Local Recurrence After Breast-Conserving Surgery and Radiation
Characteristics
Approximately 10% to 20% of patients with stage I/II invasive breast cancer will develop an IBTR at 10 years after breast-conserving surgery and radiation.[1-14] The median interval to IBTR is 3 to 4 years,[1,4,5,9,10,13,14,61,63,65,86,116-120] but this may be significantly prolonged to 5 to 7 years after chemotherapy and/or tamoxifen(Drug information on tamoxifen) therapy.[86]
At least two-thirds of breast recurrences are clinically detectable by physical examination with or without mammographic findings, and approximately one-third are detectable by mammography alone.[10,116-122] Philpotts et al[123] observed that, on mammogram, 81% of breast recurrences had a similar appearance to their initial tumors (eg, tumors initially presenting as masses without calcifications usually recurred as masses, and tumors initially presenting with calcifications recurred with calcifications).
More than 80% to 90% of breast recurrences are invasive,[60,116-119,122] and 60% to 85% of recurrences are of the same histologic subtype as the original primary.[4,59,60,120,123] Three-quarters of recurrences are clinically solitary[63,65,120] with an average size ranging from 1 to 2 cm.[4,59,60,123,124] Approximately 75% of patients present with an isolated recurrence in the breast only, 5% to 15% present with a clinically positive simultaneous regional nodal recurrence, and 5% to 15% present with simultaneous breast and distant metastases.[4,10,11,14,61,63,116,117,120,122,125]
The recurrence occurs in the same quadrant as the original primary tumor in approximately 50% to 90% of cases.[4,6,10,11,14,59-61,63,65,86,108,116-123] The distinction between an IBTR as a recurrence of the original tumor vs a new primary tumor arising in the breast is generally made on clinical grounds. Veronesi et al[59] considered 79% of cases as true local recurrences (ie, in the region of the original tumor) and 21% as new ipsilateral breast tumors (ie, in a separate quadrant from the original tumor).
Kurtz et al[11] also considered recurrences ³ 5 cm from the initial tumor site as new primary tumors. They reported that 32% of late failures more than 5 years after treatment were away from the primary tumor site (compared to 14% of earlier recurrences [P < .005]) and considered 64% of recurrences after 10 years and all recurrences after 15 years to be new primaries.
Recurrences elsewhere in the breast have been reported to occur at a later interval than in the original location in several other series as well.[118-120,126] Philpotts et al[123] observed that 91% of tumors recurring in the same quadrant were similar in mammographic appearance to their initial tumor, compared with only 25% of tumors recurring elsewhere (P < .02). Haffty et al[126] classified recurrences as new primaries if they occurred at a different site in the breast, had a different histology than the original tumor, or had discordant DNA flow cytometry.
The clinical significance of this distinction between a new primary vs IBTR is uncertain. Several series have found that survival decreased with local failures in the same location as the initial tumor (compared to elsewhere in the breast),[5,118,126] but others have not.[6,11,122] However, this may be related to the longer interval to failure for the recurrences elsewhere in the breast rather than the location itself.
Treatment
• Surgery—Mastectomy is the standard treatment of a clinically isolated IBTR following breast-conserving surgery and radiation.[112] At least 85% of patients will have operable disease at the time of detection of the local recurrence.[8,11,59,116,119,122,125] At the time of mastectomy, an exploration of the axilla with consideration of additional axillary dissection is advisable, since 31% to 58% of patients undergoing a dissection will have pathologically positive axillary nodes.[11,14,116,127] Salvage mastectomy for IBTR has resulted in local control rates of approximately 85% to 95%.[8,10,116,117,119,120,122,125]
The role of additional attempts at breast conservation remains investigational. Fowble et al[116] reported that only 42% of patients with an IBTR after breast-conserving surgery and radiation had no residual tumor at salvage mastectomy after a wide local excision, and half of those with residual disease had involvement of two or more quadrants. They could not identify characteristics that would permit prospective identification of an appropriate subgroup for wide local excision alone after IBTR.
Kurtz et al[127] reported a series of 50 patients with stage I or II breast cancer treated with breast-conserving surgery and radiation who subsequently underwent wide local excision for a clinically isolated IBTR, with or without axillary recurrence. Of the recurrences, 80% were less than 2 cm in size, 62% were in the vicinity of the original tumor, and all were without skin involvement. The second local failure rate in the salvaged breast was 38% at 5 years, with a 5- and 10-year survival of 67% and 42%, respectively. The only significant factors for local control on multivariate analysis were a disease-free interval greater than 5 years (92% vs 49%) and negative resection margins (73% vs 36%).
