001). The FIGO 1988 staging classification was adopted for this statistical analysis of cervical, endometrial and ovarian cancers. In regard to the clinical staging of cervical cancer, the diagnosis of stage I cervical cancer is influenced by the type of specimen examined, that is, cervical biopsy, cervical conization or total hysterectomy specimens, and it is expected that there may be differences in the interpretation among institutions as well. In addition, stage IVb is also interpreted differently among institutions, and it is possible that some patients may have been diagnosed as having stage IVb due to the presence of distant Natural Product Library solubility dmso metastases or para-aortic lymphadenopathy

on CT and other imaging diagnosis. In the analysis of endometrial and ovarian cancers, surgical staging classification was adopted and the diagnosis without surgery was performed only in a small number of cases comprising 4.5% and 2.1% of patients with endometrial and ovarian cancer, respectively. This suggested that summarized distribution of the surgical stages was still reliable. In regard to the histological types, there is a problem not in cervical, endometrial cancers or ovarian surface epithelial-stromal tumors, but in ovarian Angiogenesis inhibitor sex cord-stromal and germ cell tumors: there are a small number of patients with these

ovarian tumors and only an insufficient number of cases can be accumulated in a year. Therefore, the influence even from a single case can be large, leading to over- or under-estimation. Consequently, it seems impossible to compare and analyze the changes over time. Prognosis was analyzed by the Kaplan–Meier method. Terminal-stage patients are often transferred to other medical institutions in Japan, and in such cases, information on the patients cannot often be obtained after hospital transfer, which leads to unknown prognosis. Fatal cases are considered to account Tolmetin for most of these prognosis-unknown cases. Therefore, if all these prognosis-unknown cases are counted as alive dropouts, the prognosis may be better estimated even by the Kaplan–Meier method. Accordingly,

in the present study, information from institutions in which the prognosis was untraceable for 20% or more of the cases was excluded from the analysis. Among the patients with known prognosis, 58.7% of patients with cervical cancer, 65.9% of patients with endometrial cancer, and 60.0% of patients with ovarian cancer were included in the analysis of the prognosis. However, in this method of analysis, it tends to be more difficult to collect information on patients from larger medical institutions, and future investigations are considered necessary to allow more accurate information on the prognosis to be reflected in the Treatment Annual Reports. The Patient Annual Report and Treatment Annual Report on gynecologic tumors (cervical, endometrial, and ovarian cancers and ovarian tumors of borderline malignancy) in Japan are presented in this paper.