0360.3016/90 $3.00t 00

Capynght ‘c 1990Pergamon Press plc

??Original Contribution

ISOLATED

KAREN DELIA

LOCAL-REGIONAL MASTECTOMY: J. HALVERSON, M. GARCIA,

M.D.,

M.D.,

RECURRENCE OF BREAST CANCER RADIOTHERAPEUTIC MANAGEMENT CARLOS

A. PEREZ,

JOSEPH R. SIMPSON,

M.D.,

M.D.,

ROBERT

PH.D.

FOLLOWING

R. KUSKE,

AND BARBARA

M.D.,

FINEBERG

Radiation Oncology Center. Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, MO Two hundred twenty-four patients with their first, isolated local-regional recurrence of breast cancer were irradiated with curative intent. Patients who had previous chest wall or regional lymphatic irradiation were not included in the study. With a median follow-up of 46 months (range 24 to 241 months), the S- and IO-year survival for the entire group were 43% and 26%. respectively. Overall, 57% of the patients were projected to be loco-regionally controlled at 5 years. The 5-year local-regional tumor control was best for patients with isolated chest wall recurrences (63%), intermediate for nodal recurrences (45%), and poor for concomitant chest wall and nodal recurrences (27%). In patients with solitary chest wall recurrences, large field radiotherapy encompassing the entire chest wall resulted in a 5- and IO-year freedom from chest wall re-recurrence of 75% and 63% in contrast to 36% and 18% with small field irradiation (p = 0.0001). For the group with recurrences completely excised, tumor control was adequate at all doses ranging from 4500 to 7000 cGy. For the recurrences < 3 cm, 100% were controlled at doses > 6000 cGy versus 76% at lower doses. No dose response could be demonstrated for the larger lesions. The supraclavicular failure rate was 16% without elective radiotherapy versus 6% with elective radiotherapy (p = 0.0489). Prophylactic irradiation of the uninvolved chest wall decreased the subsequent re-recurrence rate (17% versus 27%), but the difference is not statistically significant (p = .32). The incidence of chest wall re-recurrence was 12% with doses 2 5000 cGy compared to 27% with no elective radiotherapy, but again was not statistically significant (p = .20). Axillary and internal mammary failures were infrequent, regardless of prophylactic treatment. Although the majority of patients with local and/or regional recurrence of breast cancer will eventually develop distant metastases and succumb to their disease, a significant percentage will live 5 years. Therefore, aggressive radiotherapy should be used to provide optimal local-regional control. We recommend (a) radiation therapy to the entire site of involvement, as more localized therapy is associated with an excessive incidence of re-recurrence; (b) elective irradiation of the uninvolved supraclavicular fossa to 4600-5000 cGy; (c) serious consideration for elective chest wall irradiation to at least 5000 cGy, particularly in patients with supraclavicolar or axillary involvement since chest wall failure developed in 29% and 21% of these patients respectively; (d) at least 5000 cGy for completely excised recurrences, and at least 6000 cGy for incompletely excised, small (~3 cm) recurrences. The tumor control in larger lesions was only 50% even with doses of 7000 cGy. Breast cancer, Radiation

therapy,

Local-regional

recurrence.

INTRODUCTION Local (chest wall) and/or regional (axillary, supraclavicular. or internal mammary lymph node) recurrence occurs in approximately 10% to 30% of patients undergoing mastectomy for operable breast cancer ( 13, 25. 26, 27). with 70% to 93% of these recurrences involving the chest wall ( I, 8, 1 1, 22, 29). Previous reports have demonstrated that irradiation is superior to systemic therapy (2, 20) or surgery alone (3, 13) in providing local control, yet even with comprehensive radiotherapy, 30% to 40% will fail locally ( 1, 7. 20, 32. 29).

Controversy exists as to the optimal volume of irradiation, with some authors advocating elective irradiation of all local and regional areas (34, 39) or elective irradiation of the chest wall and selected nodal regions (3, 7) or involved field radiotherapy only ( 12). Few reports provide information regarding dose response (1, 3, 8) and the complications of radiation therapy for local-regional recurrence is poorly documented (24). This retrospective study was undertaken to determine survival, local-regional outcome, optimal volume and doses of irradiation, and the value of elective XRT of uninvolved sites.

Presented at the 3 1st Annual American Society for Therapeutic Radiology and Oncology (ASTRO) Meeting. San Francisco. CA. l-6 October 1989.

