Background
Pure mucinous breast carcinoma (PMBC) is a rare type of breast carcinoma involving abundant extracellular mucin production, which accounts for approximately 1–6% of all cases of breast cancer [
1,
2]. PMBC has distinct clinicopathological and molecular features, including higher estrogen receptor (ER) and progesterone receptor (PR) expression, greater likelihood of human epidermal growth factor receptor-2 (HER2)-negative status, lower grade, and lower risk of nodal metastasis [
3‐
5], which all contribute to better outcomes compared to invasive ductal carcinoma (IDC); indeed, the 10-year disease-free survival rate is > 90% [
6‐
12]. In addition, the median age of PMBC patients was 70 years, which was significantly greater than the age of those with other histological subtypes [
11]. In clinical practice, the recommendation for adjuvant treatment of PMBC differs from that for the usual breast cancer histology [
13].
Prior studies have found that breast-conserving surgery (BCS) is an appropriate surgical procedure for most patients with early-stage PMBC [
11,
14]. In patients with invasive breast carcinoma, several prospective clinical trials have indicated that the omission of postoperative radiotherapy (RT) following BCS is safe and associated with an acceptable low risk of local recurrence and without a detriment to overall survival (OS) among female patients who are elderly (aged ≥50, 65, or 70 years), tumor size ≤5 cm (T1–2), node-negative (N0) disease, and ER-positive tumors [
15‐
18]. However, none of these trials specified whether PMBC patients were enrolled. In addition, in the trials evaluating the omission of postoperative RT, endocrine treatment was mandatory and in fact, there were statistically significant differences in local control rates, even though the recurrence rates were very low in general [
15‐
18]. Until today, there is no clear recommendation on the best management of elderly PMBC patients with low risk for local recurrence. In light of this, we used data from the large and contemporary Surveillance, Epidemiology, and End Results (SEER) program to determine population-based practice patterns and survival outcomes in PMBC patients receiving postoperative RT, particularly among the elderly population.
Discussion
In the present study, we aimed to assess the practice patterns and survival outcomes of postoperative RT following BCS in elderly PMBC patients, and found that the use of postoperative RT has decreased in recent years, even though it was associated with better BCSS before and after PSM. To our knowledge, this is the first study to assess the effect and administration of postoperative RT after BCS among elderly PMBC patients.
Although the present study only included patients aged ≥65 years, the patient characteristics in the selection flowchart showed that most patients had small tumor size, lower tumor grade, and ER- and PR-positive disease, consistent with that noted in previous studies [
3‐
5]. In addition, most patients also had HER2-negative disease [
3,
5]. Due to the less aggressive behavior of PMBC, the prognosis was excellent, with a 10-year BCSS rate of up to 96.6% in the present study. A retrospective analysis from a single institute or population-based study also showed an excellent BCSS of > 90% [
3,
24].
Furthermore, the period of diagnosis impacted the decision of surgical procedure. A study by Sas-Korczyńska et al. included patients from 1952 to 2002 (
n = 94) found that most patients were treated with mastectomy (95.7%) and only 4.3% received BCS [
12]. Anan et al. also found that only 14.5% of patients received BCS between 1976 and 1998 (
n = 76) [
25]. However, 79.3, 51.3, and 64.1% of patients diagnosed in 1990–2010, 2014–2016, and 1983–2014, respectively, at 3 Chinese institutes still received mastectomy [
10,
26,
27]. Several recent studies from the US showed that > 60% of patients were treated with BCS, and that this rate remained relatively constant between 1998 and 2018 [
6,
11,
28]. The selection flowchart of our study also confirmed that approximately 60% of patients were treated with BCS, and that the prognosis in this patient subset was excellent. Our results also confirmed that BCS is an appropriate treatment for low-risk early-stage PMBC.
