- Email: [email protected]
Accelerated superfractionated radiotherapy for inflammatory breast carcinoma: complete response predicts outcome and allows for breast conservation
Int. J. Radiation Oncology Biol. Phys., Vol. 44, No. 2, pp. 289 –296, 1999 Copyright © 1999 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/99/$–see front matter
PII S0360-3016(99)00019-X
CLINICAL INVESTIGATION
Breast
ACCELERATED SUPERFRACTIONATED RADIOTHERAPY FOR INFLAMMATORY BREAST CARCINOMA: COMPLETE RESPONSE PREDICTS OUTCOME AND ALLOWS FOR BREAST CONSERVATION DOUGLAS W. ARTHUR, M.D., RUPERT K. SCHMIDT-ULLRICH, M.D., RICHARD B. FRIEDMAN, M.D., DAVID E. WAZER, M.D., LISA A. KACHNIC, M.D., CYRUS AMIR, PHD., HARRY D. BEAR, M.D., PH.D., MARY HELEN HACKNEY, M.D., THOMAS J. SMITH, M.D., AND WALTER LAWRENCE JR., M.D. Department of Radiation Oncology, Medical College of Virginia Hospitals of Virginia Commonwealth University (VCU) Richmond, VA Purpose: Chemotherapy and accelerated superfractionated radiotherapy were prospectively applied for inflammatory breast carcinoma with the intent of breast conservation. The efficacy, failure patterns, and patient tolerance utilizing this approach were analyzed. Methods and Materials: Between 1983 and 1996, 52 patients with inflammatory breast carcinoma presented to the Medical College of Virginia Hospitals of VCU and the New England Medical Center. Thirty-eight of these patients were jointly evaluated in multidisciplinary breast clinics and managed according to a defined prospectively applied treatment policy. Patients received induction chemotherapy, accelerated superfractionated radiotherapy, selected use of mastectomy, and concluded with additional chemotherapy. The majority were treated with 1.5 Gy twice daily to field arrangements covering the entire breast and regional lymphatics. An additional 18 –21 Gy was then delivered to the breast and clinically involved nodal regions. Total dose to clinically involved areas was 63– 66 Gy. Following chemoradiotherapy, patients were evaluated with physical examination, mammogram, and fine needle aspiration 3 3. Mastectomy was reserved for those patients with evidence of persistent or progressive disease in the involved breast. All patients received additional chemotherapy. Results: Median age was 51 years. Median follow-up was 23.9 months (6 – 86) months. The breast preservation rate at the time of last follow-up was 74%. The treated breast or chest wall as the first site of failure occurred in only 13%, and the ultimate local control rate with the selected use of mastectomy was 74%. Ten patients underwent mastectomy, 2 of which had pathologically negative specimens despite a clinically palpable residual mass. Response to chemotherapy was predictive of treatment outcome. Of the 15 patients achieving a complete response, 87% remain locoregionally controlled without the use of mastectomy. Five-year overall survival for complete responders was 68%. This is in contrast to the 14% 5-year overall survival observed with incomplete responders. The 5-year actuarial disease-free survival and overall survival for the entire patient cohort was 11% and 33%, respectively. All patients tolerated irradiation with limited acute effects, of which all were managed conservatively. Conclusion: Our experience demonstrates that induction chemotherapy, accelerated superfractionated radiotherapy, and the selected use of mastectomy results in excellent locoregional control rates, is well tolerated, and optimizes breast preservation. Based on our present results, we recommend that a patient’s response to induction chemotherapy guide the treatment approach used for locoregional disease, such that mastectomy be reserved for incomplete responders and avoided in those achieving a complete response. © 1999 Elsevier Science Inc. Inflammatory breast carcinoma, Chemotherapy, Mastectomy, Radiation therapy, Altered fractionation.
INTRODUCTION Inflammatory breast carcinoma (IBC) represents the most aggressive form of breast cancer. While significant improvements in the management of this disease have been made, aspects of locoregional management have remained controversial (1–20). Initial treatment approaches of mastectomy alone resulted in , 10% 5-year survival rates and high rates of local failure (21–23). With the use of more
effective, adriamycin-containing induction chemotherapy regimens, 5-year survival rates have improved dramatically, to 30 –50% (1, 3, 6, 9, 12, 15–19, 24 –30). These prolonged survival rates have placed renewed emphasis on the importance of locoregional control, with the involved breast and the draining lymphatics usually being the sites of most bulky disease. Consistent with this notion, recent publications on the management of high-risk breast carcinoma have
Reprint requests to: Douglas W. Arthur, M.D., Medical College of Virginia Hospitals of Virginia Commonwealth University, P.O. Box 980058, Richmond, VA 23298-0058. E-mail: dwarthur@ hsc.vcu.edu
Presented at the 80th Annual Meeting of the American Radium Society, Monte Carlo, Monaco, 1998. Accepted for publication 30 December 1998. 289
290
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 44, Number 2, 1999
Fig. 1. MCV/VCU decision tree for IBC.
documented the importance of locoregional treatment and its impact on patient survival (31, 32). As is evident from a number of reports, the critical issues relating to locoregional control of IBC center on the sequence of surgery and radiotherapy, once-daily versus twice-daily radiotherapy, and determining the role of mastectomy after highly effective, aggressive, accelerated superfractionated radiotherapy (1–20). Because of excellent pathological tumor response rates with induction chemotherapy (26) and aggressive radiotherapy (14), accelerated twice-daily irradiation of the breast and draining lymphatics was selected as the primary locoregional therapy, with mastectomy being reserved for patients with incomplete clinical response. In this report, we provide an update on an extended number of 38 patients, demonstrating that our treatment approach results in excellent locoregional tumor control and survival rates similar to series that include mastectomy as a part of the management of this disease.
signs of IBC (33). Pretreatment staging evaluation included detailed history and physical examination, evaluation of hematological parameters, liver function tests, alkaline phosphatase, bilateral mammograms, chest X-ray, and bone scan. Brain, chest, abdominal CT scans, skeletal survey, and bone marrow aspiration to rule out distant metastases were obtained if clinically indicated. Median age at presentation of the 38 patients was 51 years, ranging from 27 to 77 years. The patient characteristics with respect to extent of tumor in the breast- and lymph node-bearing areas are presented in Table 1. Only 6 (16%) of the patients presented with a tumor size , 5 cm; 23 patients (61%) were found to have tumors of 5–10 cm, and 9 (22%) with . 10-cm tumor size, 5 of which had disease extending beyond 15 cm. No detectable lymphadenopathy (N0) was found in 12 patients; 16 patients presented with N1, and 4 with N2 disease. No patient presented with distant metastases; however, in 6 patients, supraclavic-
METHODS AND MATERIALS
Table 1. Patient characteristics
Patients Between 1983 and 1996, 52 patients with IBC presented to the Medical College of Virginia Hospitals of VCU and the New England Medical Center. Of these, 38 were jointly evaluated in multidisciplinary breast clinics by surgical, medical, and radiation oncologists, and managed according to a defined prospectively applied treatment policy. All patients received induction chemotherapy, accelerated superfractionated radiotherapy, selected use of mastectomy when needed, and concluded with additional chemotherapy (Fig. 1). All patients had biopsy-proven poorly differentiated infiltrating ductal carcinoma and presented with the clinical
Characteristic Age (years) , 50 $ 50 Tumor size (cm) ,5 5–10 . 10 Nodal status N0 N1 N2 N3* * Representing supraclavicular lymphadenopathy.
Patient no. 21 17 6 23 9 12 16 4 6
Accelerated XRT for inflammatory breast cancer
ular lymphadenopathy was identified. Supraclavicular metastases were considered N3 prior to 1992 and were, therefore, included in this analysis (Table 1).
