This pilot study indicated that there was no evident of statistic difference in 3-year prognosis between MWA without subsequent lumpectomy and BCS for 2-5cm BC patients during an interim follow-up period. Furthermore, seven multiple additional survival models showed similar results both in adjusted and unadjusted cohort in the evaluation tumor DFS and TP. Compared to BCS, MWA had shorter operative time, less blood loss, lower postoperative complication rate, better cosmetic results while higher costs.
Several prior systemic reviews have concluded ablation treating ≤ 2 cm BC is safe and effective and the complete ablation rate has increased from 75–90% in the last decade with the technique development(11–13). However, for > 2 cm BC, which is still in its initial exploratory stages, and whether percutaneous management can achieve similar effects to surgery is debatable(9). In our study, although the preliminary prognostic results is inspiring and tumor size was not a influence factor for DFS and TP in univariate and multivariate Cox proportional hazard models, it is necessary to point out that the completion ablation is still a great challenge for 2-5cm BC in generalization because of our study being conducted in two grade-A tertiary hospitals with experienced doctors in ablation, which are the most mature medical centers for ablation technique in the treatment of BC in China.
Currently, four previous studies of US-guided ablation without subsequent lumpectomy investigated the efficacy of BC ≤ 5cm, and these studies applied high intensity focused US, radiofrequency, cryoablation and MWA with small samples for > 2cm BC, respectively(14–17). Among them, Yu et al reported the first case-control study of MWA (n = 21) compared with nipple-sparing mastectomy (n = 43), and indicated there was no statistically difference on TP between MWA and nipple-sparing mastectomy (4.8% vs 2.3%; P = 0.16) with a median follow-up of 26.7 months with no cancer-related death and major complications in either group(14). In the multicenter single-arm study using radiofrequency, a total of 386 patients with BC ≤ 3.5cm were included with a median follow-up of 50 months and the tumor size of 92% enrolled patients were ≤ 2cm(16). Local recurrence was more frequently developed in patients with initial tumor sizes > 2 cm (3/30, 10%) than in those with initial tumors ≤ 2 cm (8/355, 2.3%; P = 0.015). Wu et al revealed high-intensity focused US is safe, effective, and feasible for 2–5cm BC patients and two of 22 (9.1%) patients developed local recurrence (15). Beji et al has indicated that cryoablation were well suited for the palliative care of metastatic patients with ≤ 5 cm BC and the recurrence rate was 11.8% (2/17) in this study (17).
Several studies have proved that the number of peripheral microlesions increased with tumor size and could not be identified clearly on imaging or the naked eyes, which may be an underlying risk for TP and would increase the technical difficulty of ablation(19, 20). As an emerging minimally invasive technology, complete ablation of the peripheral microlesions is a great technical challenge. Therefore, rigorous preoperative evaluation and expanding the adequate safety ablation margin is indispensable in the treatment of > 2cm BC. In addition, second ablation session is necessary to eradicate the suspiciously residual tumor detected by contrast-enhanced images after ablation.
In BCS group, the intraoperative and postoperative pathological tissues can evaluate whether complete resection has been achieved more accurately and intuitively, so that patients can received a second extended BCS when necessary. Many studies reported the five-year recurrence rates of BCS without postoperative radiotherapy were 14–35% and 4–13% for those with postoperative radiotherapy (18). The 3-year TP in our BCS group was 4.54% after PSM, which is comparable to the previous reported data but still needs to be validated with longer follow-up.
One of the most controversial problems in MWA of BC is the management of LNs. In this study, positive LNs were ablated after multipoint core needle biopsies combined with contrast enhanced imaging assessment of the sentinel LNs and axillary LNs in MWA group and no progression on LNs was developed during the follow up period. We attribute the favorable prognosis of LNs to complete ablation of LNs that was suggested to be malignant on biopsy and the postoperative standard comprehensive treatment eradicating the potential residual micro-metastatic LNs. However, whether such assessment and treatment method have the potential to be a minimally invasive approach to LNs management is still in the early stage of exploration and requires long follow-up period with large scales.
From the aspect of major complications, 27 (14.60%) patients in the BCS group developed lymphedema in the surgical side of the upper limb, which can influence the function of patients with obstacles to a certain extent. All the patients resolved with conservative treatment after a median of three months. However, none of the patients in the MWA group developed lymphedema, which may be due to ablation inactivating the metastatic LNs in situ and minimizing the impact on lymphatic return to the least extent. Chronic pain in the operative side of the breast and axilla was the main minor complication in both two groups but was more common in the BCS group (MWA vs BCS: 10.26% vs 40.54%). Compared to the BCS group, which had an average incision of 6.33 cm, while MWA left only a dot-sized wound in the skin as a minimally invasive technique.
The costs were higher in MWA group than in BCS group in our study with the reason of organizing multidisciplinary consultation to conduct comprehensive evaluation of the patient's condition before MWA and 5.13% (2/39) of the patients underwent second ablation session for inadequate ablation margin during the comprehensive imaging evaluation within three days after initial ablation. Although the results in the aspects of costs were not satisfactory, as MWA is an emerging new technology in the local therapy of BC, it is necessary to carefully and thoroughly evaluate the patients’ condition to advance the technical success efficiency. Therefore, it is worthwhile to make such sacrifice in the early stage of MWA applied in BC treatment.
There are several limitations in this study. First, it is a retrospective study and patients were enrolled from dual tertiary-A centers and selection bias was hardly to be avoided in observational studies. However, our study is the largest cohort study of MWA to date and seven different survival models were performed to validate the reliability of outcomes. Anyway, the ablation of 2-5cm BC should be generalized cautiously to hospitals without rich experience. Second, although this study enrolled patients with 2-5cm BC and no statistical difference in the subgroup analyze of tumor size, the number of patients with 3-5cm BC was relatively small, with 69.20% of patients having tumor size 2-3cm. Therefore, it is necessary to study the efficacy of MWA specifically in 3-5cm BC patients with a large sample size in the future. Third, in order to achieve a good balance in baseline characteristics after PSM and to minimize the loss of sample size, patients receiving neoadjuvant therapy were excluded from our study. Thus, the results of studies with current standard comprehensive treatment modality are expected to further verify our results. Finally, the management of LNs in MWA group may has the potential of false negative. Although no progression was observed after ablation of metastatic LNs during the follow-up period, whether such management is effective still needs to be further explored.
In conclusion, there was no statistically significant difference in prognosis between US-guided percutaneous MWA and BCS for patients with 2-5cm BC at a midterm follow-up period in this pilot study. MWA may be a safe and potential alternative local therapy for 2-5cm BC patients who were not candidate for surgery. It's worth noting that ablation of 2–5 cm BC should be cautiously generalized to hospitals without extensive experience.