Radiation therapy for local-regional recurrent breast carcinoma

0 Original Contribution RADIATION THERAPY FOR LOCAL-REGIONAL RECURRENT BREAST CARCINOMA MELVINDEUTSCH,M.D., JOHN A. PARSONS, M.D. AND BHARAT B. MITTAL, M.D. Joint Radiation Oncology Center, University Health Center of Pittsburgh. University of Pittsburgh. School of Medic: One hundredsevenwomen with recurrent breast carcinoma involving the chest wall and/or regional lymph node regions were treated with radiotherapy between 1970 and 1979. Local-regional tumor was the initial and only evidence of recurrent breast carcinoma in ail cases. Forty-seven patients had their disease confined to the chest wall alone and sixty (56%) patients had chest wall involvement as some component of their local-regionalrecurrent disease. Within five years after the initial mastectomy, 80.5% of recurrences were manifested. All patients had radiotherapy to at least the site of involvement. Eighty-four patients (785%) had a complete response. The absolute Syear survival of all patients following local-regional recurrence was 34.6%. Five year survival was 29% in those patients who had recurrence within 5 years of the orightal mastectomy. For those patients whose localregional recurrence occurred after a S-year disease-free interval, the subsequent 5-year survival was 57%. For patients with recmremz conlined to the chest wail, subsequent S-year survival was 48.9%. Patients who had supraclnvicufar involvement as part of their local-regionalrecurrence had only a 16.1% 5-year survival. The majority of patients developed distant metastasis. Twenty-two patients developed car&mma of the contralateral breast foBowing local-regional recurrence. Five year survival following locakegional recurrence was only 43% for pa. tients whose initial treatment for their primary breast carcinoma was surgery and adjuvant chemotherapy. FOI those patients whose primary breast carcinoma was treated by surgery ahme or surgery and post-operative radio therapy, the S-year survival following local-regional recurrencewas over 49%. Breast carcinoma, Local recurrence, Node recurrence, Radiotherapy. INTRODUCTION

and patients with 4-7 positive nodes had s (8/67) recurrence rate. Patients with eight or IT tive nodes had a local recurrence rate of 36.29 His description of local recurrence included 1 volving the chest wall, internal mammary nod axilla. Supraclavicular recurrence as the first I lapse occurred in 1.9% (15/794). Fisher et al.’ local-regional recurrence as the first evidence in 10.4% of 654 patients treated by radical ma It was the first sign of failure in 16.9% of pat relapsed. Donegan ef al.’ in 1966, and Haag 1974. both recommended radiotherapy as the for local recurrence. However, 5-year survive than 12% in both series. More recently, Bet al. ‘Z have reported a 36% 5-year survival folio\; regional recurrence and have emphasized tl tance of aggressive radiotherapy in the mana local-regional recurrent breast carcinoma. This report is a retrospective review of a treated, over a IO-year period at the Univers Center of Pittsburgh, with radiotherapy fo

A significant proportion of breast cancer patients treated with mastectomy will ultimately present with recurrent tumor involving the chest wall and/or the regional lymph node areas as the first sign of relapse. Donegan ef ~1.~reported chest wall recurrence as the first sign of relapse in 80 of 704 patients undergoing a radical mastectomy (11.4%). An additional 64 patients had chest wall recurrence either concomitant with or following the appearance of distant metastasis. The percentage of patients having local recurrence on the chest wall varied from 6.5% for those patients with uninvolved axillary nodes to over 33% of patients with five or more involved axillary nodes. Haagensen’ also showed a correlation between the number of positive nodes and the incidence of local recurrence within 10 years following radical mastectomy for operable patients (Columbia Stage A and B). For patients with negative axillary nodes, the local recurrence rate was 1.5% (7/466). Those patients with l-3 positive nodes had a local recurrence rate of4.7% (9/l 92) Presented in part at ASTRO Meeting held in Miami Beach, Florida on September 30,1985. Reprint requests to: Melvin Deutsch, M.D., Joint Radiation

Accepted for publication 14 July 1986.