Previous clinical trials have sought to compare the outcomes between adjuvant endocrine treatment alone and adjuvant RT plus endocrine treatment following BCS in elderly patients with invasive breast carcinoma [
15‐
18]; however, the histological subtypes of the enrolled patients were not recorded in those studies. As there are significant differences in the clinicopathological and molecular features between IDC and PMBC, it is unclear whether RT can be safely omitted in PMBC patients. However, in the present study, 64.1% of elderly patients were treated with RT, and the percentage of receiving postoperative RT significantly decreased from 2004 to 2010 (59.5%), relative to that from 1990 to 2003 (71.1%). The results of 2 prospective clinical trials were published in 2004, which showed that adjuvant RT plus tamoxifen was associated with better locoregional control but no significant difference in the rates of distant recurrence and OS compared to tamoxifen alone [
15,
16], which could impact the treatment decision of postoperative RT after BCS in PMBC. Given the decrease in the use of postoperative RT in recent years, these findings could impact patient counseling, which is currently supported by findings from existing randomized studies conducted in IDC patients and may not be suitable due to the inherent lower aggressiveness of PMBC.
To our knowledge, no study has assessed the rates and patterns of local recurrence stratified based on whether postoperative RT was administered. Barkley et al. examined 111 PMBC patients (median age, 56 years), including 67% (
n = 64) treated with BCS and 95.5% treated with RT following BCS. Over a median follow-up duration of 63 months, they found that 2 patients with stage T1 N0 and HR-positive disease developed local recurrence after adjuvant RT [
26]. In addition, Vo et al. examined 61 patients (median age, 60 years) treated with BCS, including 90% treated with RT, and 3 patients were found to have developed local recurrence [
14]. Gwark et al. assessed 471 patients from Korea (mean age, 46 years), including 65.6% who were treated with RT, and found a total of 34 relapses, including 10 patients of local recurrence [
4]. Although we could not compare the effect of RT on local recurrence, based on the above results, the local recurrence rate of PMBC should be very low after postoperative RT. Hence, the administration of postoperative RT could potentially improve the outcomes of PMBC patients.
Limited studies were available to compare the outcomes of PMBC based on the administration of postoperative RT. A prior SEER study created a clinical nomogram to predict the outcomes of early PMBC [
24]; however, the role of postoperative RT was not included in the nomogram. Gwark et al. found that RT was related to survival outcomes in univariate analysis, whereas no significant association with outcomes was observed in the multivariate analysis [
4]; however, the surgical procedure was not specified in that study. Hence, we were unable to determine the role of RT on the outcomes on different surgical procedures. Another study from the SEER database that included 11,400 PMBC patients indicated that the addition of postoperative RT did not significantly improve OS; however, in that study as well, the surgical procedure was not analyzed [
9]. To our knowledge, this is the first study to indicate the benefit of postoperative RT on BCSS in elderly PMBC patients, compared to those without postoperative RT. A large retrospective cohort study found that approximately 30% of PMBC patients did not receive postoperative RT following BCS [
11], consistent with our findings in the elderly population. Thus, our findings show that postoperative RT should not be omitted following BCS in elderly PMBC patients.
Our study has certain limitations. First, inherent bias is possible in any retrospective study. Second, a centralized pathology review of the SEER database was lacking. Nevertheless, the survival rate of patients was similar to that of previous studies [
6‐
12]. Thus, our study might still be representative of the relevant population. In addition, the details regarding endocrine therapy, chemotherapy regimen, radiotherapy dose, completeness of chemotherapy and radiotherapy, and the pattern of locoregional and distant recurrence were also not recorded in the SEER database. Finally, the percentage of patients receiving RT was under-reported in the SEER database [
29]. Despite these limitations, our study offers new insights on the impact of postoperative RT on prognosis after BCS in PMBC patients. The SEER program provides us with a large sample size that can be used to analyze rare breast cancer histologies using robust statistical methods with sufficient power to draw conclusions. Thus, we believe that our findings will contribute to the current knowledge on the effect of postoperative RT following BCS in low-risk elderly PMBC patients.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.