Abstract
Background. To study the incidence of local recurrence (LR) of early breast cancer in the ipsilateral thoracic wall after mastectomy and outcome of patients with LR. Material and methods. A retrospective cohort study based on 2220 consecutive breast cancer patients treated at the Helsinki University Central Hospital, Finland, in 2000 to 2003. A subset of 755 (34.0%) patients had mastectomy which was usually followed by postoperative radiotherapy (51.2%) and/or systemic adjuvant therapy (79.2%). Results. During a median follow-up of 89 months, 22 (2.9%) patients treated with mastectomy had LR. The median time to LR was 27 months. None of the 12 patient- or tumour-related standard risk factors studied were independently associated with LR-free survival in a multivariate model. Six (27.3%) of the 22 patients with LR had distant metastases diagnosed either prior to or simultaneously with LR. The subset of 16 patients who were diagnosed with LR without concomitant distant recurrence had five-year breast cancer-specific survival of 77.5% as calculated from the date of LR detection, and overall survival of 59.2%. Conclusions. LR after mastectomy has become a rare event. Most women with isolated LR survive for five years after LR.
Local recurrence (LR) of breast cancer in the ipsilateral thoracic wall after mastectomy is considered to be associated with an aggressive clinical course and dismal survival [Citation1]. Yet, relatively favourable five-year overall survival rates have been reported in patient populations with isolated LR after mastectomy without concomitant systemic disease ranging from 43% to 66.4% [Citation2–4]. Such data may, however, be subject to biases including a selection bias and the publication bias. Few population-based data are available, and the exact locations of cancer recurrence after mastectomy and the details of local and systemic treatments given are rarely captured in regional and nationwide cancer registries. The rates of LR could thus be underestimated and the survival rates overestimated in the current literature.
Besides the skills of the surgical team, adjuvant treatments likely influence the LR rates substantially. According to the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) overview, radiation therapy reduces the five-year risk of LR after mastectomy and axillary lymph node dissection from 23% to 6% in node-positive patients and improves absolute survival 5.4% at 15 years after breast cancer detection, and radiotherapy reduces the LR rate from 6% to 2% in node-negative patients but does not influence survival [Citation5]. Similarly, systemic adjuvant endocrine therapy reduces substantially the risk of local recurrences in estrogen receptor (ER)-positive disease and adjuvant chemotherapy reduces the LR rate in women under 70 years of age [Citation6]. The mean absolute reduction in the five-year risk of LR was 20% among women who received systemic therapy as compared to those who did not [Citation6].
The risk of LR depends also on cancer size and its biology, the most important single factor probably being the axillary nodal status. Patients with axillary lymph node metastases have a greater risk for LR after mastectomy as compared with those without nodal metastases [Citation1,Citation7–9]. The EBCTCG overview reports as high as 23% risk of LR at five years after surgery in patients who did not receive postoperative radiotherapy which contrasts to the LR rates of 5.3–8% at five-year [Citation9,Citation10] and 6.5% at 10-year [Citation11] reported in some recent smaller series consisting of patients with node-positive breast cancer who did not receive postoperative radiotherapy. Other risk factors that are associated with an increased risk of LRs or locoregional recurrences after mastectomy include a high tumour histological grade [Citation1,Citation8,Citation12], absence of steroid hormone receptors [Citation1,Citation8], a large tumour size [Citation1,Citation8,Citation12] and young age at the time of breast cancer detection [Citation13].
The objective of the present study was to investigate the incidence of LR of early breast cancer after mastectomy and multidisciplinary treatment, the risk factors for LR and patient outcome after LR, since data from few large studies where modern adjuvant treatments have been administered are available. The study was conducted in a region where virtually all patients with early breast cancer are treated at a single university centre which allows population-based data to be obtained and reduces the risk of a selection bias.
Patients and methods
Patients
A total of 2220 new primary breast cancer patients were treated at the Breast Surgery Unit of the Helsinki University Central Hospital, Helsinki, Finland, between January 1, 2000 and December 31, 2003. Altogether 786 (35.4%) out of these 2220 patients had invasive breast cancer and underwent mastectomy and axillary surgery. Thirty-one patients were excluded from this cohort of 786 patients, since 23 (2.9%) had distant metastases at the time of surgery (M1 disease), six (0.8%) died from cardiovascular causes within one month of surgery and two (0.3%) patients were lost to follow-up. The remaining 755 patients (34.0% of the 2220 patients) form the basis of the current study. The patient and tumour characteristics are provided in . The median age at diagnosis was 61 years (range, 24–96 years). The study protocol was approved by the Ethics Committee of the Helsinki University Central Hospital.
