Breast-conserving surgery (BCS) for early breast cancer patients has been well established for a number of years [1234]. Although BCS is effective in prolonging survival compared with radical mastectomy, locoregional recurrence (LRR) rates are higher after BCS than following mastectomy [5678]. Some studies have reported rates of LRR of 11% to 16% [910]. Ipsilateral breast tumor recurrence (IBTR) following treatment with BCS and radiotherapy occurs in approximately 8% to 20% of patients [234811].

If IBTR is detected during follow-up, total mastectomy is considered standard treatment because postmastectomy radiotherapy is performed after the initial BCS and the risk of further residual recurrences increases with time after partial mastectomy [12131415]. Nonetheless, in some cases, partial mastectomy may be performed instead of total mastectomy as a second operation. There is a report that suggests the type of surgery does not affect survival, therefore, BCS can be considered in selected patients with IBTR [13]. Known risk factors for IBTR include age, race, obesity, size of the pathological tumor, time interval to recurrence, lymph node status, estrogen receptor (ER) status, progesterone receptor (PR) status, and receipt of adjuvant radiotherapy or systemic therapy [910111617]. However, risk factors affecting survival after a second operation following IBTR are not as well understood as the risk factors related to IBTR. In this study, we analyzed independent prognostic factors affecting overall survival after a second operation for IBTR.

We retrospectively analyzed the medical records of 11,073 patients who underwent breast cancer surgery at Samsung Medical Center between November 1995 and December 2011. LRR occurred in 787 (7.1%), of which 622 including 471 with distant metastases, 109 who underwent total mastectomy at the first operation and 42 with regional recurrences were excluded. Of the remaining 165 patients who developed IBTR, eight refused further treatment, leaving 157 eligible for analysis. These included 28 patients (17.8%) who underwent partial mastectomy and 129 patients (82.2%) who underwent total mastectomy. Of these, 26 patients with incomplete data were excluded from the survival analysis at the second operation (Figure 1).

Patient data, tumor characteristics, and adjuvant treatment information were stratified according to the order of diagnosis (Table 1). Patient data, tumor characteristics, and adjuvant treatment information after the second operation following IBTR were analyzed using a Fisher exact test (Table 2). Age and durations were analyzed using the Mann-Whitney U-test after applying a Shapiro-Wilk test for normality (Table 2). Univariable and multivariable analyses were conducted using Cox regression analysis model. Survival from the date of the first operation to the time of death was estimated using the Kaplan-Meier method. A 95% confidence interval (CI) for survival was generated after estimating the standard error using Greenwooding Greenwowith variables affecting overall survival. Independent prognostic factors affecting overall survival were analyzed separately for the first and second operations. Univariable and multivariable analyses used age, duration from the first to the second operation, histopathology, multiplicity, nuclear grade, tumor size, resection margin, extensive intraductal component, ER/PR/human epidermal growth factor receptor 2 statuses, in addition to, adjuvant chemotherapy, radiotherapy, and endocrine therapy as variables (Table 3). Statistical significance was established at p<0.05. All statistical analyses were performed using Statistical Analysis System (SAS) Software version 9.3 (SAS Institute Inc., Cary, USA) and R Statistical Programming Language version 2.13.2 (The R Foundation, Vienna, Austria; available at http://www.R-project.org/). The study was approved by Institutional Review Committees (2015-07-150) and work was conducted according to the principles outlined in the Helsinki Declaration.

Patient data, tumor characteristics, and adjuvant treatment information for the first and second operations are described in Table 1. The median age of the patients at the first operation was 43.6 years (range, 27-69 years). Median disease-free duration from the first to the second operation was 45.0 months (range, 2.5-164.6 months). Of the 131 patients analyzed who underwent a second operation, 23 patients (17.6%) underwent partial mastectomy and 108 patients (82.4%) underwent total mastectomy (Table 2). For the survival analysis, 26 patients with incomplete data for one or more variables for the first or second operations were excluded (Figure 1). Histopathology of the tumors revealed invasive ductal carcinomas in 121 patients (77.1%) with no invasive or in situ lobular carcinomas reported at the first operation.

