Approximately 10% to 15% of patients with stage I/II invasive breast cancer will develop a clinically isolated local recurrence. The standard management of an ipsilateral breast tumor recurrence following breast-conserving surgery and radiation is salvage mastectomy, while local excision and radiation are optimal treatment of a chest wall recurrence following initial mastectomy. Although there are few data regarding the efficacy of systemic therapy after isolated local relapse, chemotherapy and/or hormonal therapy should be considered for most patients because of the high risk of subsequent distant relapse. However, local relapse does not always herald distant metastases. A prolonged interval between initial treatment and local recurrence is the most important prognostic factor for subsequent outcome, and when combined with other favorable characteristics, can predict 5-year survival rates of 70% or higher. [ONCOLOGY 14(11):1561-1581, 2000]
The isolated local recurrence in a patient previously treated for early-stage invasive breast cancer presents a unique challenge to the oncologist. The management of each patient requires a multidisciplinary approach that depends not only on factors specific to the recurrence itself but also on factors related to the original treatment. There is a paucity of clinical information, almost none prospective or randomized, to guide the clinician in choosing the optimal combination and sequence of surgery, radiation, and/or systemic therapy.
The clinical significance of an isolated local recurrence as a first event after treatment of early-stage invasive breast cancer, and its impact on survival, remains controversial. There is a strong association between local recurrence and the appearance of simultaneous or subsequent distant metastases. In many cases, local recurrence may be a manifestation of a more aggressive tumor biology that heralds the presence of distant metastases. Regardless of this association, durable local salvage is important in preventing the consequences of uncontrolled locoregional disease. However, if distant metastases are a common but not universal outcome after clinically isolated local recurrence, there may be a subgroup of patients for whom successful local salvage could result in long-term disease-free and overall survival.
The purpose of this review is to analyze the incidence and risk factors for local recurrence after initial treatment of invasive breast cancer. The risk factors for local recurrence will be compared for stage I/II breast cancer patients treated with breast-conserving surgery and radiation and stage I–IIIA breast cancer patients treated with mastectomy. Multidisciplinary management of an isolated chest wall recurrence after mastectomy and an ipsilateral breast tumor recurrence (IBTR) after breast-conserving surgery and radiation will be discussed separately. Regional recurrences and their management will not be addressed. Outcome and independent prognostic factors after salvage of a local recurrence will be reviewed, with particular attention given to the association between the clinically isolated local recurrence and subsequent distant metastases.
Approximately 10% to 20% of patients with stage I/II invasive breast cancer will develop an IBTR within 10 years of breast-conserving surgery and radiation.[1-14] However, a similar percentage of all patients (10%–20%) with stage I–IIIA invasive breast cancer will experience chest wall failure, with or without simultaneous regional failure, within 10 years of undergoing mastectomy.[1,3,8,15-19]
Table 1 shows data from six prospective randomized trials comparing breast-conserving surgery and radiation with mastectomy in stage I/II invasive breast cancer. Four of these trials report similar risks of local failure associated with these two methods of treatment for early-stage invasive breast cancer. The National Cancer Institute (NCI) and the European Organization for Research and Treatment of Cancer (EORTC) trials reported significantly higher rates of local failure for patients treated with breast-conserving surgery and radiation, compared to those treated with mastectomy. Inadequate surgery for the primary may have contributed to the higher rate of IBTR in these trials, since only gross removal of the tumor was required. For example, in the breast-conserving surgery arm of the EORTC trial, 81% had T2 tumors, and 48% of all patients had microscopically positive margins.
Clinical Risk Factors for Chest Wall Failure
• Young Age—Age groups of 35 years or less and 40 years or less have been associated with an increased risk of locoregional recurrence after mastectomy.[20-22] Lewis and Reinhoff reported a crude local recurrence rate of 67% for patients aged 20 to 29 years and 41% for patients aged 30 to 39 years in an early radical mastectomy series, whereas in women aged ³ 40 years, local failure rates were 21% to 25%. In another radical mastectomy series, Donegan et al observed a similar crude failure rate of 67% for ages 20 to 29 years and 46% for ages 20 to 39 years, compared with < 25% for those ³ 40 years of age.
Mathews et al reported that the crude rate of locoregional failure after mastectomy doubled with younger age; ie, from 6% to 7% for ages > 35 years to 12% for £ 35 years. However, in a more recent report from the same institution, the locoregional failure rate at 5 years was only 7.4% after modified radical mastectomy in 140 patients aged £ 35 years.
Studies using multivariate analysis to account for other known prognostic factors have shown that age may not be an independent predictor of locoregional recurrence.[16, 24-26] Recht et al demonstrated on multivariate analysis that the number of positive axillary nodes and total number of nodes examined—but not age—were significant independent factors for locoregional recurrence. Pisansky et al also used multivariate analysis to show that tumor size, nodal status, and estrogen-receptor (ER) status, not age, were significant independent factors for locoregional recurrence.