Reprint requests to: Carlos A. Perez, M.D., Radiation Oncology Center. Suite 4 I I. 45 I I Forest Park. St. Louis, MO 63 108. Accepted for publication 29 March 1990. 851

1. J. Radiation

852

METHODS

Oncology

0 Biology 0 Physics

October

1990, Volume

I Y. Number

4

AND MATERIALS

Patient cliaracteristics Between January 1, 1964 and December 3 1. 1986,224 patients with isolated local-regional recurrence of breast cancer were treated with curative intent in the Radiation Oncology Center, Mallinckrodt Institute of Radiology and affiliated hospitals. We chose to review as homogeneous a population as possible and therefore, excluded patients who had received postoperative irradiation, regardless of how inadequate that treatment may have been: patients referred to Washington University after failing previous therapies for local-regional recurrences at other institutions: and those with clinical or radiographic evidence of distant metastases. The anatomical distribution of recurrence is shown in Table 1. The initial surgical procedure was a radical, modified radical, or total mastectomy in 34%, 61%#, and 5% of the patients, respectively. Adjuvant systemic therapy was administered to 65 patients, including chemotherapy alone in 50. chemotherapy plus hormones in 11, and hormonal manipulation Evaluation

Fig. I. Diagram illustrating the four local-regional sites: chest wall. internal mammary nodes. axilla and supraclavicular fossa: and small versus large field chest wall irradiation portal.

alone in 4. at the time of local-regional

recurrence included chest X ray in 83%. bone scan in 64’%, liver spleen scan in 35%). and abdominal CT scan in 6.3%. In general. the search to find metastatic disease was more extensive in recent years. At the time of recurrence. 49 patients received chemotherapy, most frequently CMF. CAF. or another multi-agent regimen. Sixty-nine patients received hormonal treatment. usually tamoxifen. Because this retrospective review spans over 2 decades, the radiotherapy techniques varied considerably. Early on, small fields encompassing the recurrent lesion with only a 1-2 cm margin were commonly used. It was unusual for the uninvolved local-regional areas to be electively treated. After small field radiotherapy was found to be associated with frequent relapses. large fields encompassing the entire area of involvement were used exclusively. In addition, there was a trend towards elective irradiation of a// uninvolved adjacent regions. Most recently. our

Table I. Carcinoma of the breast local-regional involved sites at nresentation

recurrences:

Number of patients (%) Chest wall (CW) Supraclavicular fossa (SCV) Axilla (AX) Internal mammary nodes (IM) cw + scv CW+AX CW+IM IM + SCV SCV + AX CW+SCV+AX CW+SCV+IM CW + SCV + IM + AX

134 (60 ) 29(13) 17 (7) 13 (6) 13 (6) 4 (2) 2 4 5 1

(1) (2) (2) (0.4) I (0.4) 1 (0.4)

treatment policy has required elective of irradiation of the uninvolved chest wall and supraclavicular fossa only.

Radiation tlwr.up~~ In general. the techniques of irradiation used were as follows: the entire chest wall was treated with split beam and medial and lateral tangential ports with 1.5 to 6.0 MV photons to 4500-5000 cGy calculated at the midpoint. Compensating filters to optimize dose distribution and skin bolus were customary. Macroscopic disease was boosted with appositional 6-12 MeV electron portals. The supraclavicular fossa was treated with an AP field, angled 15 to 20” laterally to avoid the spinal cord. using 1.5 to 6.0 MV photons. After 4500 to 5000 cGy was delivered at a depth of 3 cm, a 9-16 MeV electron beam boost was administered to macroscopic disease. The midplane of the axilla typically receives 75% to 85%) of the supraclavicular dose. Frequently the axilla was boosted with a PA portal. 1.5 to 6.0 MV photons. to complete 4500-5000 cGy. Occasionally higher prophylactic axillary doses were given. The AP internal mammary field, which typically measures 6 X 10 cm to 6 X 12 cm, was irradiated with a 3: I combination of 12 MeV electrons and photons. Early in the study period several A-P parasternal fields were treated exclusively with 4-6 MV photons to 46005000 cGy. Approximately 90% of the treatments were given at 180-200 cGy per day. No patient received hyperfractionated or electron arc therapy. Only four patients were treated with hyperthermia. as this modality was in general reserved for patients who had failed previous RT. Local-regional tumor control was analyzed in detail. For this analysis the region was divided into four specific sites: chest wall. supraclavicular fossa, axilla, and internal mammary chain, as shown in Figure I. “Recurrence” refers to the first appearance of tumor after mastectomy. A

