Skip to main content
SIR 2026 Event Main banner

SIR 2026

Scientific Session

Interventional Oncology

Session: IO - Ablation Science

341 - Volumetric Minimum Ablation Margins Predict Local Control Outcomes for Patients Undergoing Renal Cryoablation

Wednesday, April 15, 2026
12:09 PM - 12:18 PM ET
Purpose: To analyze and establish a volumetric Minimum Ablation Margin (MAM) for T1a renal cell carcinomas treated with CT-guided cryoablation

Materials and Methods: A retrospective analysis of all CT-guided cryoablations performed for curative intent of T1a renal cell carcinomas was performed (n= 125). Fourty- eight were excluded due to MRI guidance or inadequate intraprocedural imaging, yielding a final cohort of 77 patients (mean age 69.0 ± 10.7; 61.5%, male). Intraprocedural images were analyzed in a commercial volumetric software platform (MIM). Both the target lesion and the final (i.e., largest) ice ball were manually segmented, and masks were registered to a common coordinate system. The iceball edge (0C isotherm) was considered the ablation margin. Utilizing point-to-point geometrical analysis, a computational script was used to identify the closest point on the iceball segmentation surface to every point on the surface of the tumor segmentation. The 0th percentile Hausdorff distance was recorded as the true MAM value. Local recurrence was determined by longitudinal assessment of all available follow up imaging.

Results: In the final cohort, 9 patients developed local recurrence (11.5%). The median lesion size was 3.0 cm (IQR: 1.0). Lesions without recurrence had significantly larger MAMs (3.12 ± 2.13) compared with recurrent lesions (-1.78 ± 3.97 mm; p=0.0001). ROC analysis identified an optimal MAM cutoff of 1.34 mm (AUC 0.89, 95% CI 0.77–1), achieving sensitivity 0.77, specificity 0.82, PPV 0.36, NPV 0.96, and accuracy 0.82. On univariable analysis, larger lesion size (OR 2.5 cm; p=0.033) and smaller MAM (OR 0.39 mm; p=0.003) predicted recurrence, while age was not significant (p=0.84). In multivariable analysis, MAM was the only independent predictor of recurrence (OR = 0.36mm; p = 0.004); lesion size was not significant (p = 0.506). A full model (lesion size + Age + MAM) had an AUC value of 0.89 (95% CI: 0.80 - 0.99), similar to the AUC of MAM alone (0.88, 95% CI: 0.77-1). With MAM ≥ 1.34mm recurrence-free survival was significantly higher, as confirmed by a Cox Proportional Hazard Model (log rank p< 0.001). Each 1mm increase in MAM reduced recurrence by 32% (HR = 0.68; p < 0.001; c-index = 0.81).

Conclusion: MAM is a key indicator for lesion control. When MAM ≥ 1.34mm, local recurrence-free survival is significantly higher.