Oncology Center, 230 Lothrop Street, Pittsburgh, PA 152 13. 206 I

I. J. Radiation Oncology 0 Biology 0 Physics

2062

December 1986, Volume 12, Number 12

Table I. Sites of recurrence

Table 3. Interval primary breast c; to local-regional recurrence

Single

No.

sites

patients

%

Mult. sites

No. patients

No. patients

Interval

cw

47 18 11

SC IM AX

IF Total

43.9 16.8 10.3 7.5

!

2.8

87

81.3

cw, SC

5 4 I

CW,AX CW, IM

CW, IF SC, AX SC,IM SC, IF AX, IM CW, SC, AX CW, SC, IM, IF Total

: 1 1 1 2 20 (1:.7%)

gional breast carcinoma presenting as the first sign of relapse.

METHODS AND MATERIALS One hundred and seven women with recurrent carcinoma of the breast were treated with radiotherapy at the University Health Center of Pittsburgh between 1970 and 1979. In all cases, ipsilateral local and/or regional node recurrent breast cancer was the first and only site of relapse. Seven patients had bilateral breast cancer prior to developing local-regional recurrent tumor. in these seven cases, recurrent tumor was confined to one side and it was assumed that the recurrence was from the ipsilateral carcinoma of the breast. Ali patients have been followed until death or are known to be alive more than 5 years since the diagnosis of recurrent tumor. All patients were initially treated with mastectomy f adjuvant therapy, except for one patient who had a lumpectomy, axillary dissection, and post-operative breast irradiation. The local-regional area of the breast was divided into five sites: 1. 2. 3. 4. 5.

Chest Wall (CW); Supraclavicular Area (SC): Internal Mammary Node Area (IM); Axilla (AX); Infraclavicular Area (IF).

Eighty-seven patients (8 I .3%) had recurrent disease confined to only one site. The chest wall was the most common site of local-regional recurrence (Table I). Twenty patients ( 18.7%) had recurrent disease involving

51 year 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 IO-15 >I5

31 29 17 7 2 5 2 1 : 5 4

% of total

‘i 72% a fS, 15:9

3 y1

6.5 1.8 4.7 1.8 1.0 1.0 2.8 4.1 3.7

two or more sites. Chest wall involvement other regional recurrence, occurred in 6 1 1 (Table 2). The supraclavicular area was t common site of involvement with 31 pi Neither age at diagnosis of breast cancer I time of recurrence was associated with any of recurrent tumor. The majority of cases of recurrent tumo dent within 3 years of the mastectomy (T years, over 80% of the patients had recurrer interval from initial diagnosis to recurrent t vary much with the site of recurrence. Pa single site of local-regional recurrent turn with recurrent disease before 5 years in 779 the 20 patients with multiple sites of recul 19 (95%) had recurrence presenting within : initial mastectomy, whereas only one patiel pie site recurrence presenting more than 5 yc initial mastectomy. Therapy for the initial breast carcinoma surgery alone in 63 patients (58.9%) (Tablc and chemotherapy was the initial treatme tients (2 1.SW). Surgery and post-operative I was administered to 17 patients ( 15.9%) al the initial diagnosis. Two patients were trea gery and hormone manipulation and an ad patients had surgery, radiotherapy. and ch Table 4. Therapy of primary breast C, and survival after recurrence No.

Table 2. Sites of recurrence: Single and multiple

All CW AI1 SC All IM All AX All IF

No. patients

%

61 31 15 18 6

57 29 14 16.8 5.6

initial therapy

surgery

Surgery Surgery Surgery Surgery

+ + + +

XRT Chem Hor XRT + Chem

XRT = Radiotherapy: monal therapy.

treated

5

aftc

63 17 23 2 2 Chem = Chemotherapy;

‘

Local-regional breast carcinoma 0 M. DEUTSCH et al.