Salvadori et al[125] reported a second local failure rate of 19% at 5 years after reexcision, compared to 4% after mastectomy. However, the 5-year survival rate was 85% after reexcision, compared to 70% after mastectomy.
Abner et al[122] reported a study of 17 patients who refused a salvage mastectomy after IBTR; 11 had noninvasive and 6 had invasive tumors. Two of the 10 noninvasive tumors treated with excisional biopsy experienced a second local failure, and two others developed distant metastases and died. One patient with a noninvasive tumor treated by incisional biopsy had a second local failure treated with mastectomy. Of the six patients with invasive tumors, three had a second local relapse, and one other developed isolated distant metastases and died. Overall, 5 of 16 patients (31%) had a second local recurrence after wide excision for IBTR.
Stotter et al[10] reported local control in five of seven patients treated with local excision for breast only relapse. Voogd et al[118] reported that two of four patients treated for noninvasive recurrences with wide excision had a second local failure, compared to none of 21 patients treated with mastectomy. In the same series for invasive tumors, the recurrence rates were 38% for wide excision and 25% for mastectomy (P = .27).
• Radiation Therapy—There has been limited experience with salvage therapy of an IBTR following breast-conserving surgery and radiation by means of additional radiation. In the previously mentioned series of wide local excision for IBTR by Kurtz et al[127], 11 of 50 patients who had recurrences away from the original tumor bed were given additional radiation. Seven patients received a boost by electron beam to doses of 20 to 30 Gy, and four underwent an interstitial implant to a dose of 50 Gy. Second local failures occurred in 36% of patients given the supplemental irradiation, compared with 31% of those not given additional radiation.
Deutsch[128] reported on a series of 26 women treated for IBTR by repeat wide local excision and an electron boost of an additional 50 Gy. Of the 26, 16 also received tamoxifen after treatment for their IBTR. The subsequent breast recurrence rate was 19% at a follow-up of 7 to 139 months, without reported serious sequelae from the additional radiation.
Maulard et al[129] studied 15 patients treated for IBTR by second tumorectomy and brachytherapy (30-Gy single implant) and 23 patients treated for IBTR by brachytherapy alone (60 to 70 Gy in two implants). After an average of 40 months follow-up, the rate of second local failures was 21% and 5-year survival was 55%. The cosmetic result was good or acceptable in 16 patients, but there were serious complications from treatment in three patients, who required salvage mastectomy.
Jolicoeur et al[130] reported a series of 32 patients with isolated IBTR treated by repeat wide local excision and perioperative interstitial brachytherapy to a mean dose of 33 Gy (range: 29 to 50 Gy). The 5-year rates of second local relapse and overall survival were 27% and 80%, respectively. A good or acceptable cosmetic result was achieved in 75%, with only one case of skin necrosis requiring salvage mastectomy.
In summary, the risk of second local failure in these retrospective series using wide excision and radiation for salvage of an IBTR does not appear to be significantly different than it is in series using wide excision alone. Moreover, the combined doses of irradiation that result in high-dose regions receiving 90 Gy or higher may be a risk factor for radiation-related sarcoma, which has been previously reported to occur in the region of high-dose irradiation created by overlap at the junction of incorrectly matched fields.[131,132]
The high 5-year survivals in these retrospective series of reirradiation, which are attributable to selection of patients with favorable prognostic favors for survival after IBTR (ie, isolated breast relapse, location away from the original tumor, long interval to failure, or no skin involvement), make the avoidance of serious local complications from salvage therapy clinically relevant.
• Systemic Therapy—Given the absence of data in the setting of an IBTR, systemic therapy may be appropriately considered following salvage mastectomy based upon its efficacy in the adjuvant setting.[112] Fortin et al[5] found that antiestrogen therapy rather than chemotherapy improved 10-year survival following IBTR (72% vs 47%, P = .04).
Estrogen-receptor negative recurrences have been shown to have a worse prognosis than ER-positive recurrences.[121] Other retrospective series have been unable to show an improvement in survival with systemic therapy at the time of IBTR.[116, 122] The interval to IBTR, and therefore, the perceived magnitude of the risk of subsequent distant metastatic disease, may be useful in selecting patients for immediate systemic therapy rather than local therapy alone.
