Background: Cryoablation is currently being investigated as a minimally invasive alternative to breast-conserving surgery. This meta-analysis investigates the local recurrence and residual tumor rates after cryoablation for small early-stage breast cancers.

Methods: A systematic search was conducted on Embase, PubMed, Google Scholar, and the International Clinical Trials Registry Platform from inception to 16 June 2024. Studies of patients with breast cancers ≤ 20 mm treated with cryoablation only or cryoablation followed by surgery were included. Pooled local recurrence rates (cryoablation only) and pooled residual tumors rates (cryoablation followed by surgery) were estimated with mixed-effects models. Between-study heterogeneity was assessed using I² statistics. Where I² exceeded 50%, outlier and influence analysis, followed by sensitivity analysis excluding outliers, were conducted.

Results: Twelve studies met inclusion criteria, of which 7 studies (530 female patients, 531 breast tumors) reported on patients treated with cryoablation only and 5 studies (220 female patients, 222 breast tumors) reported on patients treated with cryoablation followed by surgery. For studies on cryoablation only, pooled local recurrence rate was 1.1% (95% CI 0.42-3.03%) with low between-study heterogeneity (I² value = 0%; 95% CI 0.0-70.8%; p = 0.95). For studies on cryoablation followed by surgery, pooled residual tumor rate was 12.0% (95% CI 3.85-31.64%); however, substantial between-study heterogeneity (I² value = 76.1%; 95% CI 41.7-90.2%; p < 0.01) was present. Influence analysis revealed 1 outlier study. When this study was excluded, pooled residual tumor rate was 8.2% (95% CI 3.84-16.68%) with improvement in heterogeneity (I² value = 0%; 95% CI 0.0-84.7%; p = 0.64).

Conclusion: Pooled local recurrence and residual tumor rates after cryoablation are comparable to local recurrence rates after breast-conserving therapy and re-excision rates following breast-conserving surgery, respectively. These results are encouraging but should be interpreted with caution due to lack of comparative studies.