At the time of the first operation, most patients had stage I (n=96, 61.1%) or stage II (n=52, 33.1%) breast cancer. Systemic treatments included chemotherapy for 69 patients (44.0%) and endocrine therapy for 59 patients (37.6%) after the second operation (Table 1). Patient characteristics stratified according to the second operation method following IBTR were analyzed (Table 2). Histopathology of the tumors revealed mainly invasive ductal carcinomas in both groups. Adjuvant radiotherapy was administered to 13 patients (56.5%) after partial mastectomy in contrast to three patients (2.7%) after total mastectomy (Table 2).

Univariable analyses showed duration from the first to the second operation, multiplicity, nuclear grade, lymph node status, tumor stage, ER/PR/human epidermal growth factor receptor statuses, and adjuvant radiotherapy at the first operation, in addition to, nuclear grade, ER/PR statuses, and endocrine therapy at the second operation to be significantly associated with survival outcome (Table 3). In the multivariable analyses, duration from the first to the second operation (hazard ratio [HR], 0.98; 95% CI, 0.97-0.99; p<0.001), histopathology (HR, 0.42; 95% CI, 0.23-0.78; p=0.006), and adjuvant chemotherapy (HR, 0.38; 95% CI, 0.21-1.06; p=0.002), and radiotherapy (HR, 0.47; 95% CI, 0.24-0.89; p=0.022) at the first operation, in addition to, positive ER status (HR, 0.46; 95% CI, 0.22-0.96; p=0.039) and endocrine therapy (HR, 0.52; 95% CI, 0.30-0.91; p=0.022) at the second operation were found to be significantly associated with an overall increase in survival following IBTR. Conversely, positive lymph node status (HR, 3.71; 95% CI, 1.51-9.10; p=0.004) and endocrine therapy (HR, 1.91; 95% CI, 1.09-3.35; p=0.024) at the first operation were found to be significantly associated with an overall decrease in survival after salvage operation following IBTR (Table 3).

Overall survival rates were estimated for the 131 patients who experienced IBTR using the Kaplan-Meier method (Figure 2). One-, 3-, and 5-year overall survival rates were 98.2% (95% CI, 0.927-0.996), 90.6% (95% CI, 0.827-0.950), and 87.1% (95% CI, 0.774-0.928), respectively. The Kaplan-Meier method was also used to estimate overall survival rates of patients stratified according to the method of the second operation (Figure 3). One-, 3-, and 5-year overall survival rates of patients who underwent total mastectomy were 97.8% (95% CI, 0.918-0.994), 90.0% (95% CI, 0.808-0.949), 85.8% (95% CI, 0.748-0.923), respectively. Conversely, 1-, 3-, and 5-year overall survival rates of patients who underwent partial mastectomy were 93.3% (95% CI, 0.618-0.990), 93.3% (95% CI, 0.618-0.990), 93.3% (95% CI, 0.618-0.990), respectively. Analysis of survival between the two patient groups reported the HRs not to be significantly different according to the method of the second operation using a Cox regression analysis model (p=0.435).

This retrospective analysis of patients treated with BCS for early breast cancer revealed independent prognostic factors affecting overall survival after a second salvage operation. Several previously published reports have defined IBTR as recurrent invasive carcinoma in either the skin or parenchyma of the ipsilateral breast after BCS in the absence of clinical and radiological evidence of regional or distant metastases [910]. Reported rates of LRR after BCS are 11% to 16% [910]. IBTR rates are generally reported at 8% to 20% [12348101118]. In this study, the rate of IBTR was 21% (n=165), which is similar to other studies.

