Background:　One　of　the　main　reasons　for　most　of　the　anticancer　drugs　to　fail　in　the　late　preclinical　testing　and　early　clinical　trials　is　the　differences　in　drug　effects　observed　from　animals　and　patients,　and　the　challenge　has　been　to　find　a　balance　to　reduce　the　inherent　differences　from　species.　Objective:　Predicting　safe　starting　doses　and　dosing　schedules　for　human　clinical　trials　is　the　main　purpose　of　toxicological　studies　of　anticancer　drugs.　Methods:　Relevant　information　and　data　were　assimilated　from　manuscripts,　congress　publications,　and　online　sources.　Results:　We　systematically　overview　the　cons　and　pros　of　animal　models　and　briefed　the　ways　to　determine　human　clinical　starting　doses　derived　from　animal　toxicological　studies　for　anticancer　drugs.　Conclusion:　This　information　helps　smart　select　the　suitable　predictive　model　for　anti-cancer　drugs　with　the　different　mechanisms　and　emphasized　the　pharmaceutical　challenges　behind　and　ahead.