Recently, evidence-based medicine (EBM) has been introduced into medical practice and developed to assist practitioner and patient decisions for appropriate medical care in specific clinical circumstances, including cancer chemotherapy. At present, cancer chemotherapy is still considered an incomplete anticancer therapy, because it rarely results in cure of advanced cancers. Most cancer chemotherapy is therefore considered palliative. Moreover, cancer chemotherapy is toxic due to the side effects of anticancer drugs. Given this situation, evidence-based cancer chemotherapy may contribute to the clinical practice of medical oncology. EBM in cancer chemotherapy consists of four steps, just as in EBM in general practice. The first step is the formulation of clinical problems, the second step is to survey the literature, the third step a critical review of the literature, and the fourth step application to patient. If valid, reliable, reproducible, clinically flexible evidence for cancer chemotherapy are found in this process, it can be applied to a patient with cancer in clinical practice. It is important to realize, however, that EBM cannot always account for individual variation among patients. Furthermore, there is another big problem in obtaining evidence, since there are very few reports of randomized comparative studies of cancer chemotherapy which were carried out in Japan with Japanese patients. Most evidence is therefore derived from patients from other countries. This means that the evidence obtained in cancer chemotherapy should be applied to our patients with due caution. Given doses of drugs or administration schedules may not be suitable to Japanese people. According, we should undertake large scale clinical studies for the various evidences of cancer chemotherapy in our country.