Thermal ablation for small renal masses: Identifying the most appropriate tumor size cut-off for predicting perioperative and oncological outcomes.

ABSTRACT

OBJECTIVES: To test TRIFECTA achievement [1) absence of CLAVIEN-DINDO ≥3 complications; 2) complete ablation; 3) absence of ≥30% decrease in eGFR] and local recurrence rates, according to tumor size, in patients treated with thermal ablation (TA radiofrequency [RFA] and microwave ablation [MWA]) for small renal masses. METHODS: Retrospective analysis (2008-2020) of 432 patients treated with TA (RFA 162 vs. MWA 270). Tumor size was evaluated as 1) continuously coded variable (cm); 2) tumor size strata (0.1-2 vs. 2.1-3 vs. 3.1-4 vs. >4 cm). Multivariable logistic regression models and a minimum P-value approach were used for testing TRIFECTA achievement. Kaplan-Meier plots depicted local recurrence rates over time. RESULTS: Overall, 162 (37.5%) vs. 140 (32.4%) vs. 82 (19.0%) vs. 48 (11.1%) patients harboured, respectively, 0.1 to 2 vs. 2.1 to 3 vs. 3.1 to 4 vs. >4 cm tumors. In multivariable logistic regression models, increasing tumor size was associated with higher rates of no TRIFECTA achievement (OR1.11; P< 0.001). Using a minimum P-value approach, an optimal tumor size cut-off of 3.2 cm was identified (P< 0.001). In multivariable logistic regression models, 3.1 to 4 cm tumors (OR1.27; P< 0.001) and >4 cm tumors (OR1.49; P< 0.001), but not 2.1 to 3 cm tumors (OR1.05; P= 0.3) were associated with higher rates of no TRIFECTA achievement, relative to 0.1 to 2 cm tumors. The same results were observed in separate analyses of RFA vs. MWA patients. After a median (IQR) follow-up time of 22 (12-44) months, 8 (4.9%), 8 (5.7%), 11 (13.4%), and 5 (10.4%) local recurrences were observed in tumors sized 0.1 to 2 vs. 2.1 to 3 vs. 3.1 to 4 vs. >4 cm, respectively (P= 0.01). CONCLUSION: A tumor size cut-off value of ≤3 cm is associated with higher rates of TRIFECTA achievement and lower rates of local recurrence over time in patients treated with TA for small renal masses.