Local-Regional Breast Cancer Recurrence Following Mastectomy J. P. Crowe, Jr, MD; N. H. Gordon, PhD; A. R. and Participating Investigators
Antunez, MD; R. R. Shenk, MD; C. A. Hubay, MD; J. M. Shuck, MD, DSc;
\s=b\ Local-regional recurrence patterns were investigated in 1392 patients with breast cancer. Primary treatment for all patients included a mastectomy. Nine hundred seventeen patients had negative nodes and did not receive systemic therapy. Four hundred seventy-five patients had node metastases and were randomized to receive different combinations of chemoendocrine therapy. Follow-up ranged between 5 and 16 years. Two hundred thirty (25.8%) node-negative patients have had recurrences, with the initial recurrence being local-regional in 9.2%. Two hundred forty-two (50.9%) node-positive patients have had recurrences, with the initial recurrence being local-regional in 17.1%. Larger tumors and more extensive node involvement were associated with more first local-regional recurrences. The relative percent of first local-regional recurrence among patients in whom cancer recurred was similar for node-negative and node-positive patients (35.4% and 33.5%, respectively). In 63.6% of patients in whom cancer recurred, first local-regional recurrence were distant. Larger tumors, more extensive node involvement, and a shorter disease-free interval after mastectomy were associated with more rapid appearance of distant recurrence among these patients. (Arch Surg. 1991 ;126:429-432)
mastectomy Modified for local-regional optimal spite this, reported incidence radical
has been considered the breast cancer. De¬ the of local-regional failure after mastectomy varies from less than 5% to greater than 30%.w Patients with more extensive initial local-regional disease are thought to be at greatest risk for local-regional failure. Nu¬ merous studies have advocated the use of postmastectomy radiation therapy to decrease local-regional recurrences among this group.34 Once a local-regional recurrence is de¬ tected, treatment recommendations vary widely and fre¬ quently include different combinations of surgical resection,56 external radiation therapy,7,8 and systemic hormone therapy and/or chemotherapy.9"" This study was undertaken to deter¬ mine the incidence of local-regional failure after modified radical mastectomy and to identify factors that might relate to prognosis after such recurrence. treatment
Accepted for publication November 17,1990. From the Departments of Surgery (Drs Crowe, Gordon, Shenk, Hubay, and Shuck) and Epidemiology and Biostatistics (Dr Gordon), Ireland Cancer Center (Dr Gordon), and Department of Radiation Therapy (Dr Antunez), Case Western Reserve University and University Hospitals of Cleveland, Ohio. Read before the 43rd Annual Cancer Symposium of the Society of Surgical Oncology, Washington, DC, May 20,1990, and the American Society of Clinical Oncology 1990 Meeting, Washington, DC, May 22,1990. Reprint requests to the Department of Surgery, University Hospitals of Cleveland, 2074 Abington Rd, Cleveland, OH 44106 (Dr Crowe).
PATIENTS AND METHODS All patients in this study are part of prospective multi-institutional breast cancer trials based at Case Western Reserve University, Cleveland, Ohio. To be eligible for these trials, patients were re¬ quired to have operable breast cancer and to have undergone a modified radical mastectomy. Nine hundred seventeen patients were found to have negative axillary lymph nodes and were followed up without systemic therapy. Four hundred seventy-five patients were found to have positive axillary lymph nodes and were randomized to one of two chemoendocrine adjuvant therapy protocols that included combinations of cyclophosphamide (Cytoxan), methotrexate sodium, and fluorouracil (CMF); cyclophosphamide, methotrexate, fluoroura¬ cil, vincristine sulfate, and prednisone (CMFVP); and tamoxifen citrate (T). In the first trial, 311 axillary node-positive patients were randomly assigned to receive one of three adjuvant treatments: CMF (1 year), CMF and (1 year), or CMF and (1 year), with bacillus Calmette-Guerin (second year). This trial began in 1974, and patients were followed up for up to 16 years. In the second trial, 164 axillary node-positive patients with estrogen receptor-positive tumors were randomly assigned to receive either CMFVP (1 year) and (3 years) or alone (3 years). This trial began in 1980, and patients were followed up for up to 10 years. Premenopausal patients with estrogen receptor-positive tumors underwent a surgical oophorectomy. Axil¬ lary node-positive patients with estrogen receptor-negative tumors received CMFVP therapy (1 year). The results of both trials have been published recently. 13 Adjuvant external radiation therapy was not employed for either node-positive or node-negative patients. Estrogen receptor assays were performed on all primary tumor specimens, largely in one laboratory. Estrogen receptor values and various clinical and pathologic prognostic variables were entered into a computerized data base for subsequent analysis. All patients were followed up at regular 3-month intervals with physical examinations and blood chemistry determinations. A chest roentgenogram and bone scan were obtained at 6- and 12-month intervals, respectively. Recurrences were documented when detected, and their specific site(s) was recorded. In this analysis, first local-regional recurrences were defined as recurrence in the soft tissue of the chest wall (local), axillary nodes (regional), supraclavicular nodes (regional), or infraclavicular nodes (regional) on the mastectomy side at least 30 days before the development of a distant recurrence. Treatment following recurrence was determined by the patient's attending physician. All patients have been followed up to identify recurrence or to record death. Disease-free and overall survival estimates were compared with use of the log-rank test. Cox's proportional hazard model was used to determine the relationship to patient characteristics of dis¬ tant recurrence after a first local-regional recurrence.
