B number of patients who have received conventional

FULL THICKNESS CHEST WALL RESECTION FOR RECURRENT BREAST CARCINOMA INVOLVING THE BONY CHEST WALL JATIN P. SHAH, MD, AND JEROME A. URBAN, MD Solitary recurrent breast cancer involving the bony chest wall (parasternal recurrence) is curable when no evidence of systemic spread is found. Radical full thickness chest wall resection is an effective mode of treatment; it remains the only available treatment for patients whose chest wall recurrence fails to respond to radiation therapy, and in those who develop complications due to heavy irradiation. This report covers a review of the records of 52 patients treated at Memorial Hospital by chest wall resection for recurrent breast cancer between 1950 and 1972. The technique of chest wall resection and immediate plastic reconstruction is described. The gross and determinate 5-year survival rates for patients who underwent chest wall resection as the initial mode of therapy for chest wall recurrence were 43% and 57%, respectively. On the other hand, when chest wall resection was performed on those patients whose chest wall recurrences failed to respond to radiation therapy, the gross and determinate 5-year survival rates dropped to 16% and 19%. Full thickness chest wall resection with immediate plastic reconstruction when employed as the first mode of therapy for chest wall recurrences provides a significant 5-year survival rate, and has a definite place in the management of recurrent breast cancer. CaClr 35:567-573, 1975. REAST CANCER RECURS LOCALLY IN A SMALL B number of patients who have received conventional surgical treatment with or without radiation therapy for their primary operable tumors. The incidence of local recurrence at 10 years varies from 7-30% in different reported Most often the local recurrence on the chest wall represents one of several sites of recurrent or metastatic disease. However, in a significant number of patients the chest wall recurrence is the only evidence of recurrent disease. The recurrence may be only a skin nodule, a mass that is fixed to the underlying ribs and intercostal muscles, or a mass lying adjacent to the sternum, a so-called parasternal recurrence. Not infrequently the underlying bony chest wall is directly invaded by tumor. These solitary recurrences have been managed in a variety of ways in the past; no true prospective study has been reported evaluating the results of different modes of therapy in such a clinical setting..~ Presented at the 27th Annual Meeting of 'The James Ewing Society, Maui, HI, April 8-13, 1974. From the Breast Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY. Address for reprints: Dr. Shah, 1275 York Avenue, New York, NY 10021. Keceived for publication July 10, 1974. 5617 Snyder et al. made a serious attempt at comparing the results of two groups of patients with similar clinical presentations,' one treated by chest wall resection and the other treated by a variety of modes of therapy including radiation, ablation, hormones, and chemotherapy. With all the setbacks of a retrospective study, they reported a significantly higher salvage rate in the group of patients treated by chest wall resection. Further, the quality of survival of these patients was certainly superior in the group treated in this manner. As early as 1951, Urban reported the immediate results of major full thickness chest wall resection for recurrent mammary carcinoma and stressed the importance of including the internal mammary chain of lymph nodes in the initial extirpation of primary lesions presenting in the inner quadrants of the breasts.' This was based on the fact that the majority of patients with such chest wall-parasternal-recurrences had their primary lesions located in the inner half of the breasts. Since then a significant number of patients has been treated by chest wall resection either primarily as an extended radical mastectomy, which includes the internal mammary chain of lymph nodes, and is applied mainly to central and inner quadrant lesions; or secondarily for