Table 5. &sponse and survival survival

Response 84/107 38/84 40/84 6184 21/107 21107

(78.5%) CR (45.2%) LR control (47.6%) LR recurrence LR uncertain (19.6%) PR Response uncertain

37/107 18/107 IO/107 19/107 21107

(34.6%) (16.8%) (9.3%) (17.8%)

Alive r5 years StillsliveS-15years Alive NED 5-15 years Dead > 5 years Dead > 5 years NED

CR = Complete response; LR = Local-regional; PR = Partial response; NED = No evidence of disease. for their initial carcinoma of the breast. One of these latter patients had breast irradiation after lumpectomy and

had her first recurrence in the infraclavicular region. The other patient had irradiation of the supraclavicular area (dose and field size unknown) and had her recurrence in the supraclavicular area. Of the 17 patients who had just post-operative radiotherapy, 13 had super-voltageirradiation of the internal mammary-supraclavicular area to 4450-4500 rad/20-2 1 fractions. The remaining four patients had orthovoltage irradiation (sites and doses unknown). Of these latter four patients, three had chest wall recurrence and the fourth had recurrence in the supraclavicular region. Of the 13 patients who had supervoltage irradiation of the IM-SC area, 8 had chest wall recurrence, 1 had supraclavicular recurrence, 1 had internal mammary node recurrence, 1 had axillary node recurrence, 1 had chest wall and infraclavicular recurrence, and 1 had recurrence in the chest wall, supraclavicular region, and axilla. Thus, of the 13 patients who had supervoltage irradiation of the IM-SC area, three had recurrence within the irradiated fields. In 8 1 patients, the initial nodal status was known. Fifty-eight patients (54%) had positive nodes. Twenty-three patients (2 1%)were known to have negative nodes. In 26 patients (24%), the status of lymph nodes was unknown. Many of this latter group had just a simple mastectomy. The site of recurrent tumor did not correlate with the initial nodal status. All patients were treated with radiotherapy for their local-regional recurrent tumor. In addition, 38 patients also received systemic therapy either right before or after radiotherapy. Sixty-four patients (59.8%) received radiation to the chest wall either alone or in combination with other sites. Fifty-three patients had radiotherapy to the supraclavicular region either alone or in combination with other sites. Only 12 patients (11.2%) did not have irradiation of the chest wall or supraclavicular region. The administered tumor doses were at least equivalent to 4500 rad/25 fractions (TDF = 70) in all patients except for four, who received lower doses because of deteriorating condition. RESULTS

In 84 patients (78.5%), there was complete disappearance of all local-regional recurrent tumor following ra-

diotherapy (Table 5). Twenty-one patients ( 19.6’ a partial response, and in two patients the response was uncertain. The likelihood of complete respor not related to age at recurrent tumor nor to the therapy for the primary breast carcinoma. Likeu sponse rates did not differ significantly accordinf site of recurrence. The complete response rate w ilar in those patients with a single site (80.5%) an with multiple sites of recurrence (70%). Of the 84 patients who had a complete respc have had subsequent local/regional recurrence. one had subsequent local recurrence involving, chest wall and two additional patients had recurr volving the chest wail and axilla. Four of the 40 had subsequent recurrence involving the supracl; region and 2 additional patients had subsequen rence involving just the axilla. One patient wit1 quent recurrence in the ipsilateral breast original segmental mastectomy and axillary node dissec lowed by postoperative radiotherapy just to the Her first recurrence was in the infraclavicular ref she subsequently had recurrence in the breast ! by distant metastasis, Thirty-seven (34.6%) of 107 patients treated f regional recurrence survived at least 5 years frorr sis of recurrent tumor (Tables 5, 6, 7). Patien local/regional recurrence was limited just to 1 wall did best with a 5-year absolute survival I

Table 6.5-year survival vs. sites of recurrem No. patients r IM AX IF Total single sites Multiple Total

23/47 2118 4/11 318 O/3 32187 5120 371107

All CW All SC All IM All AX All IF

27161 5131 4115 5/18 116

1.J. Radiation Oncology 0 Biology 0 Physics

2064

Table 7. 5-year survival vs. interval to recurrence Interval 41 year :I: 3-4 4-5 Total ~5 Total >5