Chemotherapy All patients received induction chemotherapy. Chemotherapy dose schedules utilized have previously been reported (14, 24). In the majority of patients, this included 4 cycles of cyclophosphamide, doxorubicin, and 5-fluorouracil with occasional minor variations of the regimen depending on the medical oncologists involved. Following completion of locoregional treatment, cyclophosphamide, doxorubicin, and 5-fluorouracil was continued to a total doxorubicin dose of 450 mg/m2. This was followed by cyclophosphamide, methotrexate, and 5-fluorouracil. Responses to induction chemotherapy were assessed through clinical evaluation. Patients were considered complete responders if clinical evidence of disease resolved following induction chemotherapy. They were considered partial responders if , 50%, and nonresponders if . 50% of the initial tumor burden remained. In cases with disease progression during chemotherapy, radiotherapy was initiated. Most recently, in 4 patients, locoregional radiotherapy was followed by high-dose chemotherapy with autologous stem cell support (34, 35).
Radiotherapy After induction chemotherapy, radiotherapy was delivered using an accelerated superfractionated schedule (13, 14, 18). In the early utilization of this treatment policy, a 4 – 6-h interfraction interval was accepted, and in 7 patients, depending on the breast and tumor volumes treated, either 1.4 or 1.6 Gy fractions were used. The majority of patients were treated according to our present policy utilizing 1.5 Gy fractions delivered twice daily with a 6-h-interfraction interval. Irradiation was delivered with 60Co, 4 MV, or 6 MV linear accelerators. During the initial phase, the breast and ipsilateral supraclavicular, axillary, and internal mammary lymphatics were irradiated to 45 Gy over 3 weeks. Full dose to the involved dermal lymphatics was achieved by the use of a tissue equivalent bolus material placed over the entire breast. Immediately following, a boost dose was delivered to all sites clinically involved at presentation. Medial and lateral compression bridge tangents were used to deliver the boost to the breast and appositional electron fields to cover any nodal sites. Initially, once daily fractionation was used, delivering a boost dose of 20 Gy in 10 fractions to the entire breast. After 1989, based on the excellent tolerance of this regimen, twice-daily irradiation was continued, delivering a boost dose of 18 –21 Gy in 12–14 fractions of 1.5 Gy. Total radiation doses to clinically detectable disease amounted to 63– 66 Gy delivered during 4 –5 weeks. Boost doses to clinically positive lymph node regions were adjusted not to exceed brachial plexus tolerance of 60 Gy.
● D. W. ARTHUR et al.
291
Surgery The surgical management of the patient was in line with the prospective policy examining whether mastectomy added to locoregional control and/or survival in the patients with IBC. Therefore, modified radical mastectomy, after accelerated superfractionated radiotherapy, was reserved for those patients with persistent or progressive disease in the involved breast and who had not yet demonstrated evidence of distant metastases. Mastectomy was performed in 10 cases. This strategy was firmly adhered to based on pathological findings of 4 patients, who underwent mastectomy as part of the planned treatment approach in the initial phases of this study (14). The treatment policy included evaluation of each patient for residual disease in the treated breast 1 month after induction chemotherapy and accelerated superfractionated radiotherapy. This evaluation consisted of physical examination, mammogram, and 3 separate fine-needle aspirations of the tumor-bearing portion of the breast. Patients without clinical, mammographic, or pathologic evidence of residual disease avoided mastectomy, and received additional chemotherapy. Those patients with evidence of residual disease proceeded to mastectomy prior to additional chemotherapy; 6 underwent mastectomy due to evidence of residual disease in the breast 1 month after chemo-radiotherapy, and 4 underwent mastectomy as part of the planned treatment (14). Site of failure designation Failure analyses in this study were differentiated into locoregional, regional, and distant. Locoregional failures include all patients with tumor recurrence or tumor persistence in the radiation portals. Skin failures outside the original fields were considered extended locoregional. All other failures were considered distant metastases. Contralateral IBC, developing in two patients . 1 year after the first tumor were, for the purpose of this study, considered as distant failures. Statistical analyses In our analysis, Kaplan-Meier estimates of locoregional control, distant disease-free survival, disease-free survival, and overall survival functions were obtained for the entire patient group and in relation to induction chemotherapy response. Log-rank testing was applied to detect differences between survival curves. RESULTS Overall treatment outcome Follow-up and survival estimates. This report on 38 patients is based on patient analyses in April 1998. Actuarial rates of treatment outcome parameters and ultimate failure patterns were evaluated. The length of patient follow-up, to the date of last follow-up or death, for the entire patient cohort ranges from 6 to 86 months. Mean and median follow-up is 30.5 and 23.9 months, respectively. At the time
292
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 44, Number 2, 1999
Fig. 2. Patient cohort—Kaplan-Meier estimates. (222) Disease-free survival, (....) distant disease-free survival, and (- z - z - z ) overall survival.
of last follow-up, 12 patients remained alive, with a median follow-up of 34 months and a range of 14 – 86 months. For the entire group of 38 patients, Kaplan-Meier estimates of disease-free survival, distant disease-free survival, and overall survival are depicted in Fig. 2, and are comparable to published experiences on this disease entity. The 3and 5-year estimates of disease-free survival rates are 24% and 11%, respectively. This relatively low 5-year diseasefree survival is a result of the inevitable occurrence of distant metastases observed in the majority of these patients. Three- and 5-year actuarial overall survival rates for the patient cohort are 38% and 33%, respectively. However, only one third of those alive at 5 years were expected to be without evidence of disease. In 1989, a change in boosting technique from once-daily to twice-daily fractionation was made. No apparent difference on treatment outcome was observed; however, patient numbers are too small to draw any definitive conclusions. Patients presenting with supraclavicular lymphadenopathy experienced a varied treatment outcome. Although 2 patients died within 8 months of diagnosis, 1 survived 17 months, 2 survived 22 months, and 1 is still alive without evidence of disease 7 years after diagnosis. Locoregional control and breast conservation. Ultimately, 10 patients of the entire cohort experienced locoregional failures. Of these, 5 developed distant metastases as their first site of failure. Therefore, the ultimate locoregional control rate with the selected use of mastectomy was 74%. Only 13% recurred in the breast or chest wall as the site of
first failure. The overall breast conservation rate was 74%. However, as a result of in breast failures in addition to the selected use of mastectomy, the rate of locoregional control with breast conservation was reduced to 50%. Failure patterns. Despite aggressive chemotherapy and improved locoregional control, continued follow-up revealed a preponderance of distant metastases. Of the entire cohort, 26 patients failed first in a distant metastatic site, and only 7 of the 38 patients have remained disease-free. Of those that have remained disease-free, only 2 had a mastectomy. Of the 10 patients who underwent mastectomy, 1 has developed a chest wall failure, while 9 have experienced distant metastases. This data suggests that a prolonged disease-free status is difficult to achieve due to the high rate of distant metastases, and that distant failures occur despite the ability to control locoregional disease, and regardless of the use of mastectomy. Tumor recurrence in the treated nodal regions only occurred secondarily after documented failures in the breast or at distant sites. Additional failures included distant disease failure in the contralateral breast in 2 patients, and extended locoregional disease failure from tumor progression into dermal lymphatics outside the irradiation portals in 3 patients. Chemotherapy response versus overall treatment outcome Overall treatment outcome. Response to induction chemotherapy strongly correlated with treatment outcome. A clinical complete response was achieved in 15 patients. Of
Accelerated XRT for inflammatory breast cancer
● D. W. ARTHUR et al.
293
Table 2. Ultimate locoregional control (LRC) Characteristic
Total no.