• Tumor Size—Patients with tumors ³ 5 cm have a 25% or higher risk of isolated locoregional recurrence after mastectomy with or without adjuvant systemic therapy.[16,18,21,25-30] Tumors < 5 cm are not associated with an increased risk of chest wall recurrence, and there is no significant distinction between T1 or T2 tumors in most studies.[1,18,25,26]
• Gross Multifocal/Multicentric Disease—Multifocal or multicentric disease does not increase the risk of local failure after treatment by mastectomy. In a series of 57 patients with gross multicentric disease treated by mastectomy, Fowble et al  reported a low (< 10%) risk of chest wall recurrence in the absence of ³ 4 positive nodes or T3 tumor size.
• Genetic Factors—Whether the presence of a BRCA1 or BRCA2 mutation increases the risk of chest wall failure following mastectomy is presently unknown.
Histopathologic Risk Factors for Chest Wall Failure
• Nodal Status—Patients with 4 or more positive axillary lymph nodes have a 25% or higher risk of developing isolated locoregional recurrence after mastectomy with or without adjuvant systemic therapy.[16,18,21,25-29] There is controversy regarding the risk of chest wall recurrence in the subgroup of patients with 1 to 3 positive nodes, with three randomized trials from the Danish Breast Cancer Cooperative Group (DBCG) and the British Columbia Cancer Agency recently reporting rates of 30% or more.[28,29,32] This rate is significantly higher than the £ 15% risk predicted from historical mastectomy series for patients with 1 to 3 positive nodes. The low number of axillary nodes—medians of 7[28, 29] and 11—in the dissection specimens in these trials may be an important factor in the high risk of locoregional recurrence.
According to a statistical model reported by Iyer et al, inaccuracy of the staging of a patient with ³ 4 positive nodes vs 1 to 3 positive nodes increases as fewer total nodes are removed. For example, the model predicts that in order to have a 90% probability of accurately ruling out 4 or more positive nodes, a patient with 1, 2, or 3 positive nodes and a T1 tumor size would need 8, 15, or 20 nodes examined, respectively. Viewed in another way, the odds that a patient with a T1 tumor and 1, 2, or 3 positive nodes but with only 7 nodes examined actually has 4 or more positive nodes are 13%, 55%, or 93%, respectively. The model suggests that these randomized trials found higher rates of locoregional recurrence in patients with 1 to 3 positive nodes because a significant number were understaged by the small median number of nodes removed.
There is also clinical support for this theory in the literature. Benson and Thorogood reported a prospective nonrandomized trial of total mastectomy with either an axillary dissection or axillary sampling. The locoregional recurrence rate at 5 years was 11.7% after axillary dissection, compared with 19.4% in those with only axillary sampling (P = .0019). Recht et al found that a greater number of nodes examined—from 2 to 5, 6 to 10, or ³ 11—was associated with a decreased risk of locoregional failure that was independent of the number of positive nodes on multivariate analysis. In another series of 404 mastectomy patients with T1-2 tumors and 1 to 3 positive nodes, Katz et al reported a locoregional recurrence risk of 24% with less than 10 nodes removed vs 11% for 10 or more nodes removed (P = .02).
These factors may account for the 30% or higher rates of locoregional recurrence in the Danish and British Columbia series that would be more usual for patients with 4 or more positive nodes. The largest reported series of patients with 1 to 3 positive nodes treated with mastectomy and adjuvant systemic chemotherapy has recently been updated by Recht et al. Among 983 patients with T1-2 tumors and 1 to 3 positive nodes, the 10-year cumulative incidence of local failure was only 8%.
• Extracapsular Extension—The presence of extracapsular nodal extension has been associated with an increased risk of locoregional recurrence following mastectomy.[27,36,37] However, the presence of extracapsular nodal extension is a risk factor for chest wall recurrence and is not associated with an increase in the risk of axillary failure.[37,38] Kuske et al reported that in the subgroup of patients with 1 to 3 positive nodes, extracapsular extension was associated with a 17% risk of chest wall failure, compared with a 7% risk among those without extracapsular extension (P = .08). However, nodal recurrence was rare with or without the presence of extracapsular extension. For patients with 4 or more positive nodes, extracapsular extension was not associated with a further increase beyond the already high rate of chest wall failure (22% vs 26%, P = .47). In other studies, extracapsular extension has been significantly related to the number of positive nodes, but was not an independent factor for chest wall or axillary recurrence following mastectomy.[39,40]
• Margin Status—There is a paucity of data regarding the importance of margins after mastectomy for determining the risk of subsequent local relapse. Deep invasion of the pectoral fascia has been associated with an increased risk of local recurrence after mastectomy with or without chemotherapy.[21,41] Freedman et al found a 28% risk of chest wall failure at 8 years after mastectomy with a margin £ 5 mm in women aged 50 years or less, who otherwise were considered at low risk for locoregional recurrence because of a tumor size < 5 cm and 0 to 3 positive nodes.
