Radiotherapy after Breast-Preserving Surgery in Women with Localized Cancer of the Breast

Treatment Groups

Between 1987 and 1989, 579 women with carcinomas of the breast less than 2.5 cm in maximal diameter at pathological examination were randomly assigned to one of two treatments: 299 patients were assigned to quadrantectomy combined with axillary dissection and radiotherapy, and 280 to quadrantectomy combined with axillary dissection without immediate radiotherapy. Randomization to prophylactic radiotherapy or to therapeutic radiotherapy only in case of recurrence was performed immediately after surgery (quadrantectomy and axillary dissection). In 12 patients the margins of the resection were found to contain tumorous tissue (5 patients assigned to quadrantectomy with radiotherapy and 7 assigned to quadrantectomy alone); consequently, these patients were excluded, since they were candidates for a second operation or radiotherapy.

Table 1. Table 1. Characteristics of the Treatment Groups.

Table 2. Table 2. Adjuvant Treatment Administered to Patients with Nodal Involvement.

There was no significant difference between the two treatment groups in age, site and size of the primary carcinoma, histologic characteristics, and prevalence of axillary invasion (Table 1). Similarly, there was no significant difference between the groups in the distribution of patients with nodal involvement according to the adjuvant treatments administered for this condition (Table 2). The two groups of patients were therefore comparable.

Surgical Technique

Quadrantectomy consisted of extensive breast resection, including a portion of the skin overlying the tumor and the underlying fascia, generally performed with a radial incision. Axillary dissection always included total removal of all lymph nodes of the three levels up to the apex of the axilla. When the tumor was located in the upper outer quadrant, the nodes were removed en bloc with the primary carcinoma. In other locations, the dissection was performed through a separate incision.

Pathological Examination

The specimen of breast tissue was examined, and the whole surface of the resected area was stained. The primary carcinoma and the extent of the normal tissue removed around the tumor were measured on the cut surface. The minimal distance of the tumor (whether the carcinoma was infiltrating, intraductal, or both) from the inked margin was measured with the aid of a micrometric ocular after at least five histologic sections 5 micrometers thick were examined.

Radiotherapy

Patients treated with both quadrantectomy and radiotherapy received radiation therapy from a cobalt unit or a 6-MeV linear accelerator, starting four to six weeks after surgery. The breast was treated with two opposing tangential fields (a total of 50 Gy given over a five-week period, with a daily target dose of 2 Gy) and then with a boost dose to the tumor bed with an orthovoltage unit (10 Gy in five fractions).

Follow-up

Follow-up was strict and included routine clinical examination every four months and mammography, chest radiography, bone scanning, and ultrasonography of the liver every year. If a local recurrence was suspected, mammography and a fine-needle biopsy were performed. The median length of follow-up was 39 months (range, 28 to 54).

Adjuvant Treatment

Patients with positive axillary nodes (nodal involvement) were treated with adjuvant medical therapy: premenopausal patients and postmenopausal patients negative for estrogen receptors received chemotherapy (a regimen of cyclophosphamide, methotrexate, and fluorouracil), and postmenopausal patients positive for estrogen receptors received tamoxifen. Nine patients with positive nodes declined adjuvant treatment (six patients in the group treated with quadrantectomy and radiotherapy and three in the group treated with quadrantectomy).

Statistical Analysis

Life tables were constructed by the Kaplan-Meier method, and P values calculated with the log-rank test, on the basis of overall follow-up^9,10.

As in our previous trials, we defined a local recurrence as the appearance of any new tumor in the breast within 2 cm of the surgical scar, and a second primary carcinoma as the appearance of any new tumor in other quadrants of the breast more than 2 cm from the scar.

Table 3. Table 3. Local Recurrences and First Adverse Events in the Treatment Groups.

Figure 1. Figure 1. Cumulative Incidence of Local Recurrences among 567 Women with Breast Cancer, According to Treatment Group.

The values in parentheses are the number of recurrences, followed by the number of patients in the treatment group.

We observed a considerable difference in the rate of local recurrence between the two treatment groups (Table 3). Only 1 of the 294 patients treated with quadrantectomy and radiotherapy (0.3 percent) had a local recurrence, as compared with 24 of 273 patients treated with quadrantectomy (8.8 percent) (Figure 1).

Table 4. Table 4. Local Recurrences among 273 Patients Treated with Quadrantectomy Alone, According to Tumor Diameter, Age, Nodal Status, and Presence or Absence of Extensive Intraductal Component.

Figure 2. Figure 2. Cumulative Incidence of Local Recurrences among 273 Patients Treated with Quadrantectomy Alone, According to Age.

Figure 3. Figure 3. Cumulative Incidence of Local Recurrences among 273 Patients Treated with Quadrantectomy Alone, According to Nodal Status.