S-year survival

%

513 1

16.1

9129 6117 317 212 25186 12121

31 35.3 42.9 100 29 57

(23/47). Patients with recurrence in the supraclavicular region did poorly with a 16.1%5 year survival for all patients with any supraclavicular involvement. Patients with a single site of&al/regional recurrence had a 36.8% 5-year survival vs. 25% (5/20) 5-year survival for patients with multiple sites of recurrence (Table 6). Forty-seven patients had isolated chest wall recurrence. Thirty-three received radiotherapy just to the chest wall and their 5-year survival was 54.5% (18/33). Fourteen patients were treated with radiotherapy to the chest wall plus additional sites prophylactically and 5 patients survived greater than 5 years (35.7%). Patients treated with radiotherapy just to the chest wall probably had less involvement than those treated more extensively. Only three patients with a partial response survived over 5 years from the onset of local/regional recurrence. They all died with distant metastasis. Sixty-nine patients had radiotherapy as their initial treatment for local-regional recurrence. Thirty-eight patients had chemotherapy, and/or hormone manipulation, in addition to radiotherapy. Five-year survival did not differ significantly between those patients treated just with radiotherapy (3 1.9%)and those treated with radiotherapy and systemic therapy (39.5%). The initial nodal status was not a strong predictor of 5-year survival after recurrence, although there was a trend for patients whose nodes were initially positive to do a little bit worse (25.9% 5-year survival) than those whose nodes were uninvolved (43.5%) or those whose nodal status was unknown (46.2%). Neither age at initial diagnosis nor age of time of recurrence correlated with the likelihood of subsequent 5-year survival. There was a relationship between 5-year survival following local-regional recurrence and the interval from initial therapy to recurrence. Patients whose disease recurred after 5 years from mastectomy had a 57% subsequent 5 year survival versus a 29% 5-year survival for those patients whose local-regional recurrence occurred within 5 years of initial therapy. Survival was especially poor for patients with a recurrence within 1 year ofinitial treatment (Table 7). Patients whose initial therapy prior to development of recurrence with surgery alone or surgery and radiother-

December 1986. Volume 12. Number I2

apy did fairly well with post-recurrence 5 of 42.9 and 47.1% respectively (Table 4: patients whose initial therapy with surge1 therapy had a very poor 5-year survival recurrence (4.3%). However, of the 23 patj initially treated with surgery and chemoti involved lymph nodes. For 19 of these pati ber of lymph nodes involved was known tc erage 7.2). The poor survival of this grow seemed to be related to the earlier onset oft tasis. Of the 17 patients who received surge therapy as their initial treatment, 12 had in, 2 had negative nodes, and in 3 patients tht was unknown. Patients with positive nodes valved nodes (average 4.5). The majority of patients did develop dis sis, with bone being the most common site. patients developed tumor involving the 01: following local-regional recurrence. It is nc state whether the contralateral breast tumor metastatic involvement or new primaries. least 16 patients developed other evidence c Only five patients with opposite breast invo vived longer than 5 years. Two are still alivf disease. There were 37 5-year survivors (Table 5). f ber, 18 patients ( 16.8% of total) are still alil patients (9.3%) are still alive free of tumor. I’ tients died more than 5 years after local-reg rence. Two were without evidence of diseu of death. Of the 47 patients with just chest rence, 38 had a complete response and 18 h ous local-regional control (Table 8). Twel with initial chest wall recurrence are still alil are alive free of disease. Of the ten patients st of disease, eight had local-regional recurrent to the chest wall, one had recurrent disease the internal mammary nodes, and one had su lar and axillary node involvement at the tin rence. Two of these ten patients develope breast cancer. One was treated with segmen tomy and chemotherapy and the second was t a modified radical mastectomy and chemothe

Table 8. Isolatedchest wail recurrence: 47 pa 38/47 (81%) 14/38 (36.8%) I 5 18/38 (47.4%) 23147 (48.9%) 12/47 (25.6%) 8/47 (I 7%)

Complete respono Further recurrence Further recurrence Uncertain Continuous LR co 5 year survival Still alive Alive NED

LR = Local-regional; NED = No evidence of disc

Local-regional breastcarcinoma 0 M. DEUTSCHer al are alive and free of disease. Another patient who is still alive free of disease had further chest wall recurrence

treated by excision and remains well over 5 years later.

