Ovarian cancer is the second most fatal gynecological cancer. Early detection, which could be achieved through widespread screening, has not yet had an impact on mortality. The aim of our pilot study was to investigate the expression of miRNAs analyzed by a human miRNA microarray chip in urine and serum of patients with ovarian cancer. We analyzed three serum and three urine samples from healthy donors and five serum and five urine samples from patients with ovarian cancer taken at first diagnosis, before any treatment. We selected the seven miRNAs with the highest expression fold change in the microarray chip (cancer vs. control) in urine and serum, for validation by qPCR. We were able to validate two of the seven miRNAs in serum. In contrast to these findings, we were able to validate all of the top seven miRNAs identified in urine using qPCR. The top seven miRNAs in urine identified by microarray chip showed significantly greater differences in expression between patients with ovarian cancer and healthy donors compared to serum. Based on our finding, we can suggest that urine as a biomaterial is more suitable than serum for miRNA profiling by microarray chip in the search for new biomarkers in ovarian cancer.