The purpose of the study was to compare the in vitro activities of pure lFA and pure l5MTHF at equivalent drug exposure [area under curve (AUC)], taking into account their respective chemical stability in the culture medium. The in vitro growth inhibition [3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) test] was evaluated on five human tumor cell lines after methotrexate (MTX)-folate or 5-fluorouracil (5FU)-folate exposures. Not only were the activities of lFA and l5MTHF compared, but also clinically relevant mixtures of lFA + dFA + l5MTHF corresponding to the proportions found at steady state during oral (PO mixture, 4, 39 and 57%, respectively) and i.v. administrations (i.v. mixture, 7, 81 and 12%, respectively). Measurement of folates demonstrated the marked lability of l5MTHF (65.8% loss over 5 days in the culture medium) as compared to lFA (2.6% loss). Whatever the pharmacological model tested (MTX-folate or 5FU-folate), comparison of the folate effects at equivalent drug exposure taking into account their relative stability showed that l5MTHF was never more potent than lFA. Moreover, a higher efficiency of lFA was demonstrated for the cell line most sensitive to 5FU; in this case, as expected, the i.v. mixture was more potent than the PO mixture. This study shows that depending on the tumor, lFA can be more potent than its main circulating metabolite l5MTHF. Along with the limited capacity of oral absorption, the choice between oral and i.v. route for FA administration in patients should take into consideration the different pharmacological activities between lFA and l5MTHF which suggest that the oral route is potentially detrimental to the optimal activity of the 5FU-FA combination as compared to i.v. administration.