Recurrent breast cancer following mastectomy 0 K. J. HALVERSONel

‘Ye-recurrence” refers to the development of tumor within the local-regional area following treatment. A re-recurrence does not specifically indicate an infield failure; for example, a tumor which develops within the chest wall but outside a small irradiation portal is considered a chest wall re-recurrence. Similarly, the development of tumor in the uninvolved supraclavicular fossa following chest wall XRT is deemed a supraclavicular re-recurrence. The chest wall and supraclavicular fossa were considered to be electively treated when the entire region was irradiated via medial and lateral tangents (1.25-6 MV photons) or a large AP field, respectively. The internal mammary region was scored as electively treated when an AP photon or photon/electron parasternal field was used. This region was IZOZdeemed prophylactically treated when incorporated within the chest wall tangents. The axilla was scored as receiving elective treatment when the axillary midplane dose was carried to 24500. A grade 1 complication was mild or self limited in nature; grade 2 required outpatient management or life style adjustment: and grade 3 necessitated significant intervention or was life threatening. Follow-up data were obtained from records in the Radiation Oncology Center, from referring physicians. and occasionally from correspondence directly with the patient or family. The minimum, maximum, and median followup was 24 months, 241 months. and 46 months, respectively. Stutistical ana/j3is The curves for overall and disease-free survival were generated using the actuarial life table method of Cutler and Ederer (IO). Times for survival and/or failure were calculated from the time of local-regional recurrence. Statistical significance of comparisons between different pa-

853

al.

Table 2. Carcinoma of the breast local-regional overall actuarial survival

5-year %

1O-year

50 19 36

32 12 15

Chest wall alone recurrence (n = 133) Chest wall + nodal recurrence (n = 21) Nodal alone recurrence (n = 66) p = 0.0002

Gent

groups

was

evaluated

square or the generalized method (5).

recurrences:

%

of Pearson Chi(Breslow) statistical

by means

Wilcoxon

RESULTS The 5- and I O-year overall and disease-free survival for the entire study group was 43 and 26% and 25 and 15%, respectively (Fig. 2). At last follow-up 136 patients had died. with 86% of these deaths caused by breast cancer, whereas 13%) succumbed to intercurrent disease. The 5-year survival and relapse-free survival were best in patients who recurred only in the chest wall (50% and 35%). intermediate in those recurring solely in a nodal region (36% and 15%), and poor in patients with concomitant chest wall and nodal recurrence (19% and 5%). as shown in Tables 2 and 3. These differences were significant for survival (p = 0.0002) and relapse-free survival (p = 0.000 1). For patients with isolated chest wall recurrences, the 5year survival was 62%. 58%, 31%. 43%, and 20% for excised, 5 cm/diffuse (cancer en cuirrase) lesions, respectively. Local-regional tumor control Overall. there were 259 sites involved (Table 1). At last follow-up, the local status was known in 235 of which 162 (69%) were controlled. At 5 years, 57% of the study group was projected to be without evidence of local-regional disease. that is. all four sites controlled. The probability of remaining free of local-regional re-recurrence at 5 years was best for patients with chest wall recurrences (63%), intermediate for nodal recurrence (45%), and poor for those with concomitant chest wall and nodal recurrences (27%), p = 0.0084.

q

Table 3. Carcinoma of the breast local-regional overall relapse free survival Relapse

000

Free Survival

lo

TIME (years)

Fig. 2. Overall survival and disease-free survival for the study population.

Chest wall alone recurrence (n = 124) Chest wall + nodal recurrence (n = 19) Nodal alone recurrence (n = 62) p = 0.0001

recurrences:

S-year %

1O-year %

35 5 15

18 0 11

1. J. Radiation

854

Oncology

0 Biology 0 Physics

Chest wall alone recurrence (%)

Chest wall + nodal recurrence (%I)

Excised 5 cm Multiple/diffuse

55/61 14/17 6/15 l/3 l4/26

2/5 l/l l/l O/l 3/9

Total

90/122 (74)

Size

Our ability

to control

(90) (82) (40) (33) (54)

isolated

(40) (100) (100)

Nodal recurrence (%I) (82) (69) (70) (56)

64/90 (7 I )

wall or nodal

re-

in Table 4. The control rate was 85%, and 75% for excised and (3 cm recurrences, but only 53%’ for the larger tumors. Concomitant chest wall and nodal recurrences were difficult to manage, which is due in part to the substantial proportion (65%) of large recurrences in this group: yet even in the small (excised or