LRC (%)
Complete responders No mastectomy Mastectomy Partial responders No mastectomy Mastectomy Non-responders No mastectomy Mastectomy Total pts no MRM Total patients
15 15 0 12 6 6 11 7 4 28 38
13 (87) 13 (87) — 10 (83) 4 (67) 6 (100) 5 (45) 2 (29) 3 (75) 19 (68) 28 (74)
the remaining patients, 12 experienced a partial response, and 11 had no response or disease progression (Table 2). Initial analyses revealed partial responders to have a similar disease-free, distant disease-free, and overall survival to that of nonresponders; therefore, these groups were compared together against complete responders when treatment outcome was evaluated. Survival estimates. In those patients experiencing a complete response to chemotherapy, estimated survival outcomes were significantly superior to those of incomplete responders (Fig. 3). For incomplete responders, the 3-year disease-free survival and distant disease-free survival rates were 14% and 15%, respectively, and for complete responders were both 38%. There was a striking difference in overall survival between these 2 groups. Complete responders had a 3- and 5-year actuarial overall survival of 68% that was significantly improved over that of the incomplete responders, 21% and14%, respectively (p 5 0.008). Locoregional control and breast conservation. The ultimate locoregional control rates are stratified by response to chemotherapy and are presented in Table 2. Locoregional control was significantly dependent on the clinical response to induction chemotherapy (p 5 0.03, Fig. 4). Of the complete responders, 13 of 15 patients (87%) remained locally controlled without a need for mastectomy. The remaining 2 patients failed in the breast, 1 with an unresectable, isolated recurrence, and the other in conjunction with distant metastatic disease. The median follow-up in this group of patients was 29 months. Amongst the partial responders, 6 of 12 patients were without evidence of disease following radiotherapy, continued with chemotherapy, and avoided mastectomy; 4 of these patients had lasting locoregional control. The remaining 6 partial responders proceeded to mastectomy and remained free of disease recurrence on the chest wall or in the regional lymphatics. Overall, there was an 83% locoregional control rate in partial responders with the selected use of mastectomy. Locoregional control was generally poor in those patients categorized as nonresponders, 45%. However, we found it notable that in 2 patients, locoregional control after radiotherapy was achieved with resultant breast conservation and
Fig. 3. Kaplan-Meier estimates of induction chemotherapy response versus (a) disease-free survival, (b) distant disease-free survival, and (c) overall survival. (- z - z - z ) Complete responders, (222) incomplete responders.
lasting locoregional control until their deaths from distant metastases 4 and 6 years later. Four of the nonresponders to chemotherapy proceeded to mastectomy, 3 of whom remained locoregionally controlled, and 1 failed on the chest wall. Tolerance The treatment schedule of induction chemotherapy followed by accelerated superfractionated radiotherapy was within the range of expected toxicities. Acute skin toxicity ranged from brisk erythema to confluent moist desquamation. Those patients undergoing twice-daily boosting tended toward patchy moist desquamation, as compared to less severe skin reactions in patients boosted once daily. Not unexpectedly, the treatment regimen resulted in significant
294
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 44, Number 2, 1999
Fig. 4. Induction chemotherapy response versus LRC. (- z - z - z ) Complete responders, (.....) partial responders, and (222) nonresponders.
progression of skin toxicity 1–2 weeks following radiotherapy completion; however, all patients healed within 1–2 months. Of those subjected to elective mastectomy for evidence of persistent disease, there were no wound healing complications documented. In those patients avoiding mastectomy and followed a minimum of 12–18 months, a range in cosmetic outcome was observed. Minimal to moderate fibrosis was seen in most patients. However, remarkable retraction was not observed, and there was universal preservation of acceptable symmetry when compared to the contralateral breast. Telangiectasias were observed to a varying degree and attributable to the amount of moist desquamation experienced. Those patients followed longer than 18 months did not demonstrate progression of late effects, and in some there was documentation of breast tissue softening. Severe late toxicity, including pain, breast retraction, and soft tissue necrosis requiring surgery was not documented. Two patients developed significant fibrosis within the treatment fields. Both of these patients began to show signs of generalized fibrosis within 6 –9 months after completing radiotherapy and progressed to significant fibrosis over the subsequent 6 months. DISCUSSION Inflammatory breast carcinoma is an uncommon presentation of breast cancer (36) and, because of its aggressive nature, has been the focus of intense clinical investigation. Multiagent chemotherapy has dramatically affected overall survival rates and has re-emphasized the importance of locoregional control to optimize quality of life and potentially contribute to the improvement of overall survival. In patients who are able to tolerate aggressive treatment in-
volving chemotherapy, radiotherapy, and possibly mastectomy, 5-year actuarial disease-free survival and locoregional control rates are approaching 40% and 85%, respectively (9). This review of 38 patients reports our experience with chemotherapy, accelerated superfractionated radiotherapy, and the selected use of mastectomy. We are able to demonstrate that with aggressive radiotherapy and appropriate decision criteria, breast conservation without compromise of locoregional control can be achieved. Because of the relative infrequency of IBC, most data on patient management are derived from retrospective reviews of institutional experiences, frequently conducted over a long time period. Amongst institutional experiences, two institutions have pioneered the multimodality management of IBC over the past 25 years (1, 3, 5, 6, 15–18, 24, 25). The most recent publications from each of these institutions (6, 17) both advocate a multimodality approach; however, one promotes the routine use of mastectomy while the other supports that mastectomy be reserved only as salvage. Fleming et al. (6) recently reported on the treatment results of 178 patients treated for IBC over a 20-year period. Based on their evaluation of the role of mastectomy, they determined that the addition of mastectomy to chemotherapy and radiotherapy resulted in a lower locoregional failure rate of 16.3% as compared to those patients managed without mastectomy, 35.7%. However, when stratified for response to induction chemotherapy, the improvement in locoregional control with mastectomy was only significant in patients with a partial response to chemotherapy. The added benefit of mastectomy was also present for disease-specific and disease-free survival when complete and partial responders were grouped together for the analysis. Disease-specific survival rates at 5 years for patients treated with and without
Accelerated XRT for inflammatory breast cancer
the addition of mastectomy were 62% and 43%, respectively, with the respective disease free survival rates of 52.6% and 31.2%. A comparison of our experience to that of other institutions is somewhat difficult. One needs to consider that a larger proportion of our patients achieved a complete response to induction chemotherapy and that our patients experienced a high rate of distant metastases and death when a complete response was not achieved. Nevertheless, in our group of 27 complete and partial responders (the selective use of mastectomy was utilized selectively in only 6 patients), occurrence of locoregional failure was only 15%, and compares favorably to reports where mastectomy was used routinely. In addition, our 5-year disease-free survival and overall survival for complete responders was 68% and 33%, respectively. In another large institutional experience, Thomas et al. (17) reported on the treatment results of 125 women with IBC utilizing their approach of alternating chemotherapy and radiotherapy. The reported 5-year disease-free and overall survival rates were 38% and 50%, respectively. Mastectomy was not routinely used, and therefore only 10 patients underwent mastectomy for local progression of disease. The reported complete response rate following this combined modality treatment was 82% with an ultimate local failure rate of 27%. Only 18% of the patients experienced a local recurrence as the first site of failure, and all patients with local failures developed distant metastases within 12 months. This compares favorably to our own experience showing a 26% ultimate locoregional failure rate. Although mastectomy appears to be of benefit upon review of the literature (2, 3, 5–13, 16, 18 –22, 24, 28 –30, 36), this benefit is not clearly evident in all patients. Reports describing treatment approaches that avoid mastectomy (4, 10), thus optimizing breast conservation, employ conventional radiotherapy treatment schedules, and as a result, compromise locoregional control rates. In contrast, the routine use of mastectomy circumvents the opportunity to preserve the breast, and the reported data does not clearly support that it is beneficial in all patients. Despite this, those reporting on treatment approaches proposed to improve treatment outcome recommend the routine use of mastectomy. Mastectomy has also been advocated despite the use of more aggressive radiotherapy to avoid severe late effects
● D. W. ARTHUR et al.
