Risk factors for local and distant metastasis and survival characteristics of patients with locally advanced cervical cancer, revealed on the basis of analysis of historical data for the period 1998–2021

Background. Cervical cancer is the second most common female malignant tumor in the world, which seriously threatens women’s health. The prevalence of cervical cancer in developed countries is 9 cases per 100 thousand people and almost twice as much (about 18 cases per 100 thousand female population) in developing countries. The survival rate of patients depends on the stage of the disease, and if detected late, it is critically low – 5–6% over a 5-year period. After clarifying the etiology of the disease as a persistent infection of the human papilloma virus, the processes of creating and implementing a comprehensive system of prevention and combating the disease were significantly accelerated. In 2018, WHO called for the elimination of cervical cancer worldwide, and more than 70 countries and international academic societies took immediate positive action to join this process. In 2020, WHO published a global strategy to combat cervical cancer as a pressing public health problem and put forward a list of the main areas of work for the prevention and control of cervical cancer in the future, which was supported by 194 countries of the world. Radiation therapy in combination with cisplatin-based chemotherapy is the standard of care for locally advanced cervical cancer. A possible way to improve the results of treatment is the search for risk factors for a negative prognosis and the appearance of remote oncological consequences (OC). Purpose. Determine the factors and time characteristics of the risk of local and distant metastasis in patients with locally advanced cervical cancer with radical conservative treatment. Materials and Methods. The basis for the work was the catamnetic data of the medical histories of patients with locally advanced cervical cancer who were treated at the State Organization «Grigoriev Institute for Medical Radiology and Oncology of the National Academy of Medical Sciences of Ukraine» in 1998–2021, the duration of observation of which was more than 1 year or the disease manifest was recorded at the initial stage (up to 1 year) from the start of treatment. Complete, uncensored observations were considered lethal consequences and the occurrence of relapses and metastases. The database for conducting the research was created in accordance with the informational model developed at the beginning of the research 1. The WizWhy package of the Data Mining category and non-parametric statistics methods of the «STATISTICA» package (Mann–Whitney test, Pearson’s chi-square test) were used for information processing. Data by text are presented either as median or as Me (LQ;UQ), where Me is the median, LQ is the lower quartile, and UQ is the upper quartile. Patient survival was calculated using the Kaplan–Meier method. The proposed results are a continuation of the research results presented in 2. The study belongs to the category of retrospective case-control studies. Results. The possibility of using indicators of cervical volume, platelet level, and their relative dynamics as prognostic factors for the appearance of remote oncological consequences of treatment of cervical cancer is shown, and a number of optimal parameters of radiation therapy to prevent their appearance are determined. Conclusions. The methods of mathematical statistics and intellectual data analysis (Data Mining) are a powerful tool for obtaining new non-obvious knowledge from arrays of «used» medical data, provided they are properly formalized and structured.