DISCUSSION The results of this retrospective survey confirm the conclusions of others concerning the overall poor prognosis of local-regional recurrent breast carcinoma. The great majority of patients ultimately develop distant metastasis and die of tumor. However, the absolute 5-year survival in this series is 34.6%. This is rather similar to the overall 5-year survival of 36% reported by Bedwinck et al. for patients with isolated local-regional recurrent breast carcinoma.’ For patients with isolated chest wall recurrence, the survival is 48.9%. These results support the aggressive treatment of local-regional recurrent breast carcinoma in hopes of achieving at least long term local-regional control. Factors associated with a relatively high likelihood of 5-year survival following local-regional recurrence were: recurrence confined to the chest wall, a long disease-free interval following initial treatment, and initial treatment that did not include chemotherapy. Conversely, factors associated with poor survival following local-regional recurrence were: disease involving the supraclavicular region, a short disease-free interval following initial therapy, especially under 1 year, and initial therapy with chemotherapy. Factors that had little, if any, influence on survival following local-regional recurrence were: age at initial diagnosis, age at the time of recurrence, and the original nodal status.

For patients with recurrence involving the chest we advocate irradiation of a large portion of the wall including the entire mastectomy incision. V longer administer small field radiotherapy just to t mor. Unless the chest wall recurrence is extensik supraclavicular region is not irradiated prophylact Likewise, the internal mammary nodes and axilla a irradiated prophylacticaIly. For patients with isola currence involving just the supraclavicular region, therapy is administered just to the supraclavicu gion. Doses of at least 5000 rad in 25 fractions sho administered to the entire site of involvement w additional 1000-2000 rad boost using small fie rected to the actual tumor or biopsy sites. From this retrospective review, it is not possible fine the role of systemic therapy in the managen local/regional recurrence. Although there was n benefit to systemic therapy for recurrence in this the systemic therapies administered varied widt there probably were selection factors involved ir mining who received systemic therapy in additio diotherapy. It must be emphasized that patients who hav regional recurrent breast cancer are at high risk ther local/regional disease and distant metastas ther local/regional recurrence involving the chr should be treated aggressively since there have b sional long-term disease-free survivors following or re-irradiation. In addition, there is a high in of subsequent tumor involving the contralatera’ Subsequent contralateral breast cancer in the otherwise free of disease, should be treated as WC other primary breast carcinoma of similar stage.

REFERENCES 1. Bedwinek, J.M., Fineberg, B., Lee. J.. Ocwieza. M.: Analysisof

failures following local treatment of isolated local/ regional recurrence of breast cancer. Int. J. Radial. Oncol. Biol. Phys. 7: 581-585, 1981.

2. Bedwinek, J.M., Lee, J., Fineberg. B.. Ocwieza. M.: Prognostic indicators in patients with isolated local/regional recurrence of breast cancer. Cancer 47: 2232-2235, 198 1. 3. Donegan, W.L., Perez-Mesa, C.M.. Watson, F.R.: A bio-

statistical

study of locally recurrent

breast ca

Surg. GynecoL Obst. 122: 529-540, 1966. 4. Fisher, B., Redmond, C., Fisher, E.R., Bauer, N ark, N., Wickerham,

D.L., Deutsch, M., Mon

Margolese, R., Foster, R.: Ten-year results of a ized clinical trial comparing radical mastectomy mastectomy with or without radiation. N. Eng 312: 6X-681,1985. 5. Haagensen, C.D.: The choice of treatment of opt cinema of the breast. Surgery 76: 685-7 14, 1974