Prognostic Factors
Similar to the setting of chest wall failure, the interval to IBTR is an important prognostic factor following IBTR. Veronesi et al[59] reported that the risk of distant metastases within 1 year after an IBTR was 6.6 times the risk of patients developing an IBTR more than 3 years after surgery (P = .004).
Touboul et al[4] found that the only prognostic factor significant for survival after an isolated IBTR on multivariate analysis was an interval of 2 years or less vs more than 2 years (5-year survival: 38% vs 90%, P = .026). Other studies have also found that an interval of 2 years or less is an important predictor of survival,[5,6,133] but many series suggest that the interval to IBTR critical for determining prognosis may be longer.
Fourquet et al[14] reported a 5-year survival of 44% after breast recurrence with an interval to failure of 3 years or less, compared to 87% with an interval of more than 3 years (P < .01). Van Dongen et al[1] observed a 92% rate of distant metastases following isolated locoregional recurrence when the interval from primary treatment was less than 2 years, 53% when the interval was 2 to 5 years, and 22% when the interval was more than 5 years.
Haffty et al[9] reported a 50% rate of distant metastases with an interval to IBTR £ 4 years vs 17% for more than 4 years. Survival at 5 years after salvage of an IBTR was 50% for those with an interval to IBTR £ 4 years, compared to 78% for an interval greater than 4 years (P < .05). Furthermore, although the 10-year rate of distant metastases in their series was 36% vs 22% (P < .05) with and without IBTR, the rate was not significantly different between patients with an interval to IBTR greater than 4 years and those who did not develop an IBTR.
Kurtz et al[11] found the same 15-year survival between patients with a late interval to IBTR of 5 to 10 years and those who never had an IBTR. Approximately 30% to 40% of patients with an IBTR will have a favorable interval to recurrence of greater than 4 to 5 years.[1,5,6,9,11,116]
Initial tumor size also has prognostic importance following IBTR. Fowble et al[116] observed a 5-year survival of 62% for initial T2 tumors, compared with 95% for T1 tumors following salvage mastectomy for isolated IBTR (P = .03). Fortin et al[5] reported an improved 10-year post-IBTR survival of 59% for initial T1 tumors, compared with 16% for T2 tumors (P = .0009). Although the size of the recurrence was prognostic in one series,[118] the presence of invasion [118,120,122] or skin involvement[5,6,117,118] is of greater importance.
Distal carcinoma in situ or focally invasive recurrences have mastectomy salvage rates of over 90%.[118,120,122] Abner et al[122] reported no further relapses after salvage mastectomy for IBTR in 24 patients with noninvasive or minimally invasive IBTR, compared to 52% at 5 years in 99 patients with invasive IBTR (P = .001). Skin involvement has been reported in 3% to 22% of breast recurrences.[5,6,61,63,117,118,124]
In a large series reported by Gage et al,[124] only 8% of all breast recurrences were skin recurrences without associated parenchymal disease. Uncontrolled local failure was more common with a skin recurrence, compared to other recurrences (50% vs 14%, P = .0007). In addition, there was an increased risk of simultaneous (44% vs 5%, P < .0001) and sequential (60% vs 39%, P = .07) distant metastases with skin recurrences, so that their overall survival at 5 years was worse than that of women with other breast recurrences (34% vs 77%, P = .001).
Characteristics of Local Recurrence After Initial Mastectomy or Breast-Conserving Surgery and Radiation
Voogd et al[118] reported a series of 266 patients with IBTR after breast-conserving surgery and radiation. The prognostic factors for survival on multivariate analysis were skin involvement, initial tumor grade, and initial nodal status.
Van Tienhoven et al[108] studied 67 patients with first isolated locoregional recurrences after breast-conserving surgery and radiation, from the randomized trials vs mastectomy conducted by the EORTC and DBCG. On multivariate analysis, only initial pathologic nodal status was significant for survival after salvage treatment. The interval to first locoregional recurrence, and whether the recurrence was solitary and £ 3 cm, were the only prognostic factors for subsequent local control.
Local Recurrence After Mastectomy vs Breast-Conserving Surgery and Radiation
Survival After Salvage of a Local Recurrence Without Distant Metastases by Initial Treatment of Invasive Breast Cancer
A comparison of the characteristics of local failures following mastectomy or breast-conserving surgery and radiation is shown in Table 3. There is no clear association between the location of the recurrence on the chest wall and involvement of the mastectomy scar. However, the majority of IBTRs occur in the same quadrant as the original tumor, although the location is more likely to be elsewhere in the breast when intervals to recurrence are longer. While the median interval to chest wall failure is 2 to 3 years, the median interval to IBTR is 3 to 4 years and may be prolonged significantly by the use of adjuvant systemic therapy.