RESULTS A total of 1392 patients with operable breast cancer were entered into this study. Nine hundred seventeen patients (65.9%) were found to have negative axillary nodes. Four
Downloaded From: http://archsurg.jamanetwork.com/ by a University of Arizona Health Sciences Library User on 05/27/2015
Table 1 .—Patient and Tumor Characteristics
Table 2.—Pattern of First Breast Cancer Recurrence
No. of Patients Node-
Variable
(n 475)
Menopausal status Premenopausal and perimenopausal Postmenopausal Estrogen receptor status Positive {>3 fmol/mg) Negative (3
0
243
Tumor diameter, 5
444
146
380
244
49
82
hundred
seventy-five (34.1%) were found to have positive axillary lymph nodes. Patient and tumor characteristics are shown in Table 1. Two hundred thirty-seven (25.8%) of the 917 node-negative patients have had recurrences. Eightyfour patients (9.2%) have had a first recurrence in a local (ipsilateral chest wall) and/or regional (ipsilateral axillary, supraclavicular, or infraclavicular nodes) site. One hundred fifty-three (16.7%) patients have had a first recurrence at a distant site. Two hundred forty-two (50.9%) of the nodepositive patients have had recurrences, with 81 (17.1%) being first local and/or regional and 161 (33.9%) being first distant. Initial recurrence patterns are presented in Table 2. For the entire group of 1392 patients, those with larger tumors (P .003) or more involved lymph nodes (P
Antunez, MD; R. R. Shenk, MD; C. A. Hubay, MD; J. M. Shuck, MD, DSc;
\s=b\ Local-regional recurrence patterns were investigated in 1392 patients with breast cancer. Primary treatment for all patients included a mastectomy. Nine hundred seventeen patients had negative nodes and did not receive systemic therapy. Four hundred seventy-five patients had node metastases and were randomized to receive different combinations of chemoendocrine therapy. Follow-up ranged between 5 and 16 years. Two hundred thirty (25.8%) node-negative patients have had recurrences, with the initial recurrence being local-regional in 9.2%. Two hundred forty-two (50.9%) node-positive patients have had recurrences, with the initial recurrence being local-regional in 17.1%. Larger tumors and more extensive node involvement were associated with more first local-regional recurrences. The relative percent of first local-regional recurrence among patients in whom cancer recurred was similar for node-negative and node-positive patients (35.4% and 33.5%, respectively). In 63.6% of patients in whom cancer recurred, first local-regional recurrence were distant. Larger tumors, more extensive node involvement, and a shorter disease-free interval after mastectomy were associated with more rapid appearance of distant recurrence among these patients. (Arch Surg. 1991 ;126:429-432)
mastectomy Modified for local-regional optimal spite this, reported incidence radical
has been considered the breast cancer. De¬ the of local-regional failure after mastectomy varies from less than 5% to greater than 30%.w Patients with more extensive initial local-regional disease are thought to be at greatest risk for local-regional failure. Nu¬ merous studies have advocated the use of postmastectomy radiation therapy to decrease local-regional recurrences among this group.34 Once a local-regional recurrence is de¬ tected, treatment recommendations vary widely and fre¬ quently include different combinations of surgical resection,56 external radiation therapy,7,8 and systemic hormone therapy and/or chemotherapy.9"" This study was undertaken to deter¬ mine the incidence of local-regional failure after modified radical mastectomy and to identify factors that might relate to prognosis after such recurrence. treatment
Accepted for publication November 17,1990. From the Departments of Surgery (Drs Crowe, Gordon, Shenk, Hubay, and Shuck) and Epidemiology and Biostatistics (Dr Gordon), Ireland Cancer Center (Dr Gordon), and Department of Radiation Therapy (Dr Antunez), Case Western Reserve University and University Hospitals of Cleveland, Ohio. Read before the 43rd Annual Cancer Symposium of the Society of Surgical Oncology, Washington, DC, May 20,1990, and the American Society of Clinical Oncology 1990 Meeting, Washington, DC, May 22,1990. Reprint requests to the Department of Surgery, University Hospitals of Cleveland, 2074 Abington Rd, Cleveland, OH 44106 (Dr Crowe).
