Abstract
BACKGROUND:
Positive margins after Breast conserving surgery (BCS) for breast cancer can result in local recurrence (LR) requiring further surgery. This can lead to unnecessary patient anxiety, poor prognosis and impose additional economic burden to our health system. The aim of this study is to assess the rate of re-excision for positive margins after BCS using the sector resection technique.
METHODS:
This single centre retrospective cohort study included all women who underwent BCS using sector resection between the years of 2012 and 2016. A total of 456 patients underwent sector resection. We evaluated the margin status, re-excision rates and their predictive risk factors.
RESULTS:
415 (91%) patients had clear margins. 41 (9%) patients underwent further re-excision for positive or close margin. 75.6% of those patients had DCIS and 51% had invasive carcinoma involving the margins. Patient and tumour characteristics associated with an increased risk of positive margin were women under the age of 50 (p = 0.19), tumours >50 mm (p = 0.001), grade-2 (p = 0.48) and grade-3 (p = 0.63), HER-2 positivity (p = 0.02), sentinel lymph node positivity (p = 0.03), and patients undergoing axillary lymph node dissection (p = 0.01).
CONCLUSION:
BCS using the sector resection technique has a low re-excision rate for positive margins. Younger patients and aggressive tumour biology are important predictive risk factors for positive margins.
Introduction
High quality mammographic screening has led to early diagnosis of palpable and impalpable breast lesions. In majority of the cases, this early diagnosis has resulted in breast conserving surgery being possible. It is now well established that breast-conserving surgery (BCS) with radiotherapy is the standard of care for early breast cancer [1–8]. BCS with radiotherapy has similar overall survival rates as total mastectomy and offers better cosmetic and psychosocial outcomes [3–5,8–10].
Margin status is an important prognostic factor for local recurrence (LR) after BCS [1–5,11–16]. Positive margins can result in local recurrence requiring further surgery leading to a poorer prognosis, reduced survival, and increased patient anxiety and an added economic burden on the health system. Further surgery can also compromise cosmesis, increase morbidity, and delay adjuvant therapy [7,8,17].
Majority of the LR occur at the site of the primary surgery. The re-excision rates can vary between 17 to 68% in the literature [1–8,12–24]. This variation can be due to the different preoperative imaging techniques, tumour locations, tumour localisation techniques, intraoperative techniques, quality of pathology reporting and the heterogeneity of the studies. Recent randomised controlled trials have revealed that cavity shave margins decreases positive margins and re-excision rates [6,18]. Several studies have reported that no tumour on the inked margin is an acceptable excision margin and it is also suggested that wider excisions are not recommended due to poor cosmetic outcome without providing additional oncological benefit [3–6,15,17–21].
In our institution all breast cancers were managed using a multidisciplinary team (MDT) approach and breast conserving surgery is performed using the sector resection technique [9,10,25,26]. The aim of this retrospective analysis is to assess the re-excision rates for positive margins after BCS using the sector resection technique.
Methods
Patients
In this single centre, retrospective cohort study, all women who underwent BCS using the sector resection between the years 2012 and 2016 (5 years) were included. All the data was collected from an established breast surgery unit database. A total of 456 patients underwent BCS.
All women who underwent BCS for a biopsy proven invasive breast cancer were included, but patients who underwent neoadjuvant chemotherapy, multi-centric cancers and diagnostic biopsies were excluded from the study.
All patients were consented for their clinical and biological data to be used for research purposes. Baseline information including patient’s age, tumour information (size, type, grade, hormone receptor status, HER-2 status, molecular subtypes, margin status), and adjuvant therapy data were collected.
Treatment
All patients were discussed at the multidisciplinary meeting after diagnosis to assess and plan their management. Pre-operative staging included imaging (chest radiograph, whole body bone scan) and blood tests (complete blood count and liver function test). Pre-operative localisation for non-palpable tumours were performed using activated charcoal 120 mg in 3 ml injections (Charcotrace, Phebra Pty Ltd, Australia) and intra-operative specimen assessment using radiography were standard.
All patients underwent BCS using the sector resection technique down to the pectoral fascia. Intra-operatively a specimen x-ray was performed to ensure clear radiological margins. To obtain macroscopically clear margins, intraoperative re-excision of margins was performed in case of positive or close margins at the radiologic specimen evaluation or according to the surgeon’s clinical assessment. Intraoperative frozen section was not performed. A sentinel lymph node biopsy was performed in clinically and radiologically node negative axilla for staging. Patients with positive lymph node on pre-operative workup or macrometastasis in their sentinel lymph node biopsy underwent axillary lymph node dissection (ALND).
After surgery all patients were again discussed at the MDT meeting. Based on the pathological report for patients with positive or close margins further re-excision was performed. Depending on the extent of the disease they underwent either a re-excision or a completion mastectomy. Selected patients received standard adjuvant radiotherapy and chemotherapy. Patients with positive oestrogen/progesterone receptors received adjuvant endocrine therapy in accordance with their menopausal status. Patients with HER-2 positive tumours received standard Transtuzumab therapy.
After surgery patients were followed-up after 1 and 3 months then every 6 months for clinical review for the first 2 years and every year subsequently. Surveillance included clinical examination and yearly mammography.
Statistical analysis
Statistical analysis was performed using SPSS version 24 (SPSS Inc., Chicago, IL, USA). Descriptive statistics were performed for continuous and categorical variables. Qualitative variables were assessed using chi square test. Cox proportional hazards model was used for to determine the odds ratio and significance of potential variable. All statistical tests were two sided, with a 95% confidence interval and a p value <0.05 was considered as statistically significant.
