RESUMO
Purpose To compare the safety and efficacy of US-guided percutaneous radiofrequency ablation (RFA) as a local treatment for breast cancer with that of lumpectomy. Materials and Methods A prospective, randomized open-label phase II clinical trial (clinicaltrials.gov identification number NCT02281812) was conducted in a single institution from 2013 to 2017. Women with invasive ductal carcinoma of the breast measuring 2 cm or smaller were randomly assigned to receive RFA or lumpectomy alone (control group). Margin status at surgery, tumor cell viability after RFA (with nicotinamide adenine dinucleotide [NADH] and cytokeratin 18 [CK18] staining), cosmetic results, adverse events, and local recurrences were evaluated with univariable and multivariable analyses. Results Forty subjects (20 in the RFA group and 20 in the lumpectomy group) were evaluated. The mean participant age was 64 years (range, 46-86 years). NADH and CK18 staining demonstrated absence of tumor cell viability after RFA with at least one of the two techniques. The surgical margins were positive in 11 of the 20 participants in the lumpectomy group (55%) and four of the 20 in the RFA group (20%) (P = .02). Median follow-up was 25 months (range, 1-83 months). Local breast inflammation after surgery was higher in the RFA group than in the lumpectomy group (40% [eight of 20 participants] vs 5% [one of 20 participants], respectively; P = .01). Local infection occurred in three participants who underwent RFA (two of whom had undergone partial irradiation of the breast). None of the participants in the control group developed local infection. No participants had recurrence or the need for a second surgery during the study period. Conclusion This preliminary study showed that radiofrequency ablation was effective for local tumor control and that tumor-free margins were obtained more often with radiofrequency ablation than with lumpectomy. Surgical excision after radiofrequency ablation was infrequently associated with local infection. © RSNA, 2018 Online supplemental material is available for this article.