PATIENTS AND METHODS All patients in this study are part of prospective multi-institutional breast cancer trials based at Case Western Reserve University, Cleveland, Ohio. To be eligible for these trials, patients were re¬ quired to have operable breast cancer and to have undergone a modified radical mastectomy. Nine hundred seventeen patients were found to have negative axillary lymph nodes and were followed up without systemic therapy. Four hundred seventy-five patients were found to have positive axillary lymph nodes and were randomized to one of two chemoendocrine adjuvant therapy protocols that included combinations of cyclophosphamide (Cytoxan), methotrexate sodium, and fluorouracil (CMF); cyclophosphamide, methotrexate, fluoroura¬ cil, vincristine sulfate, and prednisone (CMFVP); and tamoxifen citrate (T). In the first trial, 311 axillary node-positive patients were randomly assigned to receive one of three adjuvant treatments: CMF (1 year), CMF and (1 year), or CMF and (1 year), with bacillus Calmette-Guerin (second year). This trial began in 1974, and patients were followed up for up to 16 years. In the second trial, 164 axillary node-positive patients with estrogen receptor-positive tumors were randomly assigned to receive either CMFVP (1 year) and (3 years) or alone (3 years). This trial began in 1980, and patients were followed up for up to 10 years. Premenopausal patients with estrogen receptor-positive tumors underwent a surgical oophorectomy. Axil¬ lary node-positive patients with estrogen receptor-negative tumors received CMFVP therapy (1 year). The results of both trials have been published recently. 13 Adjuvant external radiation therapy was not employed for either node-positive or node-negative patients. Estrogen receptor assays were performed on all primary tumor specimens, largely in one laboratory. Estrogen receptor values and various clinical and pathologic prognostic variables were entered into a computerized data base for subsequent analysis. All patients were followed up at regular 3-month intervals with physical examinations and blood chemistry determinations. A chest roentgenogram and bone scan were obtained at 6- and 12-month intervals, respectively. Recurrences were documented when detected, and their specific site(s) was recorded. In this analysis, first local-regional recurrences were defined as recurrence in the soft tissue of the chest wall (local), axillary nodes (regional), supraclavicular nodes (regional), or infraclavicular nodes (regional) on the mastectomy side at least 30 days before the development of a distant recurrence. Treatment following recurrence was determined by the patient's attending physician. All patients have been followed up to identify recurrence or to record death. Disease-free and overall survival estimates were compared with use of the log-rank test. Cox's proportional hazard model was used to determine the relationship to patient characteristics of dis¬ tant recurrence after a first local-regional recurrence.
RESULTS A total of 1392 patients with operable breast cancer were entered into this study. Nine hundred seventeen patients (65.9%) were found to have negative axillary nodes. Four
Downloaded From: http://archsurg.jamanetwork.com/ by a University of Arizona Health Sciences Library User on 05/27/2015
Table 1 .—Patient and Tumor Characteristics
Table 2.—Pattern of First Breast Cancer Recurrence
No. of Patients Node-
Variable
(n 475)
Menopausal status Premenopausal and perimenopausal Postmenopausal Estrogen receptor status Positive {>3 fmol/mg) Negative (3
0
243
Tumor diameter, 5
444
146
380
244
49
82
hundred
seventy-five (34.1%) were found to have positive axillary lymph nodes. Patient and tumor characteristics are shown in Table 1. Two hundred thirty-seven (25.8%) of the 917 node-negative patients have had recurrences. Eightyfour patients (9.2%) have had a first recurrence in a local (ipsilateral chest wall) and/or regional (ipsilateral axillary, supraclavicular, or infraclavicular nodes) site. One hundred fifty-three (16.7%) patients have had a first recurrence at a distant site. Two hundred forty-two (50.9%) of the nodepositive patients have had recurrences, with 81 (17.1%) being first local and/or regional and 161 (33.9%) being first distant. Initial recurrence patterns are presented in Table 2. For the entire group of 1392 patients, those with larger tumors (P .003) or more involved lymph nodes (P