In a series of node-positive postmenopausal women treated by mastectomy and tamoxifen(Drug information on tamoxifen) (Nolvadex), Fisher et al found that a positive margin was a significant predictor of locoregional recurrence on multivariate analysis. However, a close or positive margin after mastectomy has not been associated with a high risk of chest wall failures in all series.[27,44,45] For example, in a series of T1-2 node-negative women treated by mastectomy, Ahlborn et al observed a crude rate of local recurrence of 6% with margins £ 4 mm compared with 3% with margins > 4 mm, but the difference was neither clinically nor statistically significant.
Mentzer et al reported outcomes from a series of patients with stage II disease (two-thirds with positive nodes) who were treated by modified radical mastectomy, with or without systemic therapy. Two-thirds of these patients were assessed using gross margins, which is less accurate than microscopic assessment, and one-quarter also received postmastectomy radiation, which minimizes the rate of chest wall failure with a close or positive margin. The crude local recurrence rate was nearly four times higher in patients with a margin of 5 mm or less compared to patients with margins greater than 5 mm (11% vs 3%), but this finding was not statistically significant.
In another series of 608 patients treated with mastectomy, with or without systemic therapy, and postmastectomy chest wall irradiation in 8%, Jager et al found no statistical difference in locoregional recurrence between 57 patients with close (< 5 mm) or positive margins and 551 patients with negative margins.
• Extensive Intraductal Component—Extensive intraductal component has been defined as intraductal carcinoma occupying more than 25% of the area encompassed by the invasive tumor and extending beyond the infiltrating edge of the tumor into surrounding breast tissue, or present in random sections of grossly unremarkable breast tissue. While an important prognostic factor for patients treated with breast-conserving surgery, the presence of an extensive intraductal component has not been associated with an increase in chest wall failure following mastectomy.
• High Grade—Few studies have addressed the significance of high histologic grade, but this factor has been associated with an increased risk of locoregional recurrence after mastectomy in some series.[49,50] O’Rourke et al reported on a series of 966 patients treated with mastectomy, without radiation or systemic therapy, for breast cancer tumors smaller than 5 cm in diameter. More than half of the patients were node positive. The risk of a chest wall recurrence was 16%, 21%, or 27% for grade I, II, or III tumors, respectively. These differences remained significant on multivariate analysis, as did lymph node status and lymphovascular invasion.
• Lymphovascular Invasion—There is a paucity of data regarding lymphovascular invasion and the subsequent risk of chest wall recurrence after mastectomy. In the aforementioned series of mastectomies for T1-2 tumors investigated by O’Rourke et al, there was a 36% risk of chest wall recurrence with lymphovascular invasion, compared to a 19% risk without lymphovascular invasion. This difference remained significant on multivariate analysis, as did lymph node status and tumor grade.
• Oncogenes and Tumor-Suppressor Genes—Most of the literature on HER2/neu and p53 expression discusses the impact of these genes on overall recurrence or survival, with less information available about their impact on local control. Pierce et al reported on a series of 107 patients with known HER2/neu overexpression who were treated with either breast-conserving surgery and radiation or mastectomy. In those treated with mastectomy, there was no difference in the subsequent chest wall failure rate, with or without overexpression. In contrast, Zellars et al found that p53-positive patients had a higher risk of local failure after mastectomy—with or without radiation—that remained significant on multivariate analysis.
• Estrogen-Receptor Status—Negative ER status, alone or in combination with negative progesterone(Drug information on progesterone) receptors, has been associated with an increased risk of chest wall recurrence after mastectomy.[16,25,26,43]
• Adjuvant Systemic Therapy—Adjuvant systemic chemotherapy has been associated with a modest decrease in the risk of chest wall recurrence after mastectomy.[26,33,35,53,54] Bonadonna et al updated a randomized trial of CMF (cyclophosphamide [Cytoxan, Neosar], methotrexate(Drug information on methotrexate), fluorouracil(Drug information on fluorouracil)) vs observation in 386 node-positive women treated with radical mastectomy. Patients in the trial received no adjuvant radiation or endocrine therapy. At 20 years, there was no significant difference (15% vs 13%) in the rate of locoregional recurrence as a first event, with or without chemotherapy.
Tamoxifen alone or when added to chemotherapy also produces a modest reduction in the risk of chest wall recurrence.[26,33] Fisher et al reported the 10-year results from a trial by the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14 in which node-negative patients with ER-positive tumors were randomized to tamoxifen or observation. Of the 2,818 women entered into the trial, 62% were treated with mastectomy. The rate of local failures in the chest wall and scar as a first event decreased from 5% to 2% (P < .0001).
Goldhirsch et al reported a meta-analysis of five trials of adjuvant systemic therapy in node-positive patients treated by mastectomy without radiation. Approximately 2,108 patients received adjuvant systemic chemotherapy with (815) or without (1,293) tamoxifen or with oophorectomy (166), and 722 patients received no systemic therapy or only one cycle of chemotherapy. The 10-year cumulative incidence of local relapse as a first event was 8% vs 16%.