295
(9, 18). However, severe soft tissue fibrosis has not been a universal finding in our own experience or in that of others (1, 10, 14, 15, 17). Furthermore, when aggressive radiotherapy schedules have been utilized, once daily fractionation utilizing fractions larger than 1.8 –2.0 Gy (1, 15, 17) or accelerated superfractionation (14), local control rates are comparable to results reported from treatment approaches utilizing mastectomy. Our data suggest that the selected use of mastectomy optimizes both locoregional control and breast preservation. Based on our present results, we recommend that the response to induction chemotherapy can be used as a guide for the treatment of locoregional disease. In our reported experience, using twice-daily irradiation, the 87% rate of locoregional disease control in complete responders supports that mastectomy can be avoided in this group of patients. This control rate compares favorably with series utilizing mastectomy in their treatment policy. Both partial and nonresponders had a high rate of distant metastases in our study, and one could question the role of mastectomy in these groups. However, based on the contributory effect locoregional control potentially has on survival and the potential reduction in distant metastases that more aggressive chemotherapy regimens may have, we recommend mastectomy in those patients unable to achieve a complete response. The use of a clinical, pathologic, and radiographic evaluation following radiotherapy may add confidence when avoiding mastectomy in initially incomplete responders, but its value as a determinant of outcome requires additional study due to the small numbers in the subgroups of our experience. It is evident in the literature that following a complete response to chemotherapy, an aggressive once- or twicedaily treatment scheme is necessary to achieve high local control rates without the use of mastectomy. Our experience demonstrates that induction chemotherapy, accelerated superfractionated radiotherapy, and the selected use of mastectomy results in excellent locoregional control rates, is well tolerated, and optimizes breast preservation. Despite advances in locoregional management of this disease, systemic disease control continues to be a challenge. Presently, we are continuing with our local management approach and are evaluating the potential benefit of the addition of aggressive chemotherapy with autologous stem cell rescue as a part of the chemotherapy regimen.
REFERENCES 1. Arriagada R, Mouriesse H, Spielmann M, et al. Alternating radiotherapy and chemotherapy in non-metastatic inflammatory breast cancer. Int J Radiat Oncol Biol Phys 1990;19: 1207–1210. 2. Brun B, Otmezguine Y, Feuilhade F, et al. Treatment of inflammatory breast cancer with combination chemotherapy and mastectomy versus breast conservation. Cancer 1988;61: 1096 –1103. 3. Buzdar AU, Montague ED, Barker JL, et al. Management of inflammatory carcinoma of breast with combined modality approach - an update. Cancer 1981;47:2537–2542.
4. Chu AM, Wood WC, Doucette JA. Inflammatory breast carcinoma treated by radical radiotherapy. Cancer 1980;45: 2730 –2737. 5. Fastenberg NA, Buzdar AU, Montague ED, et al. Management of inflammatory carcinoma of the breast. Am J Clin Oncol 1985;8:134 –141. 6. Fleming RY, Asmar L, Buzdar AU, et al. Effectiveness of mastectomy by response to induction chemotherapy for control in inflammatory breast carcinoma. Ann Surg Oncol 1997; 4:452– 461. 7. Hagelberg RS, Jolly PC, Anderson RP. Role of surgery in the
296
8. 9. 10. 11. 12. 13. 14.
15.
16. 17.
18. 19. 20. 21. 22. 23.
I. J. Radiation Oncology
●
Biology
●
Physics
treatment of inflammatory breast carcinoma. Am J Surg 1984; 148:125–131. Knight CD, Martin JK, Welch JS, et al. Surgical considerations after chemotherapy and radiation therapy for inflammatory breast cancer. Surgery 1986;99:385–390. Kuske RR. Diagnosis and management of inflammatory breast cancer. Semin Radiat Oncol 1994;4:270 –282. Lamb CC, Eberlein TJ, Parker LM, et al. Results of radical radiotherapy for inflammatory breast cancer. Am J Surg 1991; 162:236 –242. Morris DM. Mastectomy in the management of patients with inflammatory breast cancer. J Surg Oncol 1983;23:255–258. Perez CA, Fields JN, Fracasso PM, et al. Management of locally advanced carcinoma of the breast. II. Inflammatory carcinoma. Cancer 1994;74:466 – 476. Pisansky TM, Schaid DJ, Loprinzi CL, et al. Inflammatory breast cancer: Integration of irradiation, surgery, and chemotherapy. Am J Clin Oncol 1992;15:376 –387. Radie-Keane K, Wazer DE, Schmidt-Ullrich R. Locoregional control of inflammatory carcinoma of the breast with accelerated superfractionated radiotherapy. Breast Dis 1994;7:59 – 68. Rouesse J, Friedman S, Sarrazin D, et al. Primary chemotherapy in the treatment of inflammatory breast carcinoma: A study of 230 cases from the Institut Gustave-Roussy. J Clin Oncol 1986;4:1765–1771. Singletary SE, Ames FC, Buzdar AU. Management of inflammatory breast cancer. World J Surg 1994;18:87–92. Thomas F, Arriagada R, Spielmann M, et al. Pattern of failure in patients with inflammatory breast cancer treated by alternating radiotherapy and chemotherapy. Cancer 1995;76: 2286 –2290. Thoms WW, McNeese MD, Fletcher GH, et al. Multimodality treatment for inflammatory breast cancer. Int J Radiat Oncol Biol Phys 1989;17:739 –745. Fields JN, Perez CA, Kuske RR, et al. Inflammatory carcinoma of the breast: Treatment results on 107 patients. Int J Radiat Oncol Biol Phys 1989;17:249 –255. Schafer P, Alberto P, Forni M, et al. Surgery as part of a combined modality approach for inflammatory breast carcinoma. Cancer 1987;59:1063–1067. Haagenson CD. Diseases of the breast. 2nd ed. Philadelphia: WB Saunders; 1971. p. 576 –584. Camp E. Inflammatory carcinoma of the breast. Am J Surg 1976;131:583–586. Treves N. The inoperability of inflammatory carcinoma of the breast. Surg Gynecol Obstet 1959;109:240 –242.
Volume 44, Number 2, 1999
24. Koh EH, Buzdar AU, Ames FC, et al. Inflammatory carcinoma of the breast: Results of a combined-modality approach – M.D. Anderson Cancer Center experience. Cancer Chemother Pharmacol 1990;27:94 –100. 25. Krutchik AN, Buzdar AU, Blumenschein GR, et al. Combined chemoimmunotherapy and radiation therapy of inflammatory breast carcinoma. J Surg Oncol 1979;11:325–332. 26. Sataloff DM, Mason BA, Prestipino AJ, et al. Pathologic response to induction chemotherapy in locally advanced carcinoma of the breast: A determinant of outcome. J Am Coll Surg 1995;180:297–304. 27. Maloisel F, Dufour P, Bergerat JP, et al. Results of initial doxorubicin, 5-fluorouracil, and cyclophosphamide combination chemotherapy for inflammatory carcinoma of the breast. Cancer 1990;65:851– 855. 28. Sherry MM, Johnson DH, Page DL, et al. Inflammatory carcinoma of the breast: Clinical review and summary of the Vanderbilt experience with multi-modality therapy. Am J Med 1985;79:355–364. 29. Donegan WL, Padrta B. Combined therapy for inflammatory breast cancer. Arch Surg 1990;125:578 –582. 30. Burton GV, Cox EB, Leight GS, et al. Inflammatory breast carcinoma: Effective multimodal approach. Arch Surg 1987; 122:1329 –1332. 31. Ragaz J, Jackson SM, Ne N, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 1997;337:956 –962. 32. Overgaard M, Hanson PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med 1997;337:949 –955. 33. Fleming ID, Cooper JS, Henson DE, et al. AJCC cancer staging manual. 5th ed. Philadelphia: Lippincott-Raven; 1997. p. 171–180. 34. Fields KK, Elfenbein GJ, Lazarus HM, et al. Maximum tolerated doses of ifosfamide, carboplatin, and etoposide given over six days followed by autologous stem cell rescue: Toxicity profile. J Clin Oncol 1995;13:323–332. 35. Fields KK, Zorsky PE, Hiemenz JW, et al. Ifosfamide, carboplatin, and etoposide: A new regimen with a broad spectrum of activity. J Clin Oncol 1994;12:544 –552. 36. Hortobaygi GN, Singletary SE, McNeese MD. Treatment of locally advanced and inflammatory breast cancer. In: Harris JR, Lippman ME, Morrow M, Hellman S, editors. Diseases of the breast. Philadelphia: Lippincott-Raven Publishers; 1996. p. 585–599.
