Within our overall cohort, 37% of patients with ≥ 3 positive SLNs from breast cancer did not undergo ALND. Additionally, omission of ALND increased significantly from 2018–2020. We found that omission of ALND was independently and significantly associated with older age, lower grade tumors, BCS, lack of radiation, non-academic facility, higher number of SLNs examined and lower number of positive SLNs. Within our Z11 subgroup, ALND was omitted in 49% of patients with ≥ 3 positive SLN. Though not statistically significant, omission of ALND increased from 2018 to 2020.
The ACOSOG Z0011 phase 3 randomized clinical trial included women with clinically node-negative, T1/T2 breast cancer, and 1–2 positive SLNs who underwent BCS and whole breast radiation. Patients were randomized to axillary observation versus completion ALND [1]. The 10-year overall survival, loco-regional recurrence, and disease-free survival rates were not significantly different between ALND and observation [10, 11]. Based on these findings, breast cancer management guidelines were updated to avoid ALND for patients who fulfill Z11 inclusion criteria [12].
Other studies have also assessed the impact of ALND vs observation for patients with SLN metastases. Recently, the multicenter, randomized clinical trial SINODAR-ONE reported similar 3-year survival and relapse rates with ALND vs observation for patients with T1/2 breast cancers and 1–2 positive SLNs with macrometastases [13]. IBCSG 23 − 01 reported similar 5-year and 10- year disease-free survival rates with ALND vs observation for patients with T1/T2 tumors, no palpable axillary adenopathy and ≥ 1 positive SLN with micrometastasis (≤ 2mm) [14, 15]. The multicenter, randomized clinical trial AMAROS (EORTC 10981–22023 AMAROS) reported no significant differences in disease-free survival (DFS), overall survival and axillary recurrence rates with ALND vs axillary radiotherapy for patients with no palpable lymph nodes, T1/2 breast cancer, and ≥ 1 positive SLN with micro or macrometastasis [16]. The Z11 trial only included patients who underwent BCS, while these other clinical trials included a small proportion of patients who underwent mastectomy. The inclusion criteria of these trials likely influence the surgical management of SLN metastases in the United States, because we found that patients undergoing BCS were more likely to omit ALND.
SLN biopsy is currently recommended by the National Comprehensive Cancer Network(NCCN) guidelines for patients with no palpable lymph nodes and ≤ 2 nodes that are positive through needle biopsy or are suspicious on imaging [12]. NCCN guidelines recommends that if the SLN is positive and the patient undergoes BCS, then all the following criteria must be met to avoid an ALND: cT1/T2 disease, ≤ 2 positive SLNs, no neoadjuvant chemotherapy, and planned whole breast radiation therapy. If a mastectomy is completed, similar criteria must be met, however, axillary directed radiation therapy must be included [12].
The data regarding outcomes of patients with ≥ 3 positive SLNs are limited. In the AMAROS trial, only 5% had ≥ 3 positive SLNs [16]. In the IBCSG 23 − 01 randomized controlled trial, only 1 patient with 3 positive SLNs was included [14]. A meta-analysis compared ALND vs SLNB alone in patients with T1/T2 breast cancer and included patients with ≥ 3 positive SLNs. No difference in survival was reported, however, < 9% of patients had ≥ 3 positive SLNs within each group [17]. Bonneau et al published a retrospective cohort study using the Surveillance, Epidemiology, and End Results (SEER) database in 2015 assessing overall survival and specific survival of T1-T2 breast cancer patients with ≥ 3 positive lymph nodes who underwent SLNB alone vs ALND. Overall and cancer-specific survival were not different between the two groups and only 4% of patients received SLNB alone [18]. Additionally, this study was published prior to SEER releasing the variable “SLNs positive” in 2018.
Previous NCDB studies are limited due to the lack of codes tracking the type of axillary surgery and/or the number of positive SLNs [19, 20]. A retrospective cohort study by Yao et al utilized NCDB to assess patients who fulfilled the majority of Z11 criteria similar to our study (BCS, T1-T2, ≥ 3 positive lymph nodes). The authors found that SLNB alone increased significantly from 1998 to 2011, however SLNB and ALND were arbitrarily clarified as ≤ 4 nodes vs ≥ 10 nodes respectively [4]. Recently, Stafford et al utilized the new 2018 NCDB codes reporting positive SLNs to assess trends in axillary surgery, however the study limited the cohort to patients with 1–2 positive SLNs [21].
To our knowledge, this is the first study to use the NCDB to accurately assess the trends of axillary surgery in patients with ≥ 3 positive SLNs. Previous studies are limited in the availability of NCDB codes to accurately track the type of axillary surgery completed as well as the number of positive SLNs. Additionally, most of our cohort consists of patients who underwent mastectomies when literature has focused on BCS in the past.
We recognize several limitations of our study, including lack of survival data due to the recent time period of the study. Tumor registrar coding of American Joint Committee on Cancer (AJCC) staging in the NCDB also has the potential for error. Of note, a large percentage of the cohort was recorded to have ≥ 6 SLNs examined, despite specific coding instructions on identification and separation of examined SLNs versus overall regional lymph nodes removed. A histogram depicting the distribution of reported SLNs is included (Online Resource). This observation may be related to over-counting SLNs due to the technique of cutting SLNs. Weaver describes this problem when the initial pathologist/technician cuts the SLN, but the final pathologist counts each individual cut as a node [22]. Furthermore, NCDB does not record the extent of SLN involvement: the size of metastatic deposits or the presence of extranodal extension. As this is a retrospective study, selection bias regarding which patients undergo SLNB alone or ALND is present. Despite these limitations, this study includes a large number of patients with ≥ 3 positive SLNs and represents the current management in the United States.