Table I. Patient and tumour characteristics and the seven-year Kaplan-Meier estimate for local recurrence rate.
Surgery
All patients underwent either a sentinel node biopsy (SNB, n = 106), axillary lymph node dissection (n = 491) or both (n = 158). All patients who had a SNB underwent a back-up axillary lymph node dissection regardless of the findings at the SNB from January 2000 to May 2000. From June 2000 onwards, SNB was performed in patients who had clinically node negative, radiologically unifocal breast cancer with the largest tumour diameter ≤3 cm as evaluated with a breast ultrasound examination. In these patients axillary lymph node dissection was omitted whenever the SNB was negative. The median number of histologically examined axillary lymph nodes was three (range, 1–11) in patients who underwent a SNB only and 16 (range, 6–52) when axillary lymph node dissection with or without a SNB was carried out.
Histopathological evaluation
The mastectomy specimens as well as the SNB and axillary lymph node dissection specimens were assessed carefully histologically as described in a detail elsewhere [Citation14]. The axillary nodal stage was determined using the 7th edition of the TNM staging system [Citation15].
Radiation therapy
In general, postmastectomy radiotherapy was given to patients who had a large primary tumour (pT3 or pT4) and to patients who had axillary lymph node-positive cancer (). Radiotherapy was based on computer-based dose planning and was given with a linear accelerator with approximately 2 Gy daily fractions, 5 fractions per week. The cumulative dose to the thoracic wall and the regional lymphatics was approximately 50 Gy. The ipsilateral thoracic wall including the surgical scar was irradiated with electrons from an anterior field to minimise the doses delivered to the heart and the ipsilateral lung.
Table II. Radiotherapy and systemic adjuvant treatments given.
Systemic adjuvant therapy
Systemic adjuvant treatment was selected based on the patient and disease characteristics. In general, women with node-positive disease and those considered to have moderate-to-high risk node-negative cancer were treated with systemic adjuvant therapy (). Patients ≤ 65 years of age with moderate-to-high risk cancer, node-negative or node-positive, received adjuvant systemic chemotherapy. The chemotherapy regimens used usually included an anthracycline [usually epirubicin as a component of fluorouracil, epirubicin, cyclophosphamide (FEC)] or a taxane (usually docetaxel), or both, and usually consisted of a total of six cycles administered at three-week intervals. A few patients with HER2- positive breast cancer received trastuzumab and chemotherapy within the context of a clinical trial [Citation16]. Premenopausal women with oestrogen receptor ER and/or progesterone receptor (PR)-positive cancer received tamoxifen for five years, and postmenopausal women with hormone receptor-positive disease either tamoxifen or an aromatase inhibitor for five years. Hormonal therapy was initiated after chemotherapy.
Follow-up
Planned follow-up visits took place at one, three and five years after breast surgery. Whenever there was concern of cancer recurrence the patients had an access to an outpatient unit of the hospital which treats almost all (over 95%) breast cancer patients and their recurrences diagnosed in the region. The data on cancer recurrence and survival were collected from the hospital records and from the Finnish Cancer Registry that has a coverage approaching to 100% [Citation17].
Physical examination, blood cell counts and blood chemistry, and a bilateral mammogram with or without breast ultrasound examination were performed at the planned visits and whenever considered indicated. When a recurrence was suspected, an isotope bone scan and computed tomography were usually carried out as the initial staging examinations. After the first five years, follow-up was continued at a local health-care centre or at a private health-care company based on the patient preference.
Statistical analysis
LR was defined as any cancer recurrence in the ipsilateral thoracic wall, and locoregional recurrence as any cancer recurrence in the ipsilateral thoracic wall, in the regional lymphatics or in both. LR-free survival was calculated from the date of breast surgery to the date of first detection of LR censoring patients who did not have LR on the date of the last follow-up visit or the date of death. Breast cancer specific-survival (BCSS) was calculated from the date of surgery to the date of death considered to be caused by breast cancer censoring patients who were alive and those who died from an intercurrent cause on the date of death. Overall survival was calculated from the date of surgery to the date of death from any cause censoring patients who were alive on the date of the last follow-up visit. Life tables were constructed according to Kaplan-Meier method and the survival between the groups was compared with the log-rank test or the Cox proportional hazards model. P-values are two-tailed.
