Aim. To evaluate the clinical efficacy of radiofrequency ablation in high-risk patients with localized renal cell carcinoma and to assess the limitations of glomerular filtration rate in post-ablation renal function assessment.

Materials and methods. This single-center retrospective cohort included 24 patients with localized renal tumors treated with radiofrequency ablation between 2008 and 2019 at the Zaporizhzhia Regional Antitumor Center. Indications comprised solitary kidney (n = 5), bilateral tumors (n = 3), and local treatment in the setting of recurrent or metastatic disease (n = 18); categories were not mutually exclusive. A percutaneous approach was used in 21 (87.5 %) patients, laparoscopic in 1 (4.2 %), and open in 2 (8.3 %). Core-needle tissue samples from macroscopically intact parenchyma adjacent to the ablation zone were obtained intraoperatively and 72 hours after ablation. Immunohistochemical analysis of CD34 and HIF-1α expression was performed and correlated with estimated glomerular filtration rate. The Wilcoxon signed-rank test was used for statistical evaluation.

Results. Complete tumor necrosis was observed in 75–100 % of cases. Most complications were minor (Clavien–Dindo I–II: 29.2 %), while serious adverse events (IIIa–IV) occurred in three patients (12.5 %). The mean hospital stay was 3.0 ± 0.8 days. Clear cell carcinoma was the predominant histology – (83.3 %), with papillary carcinoma in 16.7 %. Tumor size ranged from 2.1 cm to 4.0 cm (mean 3.2 ± 0.5 cm); 37.5 % of patients had lesions >3 cm. Postoperative immunohistochemistry showed a significant decrease in CD34 expression (100 ± 15 vs. 58 ± 12; p = 0.014) and an increase in HIF-1α levels (25 ± 8 vs. 78 ± 14; p = 0.008) despite a stable estimated glomerular filtration rate (62.4 ± 7.8 mL/min/1.73 m2 vs. 61.9 ± 8.1 mL/min/1.73 m2, p = 0.74).

Conclusions. Radiofrequency ablation is an effective nephron-sparing option for high-risk patients with localized renal cell carcinoma. However, stable estimated glomerular filtration rate values may mask subclinical parenchymal injury; tissue-level biomarkers capture structural and microvascular alterations. These findings support prospective validation.