Previously identified risk factors known to affect overall survival after IBTR include age, race, body mass index, size of the pathological tumor, ER status, diploidy, a BRCA1/2 mutation, lymph nodal status at relapse, and adjuvant therapy [910121519202122]. Anderson et al. [9] have reported that IBTR has a greater impact on mortality in ER-negative than ER-positive patients and that older patients with larger tumors (>2.0 cm) have significantly higher mortality rates compared to younger patients with smaller tumors. Body mass index and black ethnicity were also reported to be significantly associated with mortality. Adjuvant therapy after IBTR significantly influences overall survival. The use of adjuvant radiotherapy and systemic treatments, such as adjuvant chemotherapy or endocrine therapy, has reduced the incidence of IBTR and increased the overall survival following IBTR [1223]. Histopathology, lymph node status at the first operation, hormonal receptor status, and adjuvant chemotherapy were also significantly associated with overall survival in this study. Furthermore, we also identify disease-free durations between the first and the second operations as factors affecting overall survival after a second operation following IBTR in our study. Data are limited on resection margins for second operations in other studies because total mastectomy is usually performed for IBTR [910121324]. Resection margins were also not associated with survival rates in our study. Time intervals between treatment of the primary breast cancer and IBTR greatly affected overall survival [101119212225]. Our results on overall survival after IBTR were in concordance with previous studies.

Recurrent cancer is generally treated by total mastectomy after IBTR because curative resection is regarded as the standard treatment [10141526]. Approaches that use partial mastectomy for reoperation with repeated radiation exposure to the residual breast are an emerging problem. Alpert et al. [12] reported that while mastectomy remains the standard of treatment for in-breast relapse after BCS and radiotherapy, partial mastectomy appears to be feasible for selected patients with favorable tumor biology and early detection of local relapse. They also suggested that partial mastectomy may be acceptable as a second operation for selected patients with tumors <3 cm in diameter, tumors confined to the biopsy site without skin or lymphovascular invasion, and fewer than three positive lymph nodes. According to Alpert et al. [12] approximately 20% of patients with IBTR received partial mastectomy as a second operation. In our study, partial mastectomy was performed on 17.6% (n=23) of patients as a second operation after IBTR, similar to the previous study. Most patients in other studies had early T stage tumors at diagnosis with no multiplicity. In our study, 82.6% of tumors were <2 cm, 17.4% were 2 to 5 cm and 95.8% were tumors without multiplicity in the second partial mastectomy group (Table 2). Alpert et al. also reported 20.5% T1 and T2 stage tumors because surgeons do not usually select patients with large tumor or tumors with multiple sites for second BCS. Comparing partial mastectomy and total mastectomy groups for overall survival is difficult because of differences in median follow-up durations. However, partial mastectomy can be done instead of total mastectomy for second operations for IBTR in selected patients with a long disease-free follow-up period and patients given endocrine therapy whose ER status is positive.

Due to its retrospective design, our study has several limitations. First, randomization was not applied when patients with IBTR underwent a second operation. The size of subgroup populations and conduct at a single center limited the statistical power for determining prognostic factors. Second, partial mastectomy as a second operation after IBTR is a recent surgical intervention. This meant relatively few patients underwent partial mastectomy. Further randomized controlled studies are needed on partial mastectomy as a standard treatment instead of total mastectomy. Third, we could not include genetic information and race, which are important prognostic factors for overall survival after a second operation, because genetic analysis was not prevalent in Korea at the time of the operations and because most patients were Asian. Published data on second operations for IBTR were limited and few published reports addressed survival outcomes of patients with second operations. Consequently, second operations for IBTR were unfamiliar and prognostic factors affecting overall survival after second operations were not well understood [121319]. The strength of this study was that we evaluated IBTR and independent factors affecting survival and identified disease-free duration, histopathology, lymph node status and adjuvant therapy at the first operation, and ER status, and endocrine therapy at the second operation as important prognostic factors. Furthermore, this study was conducted using large numbers of patients with IBTR (131 patients) as compared to existing studies and we assessed predictors affecting overall survival through the analysis of multiple variables. In future, additional studies are needed to evaluate recurrence rates and prognoses according to surgery type after second operations for IBTR.

We evaluated prognostic factors affecting overall survival after a second operation following IBTR. The time interval to IBTR after BCS, histopathology, lymph node status, and adjuvant therapy at the first operation, and ER status and endocrine therapy at the second operation were related to overall survival after surgical intervention for IBTR. We could see better prognoses after a second operation following IBTR in patients with a long disease-free interval and recognized the importance of performing optimal treatment, such as adjuvant chemotherapy, radiotherapy, and endocrine therapy according to pathologic status, lymph node status, and hormonal receptor status after the first operation because those risk factors affect overall survival after a second operation for IBTR following BCS.