Use of the agar diffusion chamber for the exposure of human tumor cells to drugs.

Human melanoma xenografts in immune-deprived mice have been used to assess the value of the agar diffusion chamber for chemosensitivity testing. Tumor cells were treated with melphalan, Adriamycin, or methyl trans-1-(2-chloroethyl)-3-(4-methylcyclohexyl)-1-nitrosourea either as solid tumors growing in mice or as suspensions in agar in i.p. diffusion chambers. Survival of clonogenic human tumor cells was measured by the agar diffusion chamber assay in both cases. Cell survival curves were log-linear for treatment of tumor cells in vivo or in the chambers. For melphalan the slopes of survival curves were significantly greater for treatment in the chambers than as solid tumors in vivo, but for methyl trans-1-(2-chloroethyl(-3-(4-methylcyclohexyl)-1-nitrosourea or Adriamycin, they were indistinguishable. Experiments with [14C]melphalan showed that the levels of drug achieved were less inside the diffusion chambers than in the tumors in vivo so that the sensitivity of tumor cells to melphalan was much greater when they were treated in chambers. The differences in drug exposure and in cellular chemosensitivity between chambers and tumors suggest caution in the interpretation of drug testing using this system, but the log-linear nature of the dose-response curves is an important feature which may be useful in the eventual development of optimal chemosensitivity testing systems.