Klin Onkol 2026; 39(5): 337-343. DOI: 10.48095/ccko2026337.
Background: Surgical resection of residual retroperitoneal masses (RPLND) is the standard of care for advanced non-seminomatous germ cell tumors of the testis (NSGCT) following curative chemotherapy. However, only necrosis or fibrosis is present in the resectate in approximately 40–50% of patients. Serum microRNAs represent promising biomarkers of germ cell tumors that could contribute to the identification of a residual viable tumor after systemic treatment. The aim of this study was to evaluate the diagnostic yield of selected microRNAs in predicting the histopathological findings in patients indicated for RPLND after chemotherapy. Patients and methods: A cohort of 18 patients with clinical stage II and III NSGCT was analyzed at the Masaryk Memorial Cancer Institute between 2021 and 2023 after completing curative chemotherapy. In all patients, levels of five specific serum microRNAs (miR-371a-3p, miR-223-3p, miR-383-3p, miR-517a-3p, and miR-519a-3p) were analyzed before beginning chemotherapy, and subsequently before and after the planned surgical procedure (RPLND). Results were correlated with the definitive histopathological findings in the resectate to assess the diagnostic yield and predictive value of these biomarkers for the presence of a viable tumor, teratoma, or necrosis. Results: Eighteen patients with advanced NSGCT were included in the analysis. The most frequent baseline clinical stages were stage IIB (33%; N = 6) and IIC (22%; N = 4). Histopathological examination of the residual tissue after PC-RPLND demonstrated necrosis in 55% (N = 10), teratoma in 38% (N = 7), and a viable tumor in 5% (N = 1). Serum tumor markers (alpha-fetoprotein – AFP, human chorionic gonadotropin – hCG, and lactate dehydrogenase – LDH) showed the expected decrease following the initiation of systemic therapy, which persisted after RPLND. The median AFP level decreased from 17.4 before therapy to 3.7 before RPLND and further to 2.5 after RPLND. A similar trend was observed for hCG and LDH. Receiver operating characteristic analysis of miRNAs as potential predictors of a teratoma or viable tumor after RPLND did not demonstrate sufficient discriminatory ability for any of the evaluated variables. Area under the curve values for most miRNAs ranged between 0.45 and 0.70, and none of the tested models reached statistical significance. Conclusion: Currently, serum microRNA levels do not reach sufficient sensitivity or specificity for decision-making regarding the indication of surgical resection of residual masses after chemotherapy in NSGCT patients. Although these biomarkers exhibit high diagnostic potential in other stages of the disease, their use in post-chemotherapy remains limited, and surgical resection of residual masses continues to represent a standard component of the curative approach.