Cellular basis for the inefficacy of 5-FU in human colon carcinoma.

Five established human colorectal carcinoma cell lines with distinct phenotypic properties were exposed to different concentrations of 5-FU for varying time intervals. Effects were measured by sequential cell counts and by inhibition of colony formation. Treatment for 1 hr with 1 microgram/ml barely decreased survival of all cell lines as measured by colony formation; at a concentration of 100 microgram/ml, survival was modestly reduced for all cell lines, and for concentrations of 1000 microgram/ml, survival was decreased by greater than 50%. Extending the length of the treatment interval markedly increased the degree of cell kill for all concentrations of 5-FU. Treatment for greater than 24 hrs resulted in almost complete extermination of colony-forming cells, even for relatively resistant cell lines. The effect of 5-FU treatment on cell number was more complex and depended on drug concentration, length of treatment, and type of cell line. In general, decrements in cell numbers were somewhat related to both drug concentration and length of treatment interval, especially if performed 7 days after terminating drug treatment. Earlier cell counts were inconclusive and the same result could be obtained for different drug concentrations or treatment intervals. Furthermore, these results would change on a daily basis. More important, cell count results never correlated with the survival endpoint measured by inhibition of colony formation. Our results suggest that enhancement of the currently poor performance of 5-FU in the treatment of human colon carcinoma could originate from changing the administration modality to long-term infusion.