Over a decade ago, use of tyrosine kinase inhibitors (TKIs) for the treatment of malignancies was found to cause left ventricular dysfunction, a finding that was unexpected and not well predicted by standard preclinical studies. Subsequently, several preclinical approaches were proposed to address this issue. Over the last 5 years, several approaches for preclinical evaluation of cardiac function using isolated perfused hearts, engineered heart tissue and human-induced pluripotent stem cell-derived cardiac myocytes have been shown to be relatively predictive of the cardiotoxic potential of TKIs. Further, preclinical studies submitted for regulatory review for recently approved KIs have demonstrated various forms of KI-induced cardiotoxicity. Thus, early identification and assessment of cardiotoxicity in the preclinical setting is now possible. Given that kinases are involved in diverse cellular processes common to both normal and tumor cells, KI-induced toxicity, particularly in the heart, appears difficult to avoid. To develop drugs with fewer adverse effects, better efficacy and safety assessments, such as pharmacological separation of targets for cancer from heart, and/or wider separation of the drug concentrations for antitumor activity from cardiac toxicity, may be helpful. Additional preclinical approaches for assessing drug efficacy and toxicity in parallel may include use of animal cancer models and a 3D integrated in vitro model of perfused tumor and heart tissues. Minimizing and predicting potential KI-induced cardiotoxicity is still an important regulatory challenge, and better preclinical approaches may help achieve this goal.