PII S0360-3016(99)00019-X
CLINICAL INVESTIGATION
Breast
ACCELERATED SUPERFRACTIONATED RADIOTHERAPY FOR INFLAMMATORY BREAST CARCINOMA: COMPLETE RESPONSE PREDICTS OUTCOME AND ALLOWS FOR BREAST CONSERVATION DOUGLAS W. ARTHUR, M.D., RUPERT K. SCHMIDT-ULLRICH, M.D., RICHARD B. FRIEDMAN, M.D., DAVID E. WAZER, M.D., LISA A. KACHNIC, M.D., CYRUS AMIR, PHD., HARRY D. BEAR, M.D., PH.D., MARY HELEN HACKNEY, M.D., THOMAS J. SMITH, M.D., AND WALTER LAWRENCE JR., M.D. Department of Radiation Oncology, Medical College of Virginia Hospitals of Virginia Commonwealth University (VCU) Richmond, VA Purpose: Chemotherapy and accelerated superfractionated radiotherapy were prospectively applied for inflammatory breast carcinoma with the intent of breast conservation. The efficacy, failure patterns, and patient tolerance utilizing this approach were analyzed. Methods and Materials: Between 1983 and 1996, 52 patients with inflammatory breast carcinoma presented to the Medical College of Virginia Hospitals of VCU and the New England Medical Center. Thirty-eight of these patients were jointly evaluated in multidisciplinary breast clinics and managed according to a defined prospectively applied treatment policy. Patients received induction chemotherapy, accelerated superfractionated radiotherapy, selected use of mastectomy, and concluded with additional chemotherapy. The majority were treated with 1.5 Gy twice daily to field arrangements covering the entire breast and regional lymphatics. An additional 18 –21 Gy was then delivered to the breast and clinically involved nodal regions. Total dose to clinically involved areas was 63– 66 Gy. Following chemoradiotherapy, patients were evaluated with physical examination, mammogram, and fine needle aspiration 3 3. Mastectomy was reserved for those patients with evidence of persistent or progressive disease in the involved breast. All patients received additional chemotherapy. Results: Median age was 51 years. Median follow-up was 23.9 months (6 – 86) months. The breast preservation rate at the time of last follow-up was 74%. The treated breast or chest wall as the first site of failure occurred in only 13%, and the ultimate local control rate with the selected use of mastectomy was 74%. Ten patients underwent mastectomy, 2 of which had pathologically negative specimens despite a clinically palpable residual mass. Response to chemotherapy was predictive of treatment outcome. Of the 15 patients achieving a complete response, 87% remain locoregionally controlled without the use of mastectomy. Five-year overall survival for complete responders was 68%. This is in contrast to the 14% 5-year overall survival observed with incomplete responders. The 5-year actuarial disease-free survival and overall survival for the entire patient cohort was 11% and 33%, respectively. All patients tolerated irradiation with limited acute effects, of which all were managed conservatively. Conclusion: Our experience demonstrates that induction chemotherapy, accelerated superfractionated radiotherapy, and the selected use of mastectomy results in excellent locoregional control rates, is well tolerated, and optimizes breast preservation. Based on our present results, we recommend that a patient’s response to induction chemotherapy guide the treatment approach used for locoregional disease, such that mastectomy be reserved for incomplete responders and avoided in those achieving a complete response. © 1999 Elsevier Science Inc. Inflammatory breast carcinoma, Chemotherapy, Mastectomy, Radiation therapy, Altered fractionation.
INTRODUCTION Inflammatory breast carcinoma (IBC) represents the most aggressive form of breast cancer. While significant improvements in the management of this disease have been made, aspects of locoregional management have remained controversial (1–20). Initial treatment approaches of mastectomy alone resulted in , 10% 5-year survival rates and high rates of local failure (21–23). With the use of more
effective, adriamycin-containing induction chemotherapy regimens, 5-year survival rates have improved dramatically, to 30 –50% (1, 3, 6, 9, 12, 15–19, 24 –30). These prolonged survival rates have placed renewed emphasis on the importance of locoregional control, with the involved breast and the draining lymphatics usually being the sites of most bulky disease. Consistent with this notion, recent publications on the management of high-risk breast carcinoma have
Reprint requests to: Douglas W. Arthur, M.D., Medical College of Virginia Hospitals of Virginia Commonwealth University, P.O. Box 980058, Richmond, VA 23298-0058. E-mail: dwarthur@ hsc.vcu.edu
Presented at the 80th Annual Meeting of the American Radium Society, Monte Carlo, Monaco, 1998. Accepted for publication 30 December 1998. 289
290
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 44, Number 2, 1999
Fig. 1. MCV/VCU decision tree for IBC.
documented the importance of locoregional treatment and its impact on patient survival (31, 32). As is evident from a number of reports, the critical issues relating to locoregional control of IBC center on the sequence of surgery and radiotherapy, once-daily versus twice-daily radiotherapy, and determining the role of mastectomy after highly effective, aggressive, accelerated superfractionated radiotherapy (1–20). Because of excellent pathological tumor response rates with induction chemotherapy (26) and aggressive radiotherapy (14), accelerated twice-daily irradiation of the breast and draining lymphatics was selected as the primary locoregional therapy, with mastectomy being reserved for patients with incomplete clinical response. In this report, we provide an update on an extended number of 38 patients, demonstrating that our treatment approach results in excellent locoregional tumor control and survival rates similar to series that include mastectomy as a part of the management of this disease.
signs of IBC (33). Pretreatment staging evaluation included detailed history and physical examination, evaluation of hematological parameters, liver function tests, alkaline phosphatase, bilateral mammograms, chest X-ray, and bone scan. Brain, chest, abdominal CT scans, skeletal survey, and bone marrow aspiration to rule out distant metastases were obtained if clinically indicated. Median age at presentation of the 38 patients was 51 years, ranging from 27 to 77 years. The patient characteristics with respect to extent of tumor in the breast- and lymph node-bearing areas are presented in Table 1. Only 6 (16%) of the patients presented with a tumor size , 5 cm; 23 patients (61%) were found to have tumors of 5–10 cm, and 9 (22%) with . 10-cm tumor size, 5 of which had disease extending beyond 15 cm. No detectable lymphadenopathy (N0) was found in 12 patients; 16 patients presented with N1, and 4 with N2 disease. No patient presented with distant metastases; however, in 6 patients, supraclavic-
METHODS AND MATERIALS
Table 1. Patient characteristics
Patients Between 1983 and 1996, 52 patients with IBC presented to the Medical College of Virginia Hospitals of VCU and the New England Medical Center. Of these, 38 were jointly evaluated in multidisciplinary breast clinics by surgical, medical, and radiation oncologists, and managed according to a defined prospectively applied treatment policy. All patients received induction chemotherapy, accelerated superfractionated radiotherapy, selected use of mastectomy when needed, and concluded with additional chemotherapy (Fig. 1). All patients had biopsy-proven poorly differentiated infiltrating ductal carcinoma and presented with the clinical
Characteristic Age (years) , 50 $ 50 Tumor size (cm) ,5 5–10 . 10 Nodal status N0 N1 N2 N3* * Representing supraclavicular lymphadenopathy.
Patient no. 21 17 6 23 9 12 16 4 6
Accelerated XRT for inflammatory breast cancer
ular lymphadenopathy was identified. Supraclavicular metastases were considered N3 prior to 1992 and were, therefore, included in this analysis (Table 1).