Results
Sites of breast cancer recurrence and survival
The median follow-up time after breast surgery was 89 months (range, 2–130 months). The seven-year breast cancer-specific survival was 85.9% and overall survival 74.5% in the cohort of 755 patients treated with mastectomy. A total of 114 (15.1%) patients died from breast cancer and 100 (13.2%) patients from an intercurrent or unknown cause ().
Table III. Events recorded.
Twenty-two 22 (2.9%) patients had a LR and 34 (4.5%) a locoregional recurrence during the follow-up. The seven-year LR-free survival was 96.9% and locoregional recurrence-free survival 95.4%. The median time to LR was 27 months (range, 1–87 months) and to locoregional recurrence 29 months (range, 2–130 months) as calculated from the date of surgery.
A recurrence in the ipsilateral axilla was detected in seven (0.9%) patients, two of these occurred concomitantly with a LR in the ipsilateral thoracic wall. The median time to recurrence in the ipsilateral axilla was 48 months (range, 13–124 months) from the date of breast surgery. Ipsilateral supraclavicular recurrence was found in eight (1.1%) patients, one of these concomitantly with a LR. The median time to a supraclavicular recurrence was 28 months (range, 1–59 months).
Contralateral breast cancer was detected in 35 (4.6%) patients and distant metastases in 133 (17.6%) patients. Eleven (50%) of the 22 patients with LR had distant metastases diagnosed either prior to LR (n = 2), concomitantly with LR (n = 4) or after LR (n = 5). Distant metastases as the first event (without a locoregional recurrence) occurred 113 (15.0%) patients.
Risk factors for local recurrence
Patients with PR-negative cancer had a 4.9% seven-year risk for LR as compared to a 1.8% risk in patients with PR-positive tumour in a univariate survival analysis (p = 0.015), and patients with ER−negative breast cancer tended to have a greater risk compared to patients with ER-positive cancer (6.0% vs. 2.4%, p = 0.059). Tumour biological groups formed by tumour ER and HER2 expression also tended to be associated with LR. Patients with either the ER−, HER2+ phenotype (9.5%) or the ER−, HER2− phenotype (4.9%) had a higher risk as compared to the ER+, HER2− (2.1%) and the ER+, HER2+ (2.3%) phenotypes (p = 0.056, ). When these three factors (PR, ER, the biological group) were entered into a Cox multivariate model as covariables, none of the factors was independently associated with LR-free survival. Age at diagnosis, tumour site in the breast, histological type or grade, axillary lymph node status or administration of systemic adjuvant treatment were not associated with the LR rate.
Survival after LR
The seven-year breast cancer-specific survival rate of patients without LR during the follow-up was 86.9% and that of the patients with LR 56.7% (p < 0.0001), and the seven-year overall survival figures were 75.7% and 45.5%, respectively (p = 0.001). The median follow-up time after LR was 38 months (range, 1–103 months). The five-year breast cancer-specific survival rate after LR as calculated from the date of LR detection was 54.8% and five-year overall survival 42.5%.
Sixteen (72.7%) of the 22 patients with ipsilateral chest wall recurrence did not have prior or concomitant distant recurrence. Five (31.3%) of these 16 patients developed distant metastases after detection of the LR during a median follow-up of 61 months (range, 1–103 months), and three died from breast cancer and three from an intercurrent cause during the follow-up. The five-year breast cancer-specific and overall survival rates of the 16 patients with isolated chest wall recurrence were 77.5% and 59.2%, respectively, as calculated from the date of detection of the recurrence. The six patients who had distant metastases diagnosed either prior to or simultaneously with LR died from breast cancer within one to 14 months detection of the chest wall recurrence.
Discussion
Ipsilateral chest wall recurrences were infrequent after mastectomy in this series of breast cancer patients who were frequently treated also with postoperative radiotherapy and systemic adjuvant therapies. This finding, based on a population-based series, is supported by several other studies [Citation1,Citation8,Citation9,Citation18]. The LR rate was low even in pN2 or pN3 disease (4.4%), which contrasts with the five-year LR risk of 12% in N2–3 patients who had postmastectomy RT and the five-year LR risk of 26% in N2–3 patients without RT reported in the EBCTCG overview [Citation5]. Modern radiation therapy and systemic adjuvant treatments reduce LR substantially [Citation5,Citation6], and many of the studies included in the EBCTCG overview are old and date back to times when these treatments were used less frequently, the radiation therapy techniques were suboptimal as considered from the present standards, and the endocrine and chemotherapies available were less effective.
