A search for efficient biomarkers of ovarian cancer is one of the current trends in gynecologic oncology. Metabolic profiling by ultra high-performance liquid chromatography and mass spectrometry (UHPLC-MS) yields information about the total set of low-molecular-weight metabolites of a patient's biological fluid sample. The metabolites may provide potential disease markers, and their combination with microRNA level data significantly increases the diagnostic value. To identify the potential noninvasive diagnostic markers of serous ovarian adenocarcinoma, the metabolomic profile and microRNA transcript levels were studied in urine samples of patients. The study included 60 patients diagnosed with serous ovarian adenocarcinoma and 20 women without a cancer history. Chromatographic separation of metabolites was performed on a Vanquish Flex UHPLC system coupled to an Orbitrap Exploris 480 mass spectrometer. A search for gene regulators of metabolites and microRNA regulators of genes was carried out using the Random forest machine learning method. The microRNA transcript levels in the urine were determined by real-time PCR (qPCR). LASSO-penalized logistic regression was used to build predictive models. In total, 26 compounds showed abnormal concentrations in the ovarian cancer (OC) patients compared with the control group, the set including kynurenine, phenylalanyl-valine, lysophosphatidylcholines (18:3, 18:2, 20:4, and 14:0), alanylleucine, L-phenylalanine, phosphatidylinositol (34:l), 5-methoxytryptophan, 2-hydroxymyristic acid, 3-oxocholic acid, indoleacrylic acid, lysophosphatidylserine (20:4), L-β-aspartyl-L-phenylalanine, myristic acid, decanoylcarnitine, aspartyl-glycine, malonylcarnitine, 3-hydroxybutyrylcarnitine, 3-methylxanthine, 2,6-dimethylheptanoylcarnitine, 3-oxododecanoic acid, N-acetylproline, L-octanoylcarnitine, and capryloylglycine. Metabolite-gene regulator (47 genes) and metabolite-microRNA regulator (613 unique microRNAs) relationships were established by the Random forest method. Levels of 85 microRNAs were validated by qPCR. Changes in transcript levels in the OC patients compared with the controls were observed for miR-382-5p, miR-593-3p, miR-29a-5p, miR-2110, miR-30c-5p, miR-181a-5p, let-7b-5p, miR-27a-3p, miR-370-3p, miR-6529-5p, miR-653-5p, miR-4742-5p, miR-2467-3p, miR-1909-5p, miR-6743-5p, miR-875-3p, miR-19a-3p, miR-208a-5p, miR-330-5p, miR-1207-5p, miR-4668-3p, miR-3193, miR-23a-3p, miR-12132, miR-765, miR-181b-5p, miR-4529-3p, miR-33b-5p, miR-17-5p, miR-6866-3p, miR-4753-5p, miR-103a-3p, miR-423-5p, miR-491-5p, miR-196b-5p, miR-6843-3p, miR-423-5p and miR-3184-5p. Thus, significant metabolomic imbalance in the urine was observed in the OC patients and was associated with changes in the levels of microRNAs that regulate the signaling pathways of the metabolites. The 26 compounds with abnormal concentrations and the levels of the microRNAs miR-33b-5p, miR-423-5p, miR-6843-3p, miR-4668-3p, miR-30c-5p, miR-6743-5p, miR-4742-5p, miR-1207-5p, and miR-17-5p in the urine were considered to be suitable as noninvasive diagnostic markers of OC.