Chemotherapy All patients received induction chemotherapy. Chemotherapy dose schedules utilized have previously been reported (14, 24). In the majority of patients, this included 4 cycles of cyclophosphamide, doxorubicin, and 5-fluorouracil with occasional minor variations of the regimen depending on the medical oncologists involved. Following completion of locoregional treatment, cyclophosphamide, doxorubicin, and 5-fluorouracil was continued to a total doxorubicin dose of 450 mg/m2. This was followed by cyclophosphamide, methotrexate, and 5-fluorouracil. Responses to induction chemotherapy were assessed through clinical evaluation. Patients were considered complete responders if clinical evidence of disease resolved following induction chemotherapy. They were considered partial responders if , 50%, and nonresponders if . 50% of the initial tumor burden remained. In cases with disease progression during chemotherapy, radiotherapy was initiated. Most recently, in 4 patients, locoregional radiotherapy was followed by high-dose chemotherapy with autologous stem cell support (34, 35).
Radiotherapy After induction chemotherapy, radiotherapy was delivered using an accelerated superfractionated schedule (13, 14, 18). In the early utilization of this treatment policy, a 4 – 6-h interfraction interval was accepted, and in 7 patients, depending on the breast and tumor volumes treated, either 1.4 or 1.6 Gy fractions were used. The majority of patients were treated according to our present policy utilizing 1.5 Gy fractions delivered twice daily with a 6-h-interfraction interval. Irradiation was delivered with 60Co, 4 MV, or 6 MV linear accelerators. During the initial phase, the breast and ipsilateral supraclavicular, axillary, and internal mammary lymphatics were irradiated to 45 Gy over 3 weeks. Full dose to the involved dermal lymphatics was achieved by the use of a tissue equivalent bolus material placed over the entire breast. Immediately following, a boost dose was delivered to all sites clinically involved at presentation. Medial and lateral compression bridge tangents were used to deliver the boost to the breast and appositional electron fields to cover any nodal sites. Initially, once daily fractionation was used, delivering a boost dose of 20 Gy in 10 fractions to the entire breast. After 1989, based on the excellent tolerance of this regimen, twice-daily irradiation was continued, delivering a boost dose of 18 –21 Gy in 12–14 fractions of 1.5 Gy. Total radiation doses to clinically detectable disease amounted to 63– 66 Gy delivered during 4 –5 weeks. Boost doses to clinically positive lymph node regions were adjusted not to exceed brachial plexus tolerance of 60 Gy.
● D. W. ARTHUR et al.
291
Surgery The surgical management of the patient was in line with the prospective policy examining whether mastectomy added to locoregional control and/or survival in the patients with IBC. Therefore, modified radical mastectomy, after accelerated superfractionated radiotherapy, was reserved for those patients with persistent or progressive disease in the involved breast and who had not yet demonstrated evidence of distant metastases. Mastectomy was performed in 10 cases. This strategy was firmly adhered to based on pathological findings of 4 patients, who underwent mastectomy as part of the planned treatment approach in the initial phases of this study (14). The treatment policy included evaluation of each patient for residual disease in the treated breast 1 month after induction chemotherapy and accelerated superfractionated radiotherapy. This evaluation consisted of physical examination, mammogram, and 3 separate fine-needle aspirations of the tumor-bearing portion of the breast. Patients without clinical, mammographic, or pathologic evidence of residual disease avoided mastectomy, and received additional chemotherapy. Those patients with evidence of residual disease proceeded to mastectomy prior to additional chemotherapy; 6 underwent mastectomy due to evidence of residual disease in the breast 1 month after chemo-radiotherapy, and 4 underwent mastectomy as part of the planned treatment (14). Site of failure designation Failure analyses in this study were differentiated into locoregional, regional, and distant. Locoregional failures include all patients with tumor recurrence or tumor persistence in the radiation portals. Skin failures outside the original fields were considered extended locoregional. All other failures were considered distant metastases. Contralateral IBC, developing in two patients . 1 year after the first tumor were, for the purpose of this study, considered as distant failures. Statistical analyses In our analysis, Kaplan-Meier estimates of locoregional control, distant disease-free survival, disease-free survival, and overall survival functions were obtained for the entire patient group and in relation to induction chemotherapy response. Log-rank testing was applied to detect differences between survival curves. RESULTS Overall treatment outcome Follow-up and survival estimates. This report on 38 patients is based on patient analyses in April 1998. Actuarial rates of treatment outcome parameters and ultimate failure patterns were evaluated. The length of patient follow-up, to the date of last follow-up or death, for the entire patient cohort ranges from 6 to 86 months. Mean and median follow-up is 30.5 and 23.9 months, respectively. At the time
292
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 44, Number 2, 1999
Fig. 2. Patient cohort—Kaplan-Meier estimates. (222) Disease-free survival, (....) distant disease-free survival, and (- z - z - z ) overall survival.
of last follow-up, 12 patients remained alive, with a median follow-up of 34 months and a range of 14 – 86 months. For the entire group of 38 patients, Kaplan-Meier estimates of disease-free survival, distant disease-free survival, and overall survival are depicted in Fig. 2, and are comparable to published experiences on this disease entity. The 3and 5-year estimates of disease-free survival rates are 24% and 11%, respectively. This relatively low 5-year diseasefree survival is a result of the inevitable occurrence of distant metastases observed in the majority of these patients. Three- and 5-year actuarial overall survival rates for the patient cohort are 38% and 33%, respectively. However, only one third of those alive at 5 years were expected to be without evidence of disease. In 1989, a change in boosting technique from once-daily to twice-daily fractionation was made. No apparent difference on treatment outcome was observed; however, patient numbers are too small to draw any definitive conclusions. Patients presenting with supraclavicular lymphadenopathy experienced a varied treatment outcome. Although 2 patients died within 8 months of diagnosis, 1 survived 17 months, 2 survived 22 months, and 1 is still alive without evidence of disease 7 years after diagnosis. Locoregional control and breast conservation. Ultimately, 10 patients of the entire cohort experienced locoregional failures. Of these, 5 developed distant metastases as their first site of failure. Therefore, the ultimate locoregional control rate with the selected use of mastectomy was 74%. Only 13% recurred in the breast or chest wall as the site of
first failure. The overall breast conservation rate was 74%. However, as a result of in breast failures in addition to the selected use of mastectomy, the rate of locoregional control with breast conservation was reduced to 50%. Failure patterns. Despite aggressive chemotherapy and improved locoregional control, continued follow-up revealed a preponderance of distant metastases. Of the entire cohort, 26 patients failed first in a distant metastatic site, and only 7 of the 38 patients have remained disease-free. Of those that have remained disease-free, only 2 had a mastectomy. Of the 10 patients who underwent mastectomy, 1 has developed a chest wall failure, while 9 have experienced distant metastases. This data suggests that a prolonged disease-free status is difficult to achieve due to the high rate of distant metastases, and that distant failures occur despite the ability to control locoregional disease, and regardless of the use of mastectomy. Tumor recurrence in the treated nodal regions only occurred secondarily after documented failures in the breast or at distant sites. Additional failures included distant disease failure in the contralateral breast in 2 patients, and extended locoregional disease failure from tumor progression into dermal lymphatics outside the irradiation portals in 3 patients. Chemotherapy response versus overall treatment outcome Overall treatment outcome. Response to induction chemotherapy strongly correlated with treatment outcome. A clinical complete response was achieved in 15 patients. Of
Accelerated XRT for inflammatory breast cancer
● D. W. ARTHUR et al.
293
Table 2. Ultimate locoregional control (LRC) Characteristic
Total no.