Results
Study population
A total of 456 patients were included in the final analysis. Clinical and pathological characteristics of the patients are summarised in Tables 1 & 2. The patients are classified into two groups namely “No Re-Excision” or “Re-Excision”. Median age was 65 in both the groups. Majority of the tumour were invasive ductal carcinoma, Grade-2, under 20 mm in size with oestrogen (ER) and progesterone receptors (PR) positive. 12.19% of patients in the Re-Excision group had tumour >50 mm, whereas in the No Re-Excision group there were no tumours >50 mm. Ductal carcinoma insitu (DCIS) was present in 86% of the No Re-excision and 75.6% of Re-excision specimens (Tables 1 & 2).
415 (91%) patients had clear margins and did not require further re-excision. 91.6% of those patients who had clear margins had a closest peripheral excision margin of >5 mm (Table 1). 41 (9%) patients underwent further re-excision. 61% of them had a positive margin and 39% had a close margin. 31 (75.6%) of the patients who underwent re-excisions had DCIS involving the margins, when compared to 21 (51%) cases involving invasive carcinoma. Further surgeries include re-excision, completion total mastectomy or re-excision followed by completion mastectomy. Overall the re-excision rate was 9%. Among the re-excision group 48.78% had re-excision only, 46.34% has a completion mastectomy only, and 4.88% has a re-excision followed by a completion mastectomy. 70.73% of those who had re-excision had residual carcinoma in the specimen (Tables 2).
Lymph node status
In the No Re-Excision group, 93% had a sentinel lymph node biopsy (SLNB) and 18.39% of those biopsies had macrometastasis. Those patients with macrometastasis underwent ALND. 44.57% of patients who underwent ALND had further positive lymph nodes.
In the Re-Excision group, 87.8% underwent a SLNB, of which 33.33% had macrometastasis. 41.46% underwent ALND and 47% of those patients were found to have further malignancy in the lymph nodes.
Tumour characteristics and risk factors related to positive and close margins
There were several factors associated with the presence of positive margins. Patients under the age of 50 (odds ratio (OR) of 1.84, 95% CI 0.728-4.651, p = 0.197), tumour size of >50 mm (OR of 111.13, 95% CI 5.983-2027, p = 0.001), grade 2 and 3 tumours (OR of 1.216 and 1.174 respectively), and HER-2 amplified tumours (OR 2.834, 95% CI 1.155-6.953, p = 0.022) showed an increased risk of positive or close margins (Table 3).
Lymph node factors which were associated with an increased risk of positive margin were SLNB positivity (OR 1.99, 95% CI 1.034-3.849, p = 0.039), undergoing ALND (OR 2.073, 95% CI 1.123-3.825, p = 0.019), and positive nodes in the ALND (OR 2.476, 95% CI 1.066-5.753, p = 0.035) (Table 3).
Discussion
In this study all patients underwent BCS using a sector resection and patients with a positive or close margin on histopathology report underwent further re-excision. The highlight of this study was a low rate of re-excision for positive or close margins. The re-excision rate for positive or close margin in BCS patients after sector resection was approximately 9%, which is significantly lower than the published rates in the literature [1,4–8,12,13,15,16,18–24]. This study also found that higher proportion of resection margins were involved with DCIS than invasive carcinoma. This highlights the fact that up to 23% of DCIS can be mammographically occult [32].
When compared to a standard wide local excision technique, in a sector resection slightly more volume of breast tissue is excised [9,25,27]. Although the principle of taking a wider margin is still debatable, this study found that the sector resection technique has shown to decrease the rate of positive margin and the need for re-excision. Excising a wider margin during a sector resection can be compared to the concept of cavity shave margin. Several studies have reported that performing routine cavity shave margin decreases the rate of positive margin and re-operation [6,18]. Excising additional breast tissue volume and its risk of poor cosmetic outcome has been a concern among several authors [25–27]. However, randomised controlled trials and meta-analysis have shown that the excision of additional breast tissue volume after the cavity shave margin did not result in poor cosmetic outcome [6,18,30,31]. In this study we found that among the patients who underwent BCS using a sector resection 91.6% had greater than 5 mm peripheral margins. Similar to the cavity shave margin, performing a slightly wider excision in the sector resection technique has been shown to have good oncological outcome without compromising the cosmetic outcome [9,10,24–26]. The re-excision rate in this study was lower than the published randomised controlled trials for cavity shave margin (9% vs 17%) [6,18]. This rate is also lower than the other published Australian studies for impalpable lesions (9% vs 30%) [16].
There have been several reported predictive risk factors for positive margins. Studies have shown that younger patients and tumour characteristics such as larger tumour, HER-2 positive, triple negative and lymphovascular invasion have increased risk for having a positive margin. This shows that aggressive tumour biology is an important risk factor irrespective of the surgery performed (BCS or total mastectomy) [7,29,30]. This study found similar risk factors for positive margin.
Conclusion
In conclusion, this 5-year retrospective study found that the younger patients and aggressive tumour biology are important predictive risk factors for positive margins. BCS using the sector resection technique resulted in a significantly lower rate of re-excision for positive or close margins without compromising the cosmetic or the oncological outcome. This can avoid unwanted psychosocial distress and anxiety in patients who are already going through a stressful emotional journey managing their breast cancer.
Conflict of interest
None declared.