Besides radiation therapy and adjuvant treatments, also quality of surgery and pathological assessment of the tissue specimens are likely of key importance. For example, in the Helsinki metropolitan area where the current patients were treated, the quality of surgery and pathology have improved after the year of 2000 due to centralisation of breast cancer surgery resulting in improved accuracy in cancer staging and a mastectomy technique where little or no breast tissue is left in the chest wall. During the first years of the 21st century, “remastectomy”, an operation carried out to remove both the LR and the residual breast tissue left behind in the primary operation performed in the 1990s, was not uncommon. At present, such surgery is rare in our unit.
In a univariate survival analysis a lack of progesterone receptors in the breast tumour was significantly associated with an increased risk for LR, and tumour ER-negativity and the ER−/HER2− and ER−/HER2+ biological types showed a similar tendency. Yet, none of these factors had independent influence on the risk in a multivariate analysis. Neither a poor histological grade of differentiation nor a high tumour MIB-1 cell proliferation index correlated with the risk of LR in the present study. These findings need to be viewed with some caution, since the number of LRs was small in the present series despite the relatively large size of the cohort, and multiple testing may also have a role. The small number of events coupled with administration of adjuvant treatments tailored to the patient risk profile may have prevented detection of some clinically important associations.
The current study was conducted in a region where virtually all breast cancer recurrences are captured.
Patient selection likely influences the rate of LRs observed. Young women have a higher rate of locoregional recurrence than older women. In a recent series from the MD Anderson hospital, young age was the only independent predictor of locoregional recurrence in patients who had pN0 or pN1 cancer when postmastectomy radiotherapy was not given [Citation13]. In the present study the seven-year LR rate was 6.5% in patients under 40 years of age at the time of breast cancer diagnosis compared to 2.5–3.3% in older patients. Postoperative radiation therapy reduces the risk of LR substantially also in women who are diagnosed with breast cancer at a young age. In our earlier study on patients younger than 35 the LR rate after mastectomy was 15% without radiotherapy and only 1% with postmastectomy radiotherapy [Citation19]. Similarly, Beadle et al. reported a 10-year LR rate of 12.5% after mastectomy in patients under the age of 35 treated without postoperative radiotherapy, whereas the LR rate was 7.0% in similar patients after radiotherapy [Citation20].
The outcome of patients with LR after mastectomy has been regarded sinister, and also in the present series LR was associated with a survival disadvantage. However, in the subset of patients with LR but without concomitant or prior distant recurrence survival was relatively favourable, the five-year breast cancer-specific and overall survival as calculated from the date of detection of the chest wall recurrence being 77.5% and 59.2%, respectively. In previous studies five-year overall survival after isolated LR is 43–44% in series dating back to the 1980s and 1990s [Citation2,Citation3]. A somewhat better five-year survival rate of 66.4% was reported in a study where the patients were treated from 1990 to 2005 [Citation4]. These and the current data suggest that the outcome of patients with isolated thoracic wall recurrence may be slowly improving, but this conclusion needs to be viewed with caution due to several confounding factors. These include improved imaging to detect distant recurrence, and it is not known whether the biological aggressiveness of the LRs that are not eradicated by modern adjuvant treatments is similar to the LRs that surface when solely local therapies are given.
Conclusions
We conclude that LR is a rare event after mastectomy carried out with appropriate techniques and when modern radiation therapy and adjuvant systemic treatments are frequently used. Most women with isolated chest wall recurrence survive for five years after local recurrence, but this group of patients still face a high risk for distant metastases.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
The study was supported by a grant from the Helsinki University Central Hospital Research Fund and also by a grant from Doris and Kurt Palander Foundation. With this statement all authors mentioned in this study disclose that there are no financial or personal relationships with other people or organisations that could inappropriately influence this work.
References
- Medina-Franco H, Vasconez L, Jobe Fix R, Heslin M, Beenken S, Bland K, . Factors associated with local recurrence after skin-sparing mastectomy and immediate breast reconstruction for invasive breast cancer. Ann Surg 2001;235:6:814–9.