LRC (%)
Complete responders No mastectomy Mastectomy Partial responders No mastectomy Mastectomy Non-responders No mastectomy Mastectomy Total pts no MRM Total patients
15 15 0 12 6 6 11 7 4 28 38
13 (87) 13 (87) — 10 (83) 4 (67) 6 (100) 5 (45) 2 (29) 3 (75) 19 (68) 28 (74)
the remaining patients, 12 experienced a partial response, and 11 had no response or disease progression (Table 2). Initial analyses revealed partial responders to have a similar disease-free, distant disease-free, and overall survival to that of nonresponders; therefore, these groups were compared together against complete responders when treatment outcome was evaluated. Survival estimates. In those patients experiencing a complete response to chemotherapy, estimated survival outcomes were significantly superior to those of incomplete responders (Fig. 3). For incomplete responders, the 3-year disease-free survival and distant disease-free survival rates were 14% and 15%, respectively, and for complete responders were both 38%. There was a striking difference in overall survival between these 2 groups. Complete responders had a 3- and 5-year actuarial overall survival of 68% that was significantly improved over that of the incomplete responders, 21% and14%, respectively (p 5 0.008). Locoregional control and breast conservation. The ultimate locoregional control rates are stratified by response to chemotherapy and are presented in Table 2. Locoregional control was significantly dependent on the clinical response to induction chemotherapy (p 5 0.03, Fig. 4). Of the complete responders, 13 of 15 patients (87%) remained locally controlled without a need for mastectomy. The remaining 2 patients failed in the breast, 1 with an unresectable, isolated recurrence, and the other in conjunction with distant metastatic disease. The median follow-up in this group of patients was 29 months. Amongst the partial responders, 6 of 12 patients were without evidence of disease following radiotherapy, continued with chemotherapy, and avoided mastectomy; 4 of these patients had lasting locoregional control. The remaining 6 partial responders proceeded to mastectomy and remained free of disease recurrence on the chest wall or in the regional lymphatics. Overall, there was an 83% locoregional control rate in partial responders with the selected use of mastectomy. Locoregional control was generally poor in those patients categorized as nonresponders, 45%. However, we found it notable that in 2 patients, locoregional control after radiotherapy was achieved with resultant breast conservation and
Fig. 3. Kaplan-Meier estimates of induction chemotherapy response versus (a) disease-free survival, (b) distant disease-free survival, and (c) overall survival. (- z - z - z ) Complete responders, (222) incomplete responders.
lasting locoregional control until their deaths from distant metastases 4 and 6 years later. Four of the nonresponders to chemotherapy proceeded to mastectomy, 3 of whom remained locoregionally controlled, and 1 failed on the chest wall. Tolerance The treatment schedule of induction chemotherapy followed by accelerated superfractionated radiotherapy was within the range of expected toxicities. Acute skin toxicity ranged from brisk erythema to confluent moist desquamation. Those patients undergoing twice-daily boosting tended toward patchy moist desquamation, as compared to less severe skin reactions in patients boosted once daily. Not unexpectedly, the treatment regimen resulted in significant
294
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 44, Number 2, 1999
Fig. 4. Induction chemotherapy response versus LRC. (- z - z - z ) Complete responders, (.....) partial responders, and (222) nonresponders.
progression of skin toxicity 1–2 weeks following radiotherapy completion; however, all patients healed within 1–2 months. Of those subjected to elective mastectomy for evidence of persistent disease, there were no wound healing complications documented. In those patients avoiding mastectomy and followed a minimum of 12–18 months, a range in cosmetic outcome was observed. Minimal to moderate fibrosis was seen in most patients. However, remarkable retraction was not observed, and there was universal preservation of acceptable symmetry when compared to the contralateral breast. Telangiectasias were observed to a varying degree and attributable to the amount of moist desquamation experienced. Those patients followed longer than 18 months did not demonstrate progression of late effects, and in some there was documentation of breast tissue softening. Severe late toxicity, including pain, breast retraction, and soft tissue necrosis requiring surgery was not documented. Two patients developed significant fibrosis within the treatment fields. Both of these patients began to show signs of generalized fibrosis within 6 –9 months after completing radiotherapy and progressed to significant fibrosis over the subsequent 6 months. DISCUSSION Inflammatory breast carcinoma is an uncommon presentation of breast cancer (36) and, because of its aggressive nature, has been the focus of intense clinical investigation. Multiagent chemotherapy has dramatically affected overall survival rates and has re-emphasized the importance of locoregional control to optimize quality of life and potentially contribute to the improvement of overall survival. In patients who are able to tolerate aggressive treatment in-
volving chemotherapy, radiotherapy, and possibly mastectomy, 5-year actuarial disease-free survival and locoregional control rates are approaching 40% and 85%, respectively (9). This review of 38 patients reports our experience with chemotherapy, accelerated superfractionated radiotherapy, and the selected use of mastectomy. We are able to demonstrate that with aggressive radiotherapy and appropriate decision criteria, breast conservation without compromise of locoregional control can be achieved. Because of the relative infrequency of IBC, most data on patient management are derived from retrospective reviews of institutional experiences, frequently conducted over a long time period. Amongst institutional experiences, two institutions have pioneered the multimodality management of IBC over the past 25 years (1, 3, 5, 6, 15–18, 24, 25). The most recent publications from each of these institutions (6, 17) both advocate a multimodality approach; however, one promotes the routine use of mastectomy while the other supports that mastectomy be reserved only as salvage. Fleming et al. (6) recently reported on the treatment results of 178 patients treated for IBC over a 20-year period. Based on their evaluation of the role of mastectomy, they determined that the addition of mastectomy to chemotherapy and radiotherapy resulted in a lower locoregional failure rate of 16.3% as compared to those patients managed without mastectomy, 35.7%. However, when stratified for response to induction chemotherapy, the improvement in locoregional control with mastectomy was only significant in patients with a partial response to chemotherapy. The added benefit of mastectomy was also present for disease-specific and disease-free survival when complete and partial responders were grouped together for the analysis. Disease-specific survival rates at 5 years for patients treated with and without
Accelerated XRT for inflammatory breast cancer
the addition of mastectomy were 62% and 43%, respectively, with the respective disease free survival rates of 52.6% and 31.2%. A comparison of our experience to that of other institutions is somewhat difficult. One needs to consider that a larger proportion of our patients achieved a complete response to induction chemotherapy and that our patients experienced a high rate of distant metastases and death when a complete response was not achieved. Nevertheless, in our group of 27 complete and partial responders (the selective use of mastectomy was utilized selectively in only 6 patients), occurrence of locoregional failure was only 15%, and compares favorably to reports where mastectomy was used routinely. In addition, our 5-year disease-free survival and overall survival for complete responders was 68% and 33%, respectively. In another large institutional experience, Thomas et al. (17) reported on the treatment results of 125 women with IBC utilizing their approach of alternating chemotherapy and radiotherapy. The reported 5-year disease-free and overall survival rates were 38% and 50%, respectively. Mastectomy was not routinely used, and therefore only 10 patients underwent mastectomy for local progression of disease. The reported complete response rate following this combined modality treatment was 82% with an ultimate local failure rate of 27%. Only 18% of the patients experienced a local recurrence as the first site of failure, and all patients with local failures developed distant metastases within 12 months. This compares favorably to our own experience showing a 26% ultimate locoregional failure rate. Although mastectomy appears to be of benefit upon review of the literature (2, 3, 5–13, 16, 18 –22, 24, 28 –30, 36), this benefit is not clearly evident in all patients. Reports describing treatment approaches that avoid mastectomy (4, 10), thus optimizing breast conservation, employ conventional radiotherapy treatment schedules, and as a result, compromise locoregional control rates. In contrast, the routine use of mastectomy circumvents the opportunity to preserve the breast, and the reported data does not clearly support that it is beneficial in all patients. Despite this, those reporting on treatment approaches proposed to improve treatment outcome recommend the routine use of mastectomy. Mastectomy has also been advocated despite the use of more aggressive radiotherapy to avoid severe late effects
● D. W. ARTHUR et al.