568 CANCER March 1975 Vol. 35 excision of solitary chest wall or parasternal recurrences. Chest wall recurrences can be adequately controlled at times by appropriate modes of radiation therapy. However, when radiation therapy fails to control local disease, or when complications due to radiation therapy, such as necrosis of the costal cartilages or radiation-induced cancer, develop in the treatment field, then full thickness chest wall resection remains the only practical and effective mode of therapy. The purpose of this presentation is not to compare the relative efficacy of radiation vs. surgical resection in controlling recurrent breast cancer involving the bony chest wall. We are reporting the technique and results of full thickness chest wall resection performed in patients with chest wall or parasternal recurrences at Memorial Hospital. We have included only those patients in whom a full thickness chest wall resection was accomplished in order to remove disease which involved cartilages, ribs, sternum, or intercostal muscles. Locally recurrent skin nodules are not included. CLINICAL MATERIAL Between the years 1950-1972, 52 patients underwent chest wall resection at Memorial Hospital for recurrent mammary carcinoma involving the bony chest wall (Table 1). In 28 patients this was employed as the initial, and only mode of therapy when recurrence developed (Group I). In 19 patients the chest wall recurrence was initially treated by radiation therapy, which failed to control it; the resection was then performed in these radiation failures (Group 11). Three patients underwent chest wall resection for radionecrosis of the costal cartilages following extensive radiation therapy (Group 111). Two patients developed cancer in the treatment field following radiation therapy for recurrent breast carcinoma, and underwent chest wall resection for the secondarily developed, radiation-induced nonmammary cancers (Group IV). In all patients a thorough search TAIW 1. Reasons for Chest Wall Resection Done for Recurrent Breast Carcinoma in 52 Patients Resection done primarily for recurrence 28 Kesection for persistent disease after K. T. 19 (radiation failures) Resection for radionecrosis of chest wall; 3 no residual disease Resection for radiation-induced cancer 2 TOTAL 52 was made to rule out metastatic disease prior to chest wall resection. TECHSIQWE Under adequate general anesthesia, the skin incisions are marked out with liberal margins around the site of recurrent tumor mass or ulceration, and an en bloc excision of the full thickness of chest wall, including the parietal pleura, is carried out. After complete hemostasis is obtained the pleural cavity is drained with an underwater pleurevac drainage system. The full thickness bony defect is repaired with a series of heavy nylon sutures from the ribs to sternum to stabilize the chest wall and avoid paradoxical motion. The remaining defect is reconstructed with either ox fascia or marlex mesh. When the tumor involves the pericardium, it is excised en bloc with the rest of the chest wall. When there is a question of involvement of the sternum by tumor, smears from the marrow of sternum are sent for immediate examination, and if the presence of disease is confirmed, removal of the entire manubrium or lower sternum is advisable. Stabilization and repair of the defect is performed as described. Skin coverage is obtained in one of the three manners depicted (Figs. 1-6) utilizing a full thickness flap with subcutaneous tissue, either from the opposite breast or upper abdominal wall. Complete hemostasis, delicate handling of tissues, even distribution of tension on the flap while laying sutures, and adequate drainage with hemovac drainage tubes in the subcutaneous plane are all important factors in maintaining the viability of these flaps and obtaining satisfactory primary wound healing. RESULTS Since a majority of these patients had their initial surgery (mastectomy) performed elsewhere and came to us when recurrent disease developed, we are unable to report accurate details of the histologic type, as well as location in the breast, of the primary tumors in all patients, as well as information on lymph node involvement in the axilla. 'The results of the previously mentioned four groups of patients are presented separately. Group I This group contained patients with chest wall resection as the initial and only mode of treatment for recurrence (28 patients). Information

No. 3 CHEST WALL RESECTION IN BREAST CARCINOMA Shah and Urban 569 Fics. 1 (top) and 2 (bottom). Chest wall resection and reconstruction, utilizing opposite breast. as to the location of the primary tumor was not available in 10 patients. Of the remaining 18 patients, 12 had their primary tumors located in the inner quadrants of the breast, in 3 the tumor was centrally located, and in only 3 patients the primary tumor was situated in the outer half of the breast. No information was available regarding the status of axillary lymph nodes in 12 patients. Of the remaining 16, 8 had negative nodes, and in 8 axillary lymph nodes were known to contain metastatic disease. Nine of these patients had received routine postoperative radiation therapy. The time interval between initial mastectomy and chest wall recurrence ranged from 9 to 218 months, with an average of 63 months. The end results of this group of patients at 5 years are as follows. Twelve patients survived for more than 5 years after chest wall resection, 11 of whom were free of disease. Nine patients died of disease within 5 years, 4 died of causes unrelated to breast cancer while clinically free of disease, and 2 are lost to followup. The average survival time in patients who died of disease following chest wall resection was 25 months. Group I1 This group comprised patients who had chest wall resection for persistent disease following radiation therapy (19 patients). These 19 patients were initially treated by external radiation therapy for cure or control of chest wall recurrence. However, their recurrences failed to respond to

570 CANCER March 1975 Vol. 35 FIGS. 3 (top) and 4 (bottom). Chest wall resection and reconstruction utilizing upper abdominai flap based on the same side. radiation therapy, and they then underwent chest wall resection for persistent disease. Location of the primary tumors was unknown in 7 patients. From the remaining 12, 6 had their primary tumors located in the inner quadrants. In 3 the tumor was centrally placed; the remaining 3 had their primary tumor in the outer half of the breast. Information on the status of axillary lymph nodes was not available in 11 patients. At the time of mastectomy, 5 had negative nodes and 3 were known to have metastatic disease in the axillary lymph nodes. Nine of these patients had received routine postoperative radiation therapy following mastectomy. The disease-free interval ranged from 6 to 96 months, with an average of 44 months. The end results at 5 years are as follows. Three of these 19 patients were alive and well at 5 years. Thirteen patients were dead of disease within 5 years of chest wall resection. The average survival in patients who died of disease was 24 months after chest wall resection. Group I11 Included in this group were patients with chest wall resection for radionecrosis of costal cartilages (three patients). All three patients had their primary cancer treated by radical mastectomy, and all of them

