A Murine Model to Evaluate the Ability of in Vitro Clonogenic Assays to Predict the Response to Tumors in Vivo

摘要: The use of the human tumor cloning assay as a predictor clinical response tumors to drugs is predicated on hypothesis that in vivo drug can be correlated with vitro cells derived from tumor. To test this hypothesis, we utilized murine model which and responses accurately reproducibly compared. Drug activity was assessed P388 leukemia standard antitumor (i.p. tumor/i.p. administration) an wherein ascites are removed mice, treated drug, directly cloned soft agar measure clonogenic capacity. analogues within four separate classes agents, anthracyclines, anthraquinones, platinum(II) coordination complexes, phosphinogold(I) complexes evaluated. failed discriminate between highly active agents only marginal efficacy ( i.e. , doxorubicin daunorubicin versus rhodomycins A B, ametantrone NSC 276740, cisplatin transplatin, [Au(deppe)2C] [Au(depe)2]PF6. Furthermore, detect carboplatin agent . basis for these discrepancies explored by more detailed comparison rhodomycin B. In or exposure subsequent measurement cell kill tumorigenic demonstrated overestimated cytotoxic potency relative assay. Treatment at doses below maximally tolerated dose mice resulted multiple log measured whereas B levels exceeding its dose. results indicate subset stem capable forming colonies significantly sensitive effects anthracyclines than cells. Cytotoxic not accurate even exposed i.p. drug. cytotoxicity useful nonselective prescreen must used combination other indicators selectivity dose-limiting organ toxicity.