295
(9, 18). However, severe soft tissue fibrosis has not been a universal finding in our own experience or in that of others (1, 10, 14, 15, 17). Furthermore, when aggressive radiotherapy schedules have been utilized, once daily fractionation utilizing fractions larger than 1.8 –2.0 Gy (1, 15, 17) or accelerated superfractionation (14), local control rates are comparable to results reported from treatment approaches utilizing mastectomy. Our data suggest that the selected use of mastectomy optimizes both locoregional control and breast preservation. Based on our present results, we recommend that the response to induction chemotherapy can be used as a guide for the treatment of locoregional disease. In our reported experience, using twice-daily irradiation, the 87% rate of locoregional disease control in complete responders supports that mastectomy can be avoided in this group of patients. This control rate compares favorably with series utilizing mastectomy in their treatment policy. Both partial and nonresponders had a high rate of distant metastases in our study, and one could question the role of mastectomy in these groups. However, based on the contributory effect locoregional control potentially has on survival and the potential reduction in distant metastases that more aggressive chemotherapy regimens may have, we recommend mastectomy in those patients unable to achieve a complete response. The use of a clinical, pathologic, and radiographic evaluation following radiotherapy may add confidence when avoiding mastectomy in initially incomplete responders, but its value as a determinant of outcome requires additional study due to the small numbers in the subgroups of our experience. It is evident in the literature that following a complete response to chemotherapy, an aggressive once- or twicedaily treatment scheme is necessary to achieve high local control rates without the use of mastectomy. Our experience demonstrates that induction chemotherapy, accelerated superfractionated radiotherapy, and the selected use of mastectomy results in excellent locoregional control rates, is well tolerated, and optimizes breast preservation. Despite advances in locoregional management of this disease, systemic disease control continues to be a challenge. Presently, we are continuing with our local management approach and are evaluating the potential benefit of the addition of aggressive chemotherapy with autologous stem cell rescue as a part of the chemotherapy regimen.
REFERENCES 1. Arriagada R, Mouriesse H, Spielmann M, et al. Alternating radiotherapy and chemotherapy in non-metastatic inflammatory breast cancer. Int J Radiat Oncol Biol Phys 1990;19: 1207–1210. 2. Brun B, Otmezguine Y, Feuilhade F, et al. Treatment of inflammatory breast cancer with combination chemotherapy and mastectomy versus breast conservation. Cancer 1988;61: 1096 –1103. 3. Buzdar AU, Montague ED, Barker JL, et al. Management of inflammatory carcinoma of breast with combined modality approach - an update. Cancer 1981;47:2537–2542.
4. Chu AM, Wood WC, Doucette JA. Inflammatory breast carcinoma treated by radical radiotherapy. Cancer 1980;45: 2730 –2737. 5. Fastenberg NA, Buzdar AU, Montague ED, et al. Management of inflammatory carcinoma of the breast. Am J Clin Oncol 1985;8:134 –141. 6. Fleming RY, Asmar L, Buzdar AU, et al. Effectiveness of mastectomy by response to induction chemotherapy for control in inflammatory breast carcinoma. Ann Surg Oncol 1997; 4:452– 461. 7. Hagelberg RS, Jolly PC, Anderson RP. Role of surgery in the
296
8. 9. 10. 11. 12. 13. 14.
15.
16. 17.
18. 19. 20. 21. 22. 23.
I. J. Radiation Oncology
●
Biology
●
Physics
treatment of inflammatory breast carcinoma. Am J Surg 1984; 148:125–131. Knight CD, Martin JK, Welch JS, et al. Surgical considerations after chemotherapy and radiation therapy for inflammatory breast cancer. Surgery 1986;99:385–390. Kuske RR. Diagnosis and management of inflammatory breast cancer. Semin Radiat Oncol 1994;4:270 –282. Lamb CC, Eberlein TJ, Parker LM, et al. Results of radical radiotherapy for inflammatory breast cancer. Am J Surg 1991; 162:236 –242. Morris DM. Mastectomy in the management of patients with inflammatory breast cancer. J Surg Oncol 1983;23:255–258. Perez CA, Fields JN, Fracasso PM, et al. Management of locally advanced carcinoma of the breast. II. Inflammatory carcinoma. Cancer 1994;74:466 – 476. Pisansky TM, Schaid DJ, Loprinzi CL, et al. Inflammatory breast cancer: Integration of irradiation, surgery, and chemotherapy. Am J Clin Oncol 1992;15:376 –387. Radie-Keane K, Wazer DE, Schmidt-Ullrich R. Locoregional control of inflammatory carcinoma of the breast with accelerated superfractionated radiotherapy. Breast Dis 1994;7:59 – 68. Rouesse J, Friedman S, Sarrazin D, et al. Primary chemotherapy in the treatment of inflammatory breast carcinoma: A study of 230 cases from the Institut Gustave-Roussy. J Clin Oncol 1986;4:1765–1771. Singletary SE, Ames FC, Buzdar AU. Management of inflammatory breast cancer. World J Surg 1994;18:87–92. Thomas F, Arriagada R, Spielmann M, et al. Pattern of failure in patients with inflammatory breast cancer treated by alternating radiotherapy and chemotherapy. Cancer 1995;76: 2286 –2290. Thoms WW, McNeese MD, Fletcher GH, et al. Multimodality treatment for inflammatory breast cancer. Int J Radiat Oncol Biol Phys 1989;17:739 –745. Fields JN, Perez CA, Kuske RR, et al. Inflammatory carcinoma of the breast: Treatment results on 107 patients. Int J Radiat Oncol Biol Phys 1989;17:249 –255. Schafer P, Alberto P, Forni M, et al. Surgery as part of a combined modality approach for inflammatory breast carcinoma. Cancer 1987;59:1063–1067. Haagenson CD. Diseases of the breast. 2nd ed. Philadelphia: WB Saunders; 1971. p. 576 –584. Camp E. Inflammatory carcinoma of the breast. Am J Surg 1976;131:583–586. Treves N. The inoperability of inflammatory carcinoma of the breast. Surg Gynecol Obstet 1959;109:240 –242.
Volume 44, Number 2, 1999
24. Koh EH, Buzdar AU, Ames FC, et al. Inflammatory carcinoma of the breast: Results of a combined-modality approach – M.D. Anderson Cancer Center experience. Cancer Chemother Pharmacol 1990;27:94 –100. 25. Krutchik AN, Buzdar AU, Blumenschein GR, et al. Combined chemoimmunotherapy and radiation therapy of inflammatory breast carcinoma. J Surg Oncol 1979;11:325–332. 26. Sataloff DM, Mason BA, Prestipino AJ, et al. Pathologic response to induction chemotherapy in locally advanced carcinoma of the breast: A determinant of outcome. J Am Coll Surg 1995;180:297–304. 27. Maloisel F, Dufour P, Bergerat JP, et al. Results of initial doxorubicin, 5-fluorouracil, and cyclophosphamide combination chemotherapy for inflammatory carcinoma of the breast. Cancer 1990;65:851– 855. 28. Sherry MM, Johnson DH, Page DL, et al. Inflammatory carcinoma of the breast: Clinical review and summary of the Vanderbilt experience with multi-modality therapy. Am J Med 1985;79:355–364. 29. Donegan WL, Padrta B. Combined therapy for inflammatory breast cancer. Arch Surg 1990;125:578 –582. 30. Burton GV, Cox EB, Leight GS, et al. Inflammatory breast carcinoma: Effective multimodal approach. Arch Surg 1987; 122:1329 –1332. 31. Ragaz J, Jackson SM, Ne N, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 1997;337:956 –962. 32. Overgaard M, Hanson PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. N Engl J Med 1997;337:949 –955. 33. Fleming ID, Cooper JS, Henson DE, et al. AJCC cancer staging manual. 5th ed. Philadelphia: Lippincott-Raven; 1997. p. 171–180. 34. Fields KK, Elfenbein GJ, Lazarus HM, et al. Maximum tolerated doses of ifosfamide, carboplatin, and etoposide given over six days followed by autologous stem cell rescue: Toxicity profile. J Clin Oncol 1995;13:323–332. 35. Fields KK, Zorsky PE, Hiemenz JW, et al. Ifosfamide, carboplatin, and etoposide: A new regimen with a broad spectrum of activity. J Clin Oncol 1994;12:544 –552. 36. Hortobaygi GN, Singletary SE, McNeese MD. Treatment of locally advanced and inflammatory breast cancer. In: Harris JR, Lippman ME, Morrow M, Hellman S, editors. Diseases of the breast. Philadelphia: Lippincott-Raven Publishers; 1996. p. 585–599.