No. 3 CHEST WALL RESECTION IN BREAST CARCINOMA Shah and Urban 57 1 FIGS. 5 (lop) and 6 (botfom). Chest wall resection and reconstruction utilizing upper abdominal flap based on the opposite side. had metastases to axillary lymph nodes. Two patients received routine postoperative radiation therapy. One patient had a disease-free interval of 103 months, at which time chest wall recurrence developed. Additional radiation therapy at this time resulted in radionecrosis, and chest wall resection was performed. She is alive and well 4% years later. The second patient developed recurrence at 7 months following mastectomy. Additional radiation therapy caused radionecrosis for which chest wall resection was performed. This patient has been lost to followup shortly after the operation. The operative specimen of both these patients was negative for malignant cells. The third patient initially presented with a large primary inoperable tumor with skin involvement and massive axillary lymph nodes. Radiation therapy controlled the tumor, and 4 years later she underwent mastectomy. Two years following mastectomy local recurrence in the operative scar developed. This was treated by additional radiation therapy which controlled the disease. Two years later skin ulceration developed, which gradually progressed to an extensive necrotic chest wall ulcer which extended to the pericardium. (Fig. 3). She underwent chest wall resection at this time, 11 years since her initial presentation, and she is now alive and well 1!h years after the operation. Group IV This group consists of patients who underwent chest wall resection for radiation-induced cancer in the field (two patients). The two

572 CASCER March 1975 Val. 35 patients in this group had initial radical mastectomies and both had positive axillary lymph nodes. Both received routine postoperative radiation therapy. Chest wall recurrence developed at 84 months in one and 108 months in the other. Both received additional radiation therapy for recurrent disease. The recurrent breast cancer responded, but both subsequently developed radiation-induced cancer in the treatment field. Chest wall resection was performed at this time. One patient was alive and well up to 12 months and then was lost to followup. The other patient developed a new primary cancer in the opposite breast 4 years later and eventually died of metastatic disease, 6 V2 years after chest wall resection. The results of treatment in the first two groups are summarized in Table 2. The gross 5-year survival and cure rates for patients treated by chest wall resection as the initial mode of therapy for recurrent disease fixed to the bony chest wall are 43% and 39%, respectively. However, the gross 5-year cure rate in patients with similar recurrence who failed to respond to radiation therapy and underwent chest wall resection later for persistent disease is only 16%. If, however, we consider only determinate cases eligible for a 5year analysis, excluding those who are living within 5 years, lost to followup, and those who were dead of other causes, the results are as follows. For chest wall resection performed as the initial mode of therapy, the 5- year determinate survival rate is 57% (12/21); the cure rate is 52% (11/21). When the chest wall resection was performed for radiation failures, the determinate 5-year survival and cure rate is 19% (3/16). The overall determinate 5-year survival combining these two groups of patients is 41% (15/37). DISCUSSION Solitary recurrent breast cancer arising in the chest wall (parasternal lesions) often develops as a direct extension from metastatic internal mammary lymph nodes and can present in the absence of systemic spread of cancer. These lesions are curable when no evidence of systemic spread is found, and should be treated aggressively by local therapy, viz. radical radiation therapy or radical surgical excision. Systemic therapy such as hormonal or chemotherapy should be reserved for patients with proven distant metastases, and should be used for locally recurrent carcinoma only when aggressive local therapy fails to control disease. It is well established that radiation therapy can control internal mammary nodal disease. However, surgical excision is equally effective and probably superior in controlling primary or recurrent disease in the internal mammary lymph nodes." Surgical excision with immediate plastic reconstruction remains the only available treatment for patients whose chest wall recurrence fails to respond to radiation therapy, and in those who develop extensive radionecrosis or radiation-induced cancer in the chest wall as a result of heavy irradiatione6 Radical surgery for locally recurrent carcinoma of the breast dates back to the later part of the 19th century. Sauerbruch, however, was the first to report a truly extensive chest wall resection for such locally recurrent breast cancer.' The procedure was later utilized by several au- thor~.~*~*~ We previously reported on a smaller series in 1951; however, our followup at that time was rather short, and the paper was devoted mainly to technique and indications. Rarely have there been reports of long-term followup of patients treated in this manner.'*' Generally, those patients whose disease-free interval between mastectomy and recurrence is long do better after chest wall resection.' What is more important is not only prolongation of life but also the quality of survival. To rid the patient of a fungating and foul-smelling ulcerated mass of the chest wall is certainly a most satisfactory mode of therapy. Urban reported over two decades ago that the cause of parasternal recurrences was the pres- TABLE 2. Results at 5 Years for Chest Wall Resection Only and Chest Wall Kesection for Radiation Failures Resection only (28) R.T. & resection (19) 'I-orn1. (47) NO. % NO. % No. 90 NED >5 yrs. 11 39 3 16 14 30 Survived with disease >5 yrs. 1 1 1 3 Living <5 yrs. 1 4 3 16 4 8 Dead of disease <5 yrs. 9 32 13 68 22 47 Lost to followup 2 7 2 4 Dead of other causes 4 14 4 8