- Chapgar A, Langstein H, Kronowitz S, Singletary E, Merrick R, Buchholz T, . Treatment and outcome of patients with chest wall recurrence after mastectomy and breast reconstruction. Am J Surg 2004;187:164–9.
- Nielsen H, Overgaard M, Grau C, Jensen A, Overgaard J. Loco-regional recurrence after mastectomy in high-risk breast cancer and prognosis. An analysis of patients from the DBCG 82 b&c randomization trials.Radiother Oncol 2006;79:147–55.
- Chen J, Ma X, Zhou W, Feng Y, Jiang G. Radiotherapy for and prognosis of breast cancer patients with local-regional recurrence after mastectomy. Chin J Cancer 2009;28:10.
- EBCTCG. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: An overview of the randomised trials. Lancet 2005;366:2087–106.
- EBCTG. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: An overview of the randomised trials. Lancet 2005;35: 1687–717.
- Taras A, Thorpe J, Morris A, Atwood M, Lowe K, Beatty J. Second Place Tie Residents´Competition: Irradiation effect after mastectomy on breast cancer recurrence in patients presenting with locally advanced disease. Am J Surg 2011; 201:603–7.
- Bijker N, Rutgers E, Peterse J, van Dongen J, Hart A, Borger J, . Low risk of locoregional recurrence of primary breast carcinoma after treatment with a modification of the halsted radical mastectomy and selective use of radiotherapy. Cancer 1999;85:8:1773–81.
- Gentilini O, Botteri E, Rotmensz N, Intra M, Gatti G, Silva L, . Is avoiding post-mastectomy radiotherapy justified for patients with four or more involved axillary nodes and endocrine-responsive tumours? Lessons from a series in a single institution. Ann Oncol 2007;18:1342–7.
- Cosar R, Uzal C, Tokatli F, Denizli B, Saynak M, Turan N, . Postmastectomy irradiation in breast in breast cancer patients with T1–2 and 1–3 positive axillary lymph nodes: Is there a role for radiation therapy?Radiat Oncol 2011;6:28.
- Botteri E, Gentilini O, Rotmensz N, Veronesi P, Ratini S, Fraga-Guedes C, . Mastectomy without radiotherapy: Outcome analysis after 10 years of follow-up in a single institution. Breast Cancer Res Treat Epub 2012 Apr 26.
- Yildirim E, Berberoglu U. Can a subgroup of node-negative breast carcinoma patients with T1–2 tumor who may benefit from postmastectomy radiotherapy be identified?Int J Radiat Oncol Biol Phys 2007;68:4:1024–9.
- Sharma R, Bedrosian I, Lucci A, Hwang R, Rourke L, Qiao W, . Present-day locoregional control in patients with T1 or T2 breast cancer with 0 and 1 to 3 positive lymph nodes after mastectomy without radiotherapy. Ann Surg Oncol 2010;17:2899–908.
- Leidenius MHK, Vironen JH, Heikkilä PS, Joensuu H. Influence of isolated tumor cells in sentinel nodes on outcome in small, node-negative (pT1N0M0) breast cancer. Ann Surg Oncol 2010;17:254–62.
- UICC International Union Against Cancer. In: Sobin LH, Gospodarowicz MK, Wittekind CH, editors. TNM classification of malignant tumours, 7th ed. New York: Wiley-Blackwell; 2009.
- Joensuu H, Kellokumpu-Lehtinen P, Bono P, Alanko T, Kataja V, Asola R, . Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. N Engl J Med 2006;354:809–20.
- Teppo L, Pukkala E, Saxén E. Multiple cancer – an epidemiologic exercise in Finland. J Natl Cancer Inst 1985;75: 207–17.
- Veronesi U, Cascinelli N, Mariani L, Greco M, Saccozzi R, Luini A, . Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med 2002;347:1227–32.
- Liukkonen S, Leidenius M, Saarto T, Sjöström-Mattson J. Breast cancer in very young women. Eur J Surg 2011;37:1030–7.
- Beadle B, Woodward W, Tucker S, Outlaw E, Allen P, Oh J, . Ten-year recurrence rates in young women with breast cancer by locoregional treatment approach. Int J Radiat Oncol Biol Phys 2009;73:734–44.