Simultaneous regional failures are uncommon after either treatment, but there generally is a higher incidence of simultaneous distant metastases with chest wall failure than with IBTR. In a randomized trial by the NCI, 60% of locoregional recurrences following mastectomy were associated with distant metastases, compared with only 5% following breast-conserving surgery and radiation.[8] In contrast, 38% and 32%, respectively, had locoregional recurrence and simultaneous distant metastases in the EORTC randomized trial.[1] From a comparison of retrospective series, the reported frequency of isolated local recurrence without distant metastases is between 10% and 40% following mastectomy[16,17,19,26,29] and between 5% and 15% following breast-conserving surgery and radiation.[4,10,11,14,61,63,116,117,120-122,125]
A comparison of overall survival following salvage of an IBTR or chest wall recurrence, with or without systemic therapy, is shown in Table 4. Following chest wall failure, survival ranges from 35% to 80% at 5 years and 25% to 60% at 10 years. Following IBTR, survival ranges from 45% to 80% at 5 years and 40% to 65% at 10 years. Local control following salvage treatment for chest wall failure ranges from 50% to 70%, and is generally higher—from 85% to 95%—for salvage mastectomy following IBTR. A comparison of prognostic factors for survival following chest wall recurrence vs IBTR is shown in Table 5.
Prognostic Factors for Survival After Local Recurrence Following Mastectomy or Breast-Conserving Surgery and Radiation
The interval to recurrence, a larger initial tumor size, and initial node-positive disease are unfavorable prognostic factors after local recurrence following either treatment. Local recurrence only, noninvasive histology of the recurrence, small size of the recurrence, and the absence of inflammatory symptoms are also favorable prognostic factors for both. Estrogen-receptor positivity and a gross total wide excision have been shown to be favorable prognostic factors following chest wall recurrence, while the absence of skin involvement is associated with a better prognosis following IBTR. Between 30% and 40% of patients with IBTR will have a favorable combination of prognostic factors, compared with only 20% to 25% of patients with chest wall failure.
The interval to local recurrence is the most important prognostic factor following either mastectomy or conservative surgery and radiation (Table 6). Most chest wall recurrences within 2 years are associated with distant metastases and a 5-year survival of 20% to 30%. The prognosis improves to 60% to 70% at 5 years for intervals longer than 2 years. While an IBTR within 2 years is also associated with a poor prognosis, the 5-year survival following salvage of an IBTR is 80% to 90% for intervals greater than 4 to 5 years.
Conclusions
Approximately 10% to 15% of patients with invasive breast cancer treated by mastectomy or breast-conserving surgery and radiation will have a clinically isolated local recurrence. Factors predictive of a significant risk (20% or higher) of local failure after either treatment are patient age £ 35 to 40 years, tumor size £ 5 cm, lymphovascular invasion, and close or positive resection margins. Factors increasing the risk of chest wall failure (but not IBTR) include 4 or more positive axillary nodes, negative estrogen receptors, and p53 positivity, while extracapsular extension, 1 to 3 positive nodes, and high histologic grade are variably associated with increased risk. Factors increasing the risk for IBTR following breast-conserving surgery and radiation include extensive intraductal component positivity and gross multifocal or multicentric disease.
Importance of Interval to Failure on 5-Year Survival Following Salvage of a Local Recurrence
Management of a clinically isolated chest wall failure following initial mastectomy involves wide local excision of all gross disease whenever possible, followed by postoperative chest wall and supraclavicular radiation in those not previously given postmastectomy radiation. The standard salvage of a clinically isolated IBTR is mastectomy. Systemic therapy should be considered in most patients following local failure, depending upon the perceived risk of subsequent distant relapse.
After either mastectomy or breast-conserving surgery and radiation, a long interval to the development of local failure, early stage of initial disease, and isolated local recurrence are important prognostic factors. A complete wide excision of disease and ER-positive recurrence are also favorable prognostic factors following chest wall failure, as is the absence of skin involvement following IBTR. In most cases, local failure within 2 years is a marker of more aggressive disease and simultaneous distant micrometastases. In contrast, the prognosis improves for the favorable subgroups of patients with late local failures, so that salvage treatment is associated with survival rates of 70% or higher at 5 years.