No. 3 CHEST WALL RESECTION IN BREAST CARCINOMA Shah and Urban 573 ence of metastatic disease in internal mammary lymph nodes, and that therefore these nodes should be adequately treated in patients who present with primary lesions in the inner half of the breast, because of high risk of spread to these nodes. This fact has been borne out by several previous reports from Memorial Sloan-Kettering Cancer Center.''*'' Our current findings lend further support to the earlier reports. Accurate information regarding the location of the primary tumor was available in 30 patients from Groups I and I1 of this series. Eighteen of these 30 patients had their primary lesions located in the inner half of their breasts; and 6 others were centrally located (24/30 = 80%). Thus, to begin with, 80% of the patients in whom accurate information about the primary tumor was available were at a high risk of hav- ing metastatic disease to internal mammary lymph nodes. This presumption proved to be true when they presented with chest wall-parasternal recurrences. Major chest wall resection undertaken either primarily along with mastectomy to include internal mammary lymph nodes for inner quadrant or central lesions, or performed secondarily for chest wall recurrences (mainly parasternal) has a definite place in the overall management of operable breast cancer. Major full thickness chest wall resection remains the only available mode of therapy for recurrences not responding to radiation therapy and in patients developing radionecrosis of the chest wall or developing radiation-induced cancer in the irradiated field, particularly when these patients show no evidence of recurrent disease elsewhere. REFERENCES 1. Beardsley, J. M., and Cavanagh, C. R., Jr.: Radical excision of malignant chest wall tumors. J. Thorac. Cardiovarc. Surg. 29:582-596, 1955. 2. Donegan, W. L., Perez-Mesa, C. M., and Watson, F. R. : A biostatistical study of locally recurrent breast carcinoma. Surg. Gymcpl. Obsfcf. 122:529-540, 1966. 3. Maier, H. C.: Surgical management of large defects of the thoracic wall. Surgny 22:169-178, 1947. 4. Pawlias, K. T., Dockerty, M. D., and Ellis, F. H., Jr.: Late local recurrent carcinoma of the breast. Ann. Surg. 148:192-198, 1958. 5. Pickrell, K. L., Baker, M. M., and Collins, J. P.: Reconstructive surgery of the chest wall. Surg. Gynecol. Obrfcf. 84:465-476, 1947. 6. Pierce, G. W., Wiper, T., Magladry, G., Klabunde, E. H., Pennisi, V. R., and Fagella, R.: Reconstruction of a large defect on the entire thickness of the chest wall. Am. 3. Surg. 102:720-723, Nov. 1961. 7. Sauerbruch, F.: Beitrag zur Kesektion der Brustwand mit Plastik auf die Freigelegte Lunge. Dfrch. (. Chi. 86:275-280, 1906-07. 8. Snyder, A. F., Farrow, G. M., Masson, J. K., and Spencer, P. W.: Chest wall resection for locally recurrent breast cancer. Arch. Surg. 97 :246-253, 1968. 9. Urban, J. A.: Radical excision of the chest wall for mammary cancer. Canccr 4:1263-1285, 1951. 10. Urban, J. A., and Marjani, M. A.: Significance of internal mammary lymph node metastases in breast cancer. Am. J. Romfgenol Vol. 111:130-136, 1971. 11. Urban, J. A,, and Castro, E. B.: Selecting variations in extent of surgical procedure for breast cancer. Canccr 28: 1615-1623, I97 1. 12. Zimmerman, K. W., Montague, E. D., and Fletcher, G. H.: Frequency anatomical distribution and management of local recurrences after definitive therapy for breast cancer. Cancer 19~67-74, 1966.