Figure 4. Figure 4. Cumulative Incidence of Local Recurrences among 273 Patients Treated with Quadrantectomy Alone, According to the Presence or Absence of an Extensive Intraductal Component.

Although the analysis of risk factors for local recurrence was not the primary objective of the study, we found that among the many variables evaluated (Table 4 and Figure 2), age was the most important. In the quadrantectomy group, the recurrence rate was 17.5 percent among patients 45 years of age or younger and 3.8 percent among those over 55. Since a difference between subgroups in the rate of distant recurrence might obscure the risk of local recurrence, we analyzed the rate of occurrence of distant metastases. There were no differences among the various age groups. We also found no differences in recurrence rates among subgroups defined according to tumor size or the status of the axillary nodes (positive or negative) (Figure 3). Patients with an extensive intraductal component had a much higher rate of local recurrence (7 of 42, or 16.7 percent) than patients without this histologic feature (17 of 231, or 7.4 percent) (Figure 4). There was no difference between the rate of local recurrence among 184 estrogen-receptor-positive patients (16 recurrences, or 8.6 percent) and the rate among 48 estrogen-receptor-negative patients (5 recurrences, or 10.4 percent).

Second primary carcinomas of the ipsilateral breast developed in four of the patients treated with quadrantectomy, but in none of those treated with quadrantectomy and radiotherapy.

According to the protocol, patients in the quadrantectomy group who had a local recurrence were to be treated with radiotherapy after reexcision. However, of the 24 patients with recurrences, only 15 received this treatment. The other nine patients underwent total mastectomy, performed either because the recurring carcinoma was too large and the breast too small (four patients) or because the patient asked that the breast be removed (five patients). Of the 15 patients treated with reexcision and radiotherapy, 1 patient had a second local recurrence and 1 had a new carcinoma in another quadrant. Of four patients with second primary carcinomas, three underwent a second excision and radiotherapy and one underwent a mastectomy.

There was no difference between the two treatment groups in survival four years after the beginning of the trial (actuarial survival curves were calculated according to the product-limit method of Kaplan and Meier¹⁰; data not shown).

The results of this trial show clearly that postoperative radiotherapy administered directly after quadrantectomy has a protective action against local recurrences and new primary carcinomas. This was not unexpected, but the difference between the outcomes was marked (24 local recurrences and 4 new carcinomas after quadrantectomy alone, as compared with 1 recurrence and no new tumors after quadrantectomy combined with radiotherapy). In the quadrantectomy group, local recurrences occurred mainly in patients under 55 years of age and rarely in patients beyond this age. The presence of an extensive intraductal component was confirmed as an important predictor of the risk of a local recurrence¹¹.

In the quadrantectomy group, the patients who had positive nodes treated with adjuvant therapy and those who had negative nodes did not differ in the incidence of local recurrence. This finding contradicts our previous data, which showed a low rate of recurrence among patients with positive nodes that was ascribed to the protective effect of adjuvant therapy¹². The results of the present study suggest an alternative explanation: patients with positive axillary nodes may be a selected subgroup with tumors that are more radiosensitive than those of patients with negative nodes. This might explain the absence of a difference in the rate of local recurrence in the present trial between node-positive and node-negative patients in the quadrantectomy group, since the node-positive patients received adjuvant therapy but no radiotherapy to the breast.

In fact, the degree of influence of nodal status on the risk of local relapse is uncertain. The many reports in the literature that consider the problem of local recurrence after conservative treatments do not include nodal involvement in their lists of risk factors^13-17. In contrast, according to Rose et al.¹⁸ the use of adjuvant chemotherapy lowered the incidence rate from 15 percent to 4 percent for relapses during the five years after treatment of the primary tumor.

The limited number of local recurrences among postmenopausal women treated with quadrantectomy without radiotherapy may reflect the difference between the structure of the mammary glands in women in this age group and that in young women. After menopause, the complex structure of the mammary gland disappears and the breast is reduced to a fatty organ with scattered islands of fibroepithelial tissue, without connections between them.

This trial confirms that quadrantectomy combined with axillary dissection and radiotherapy is an effective treatment that does not appear to expose patients to a significant risk of local recurrence. An important issue in weighing our findings is the difference between our results and the long-term results of the National Surgical Adjuvant Breast and Bowel Project previously reported by Fisher et al. (trial B-06)¹⁹. In the latter trial, among patients treated with lumpectomy without radiotherapy there was little difference between age groups in the incidence of local recurrence: there were 89 recurrences among the 236 patients 49 years of age or younger (37.7 percent), but 104 recurrences among the 336 patients 50 years of age or older (30.9 percent)¹⁹.

The difference between the data of Fisher et al.¹⁹ and our results may be explained by the difference in surgical procedures. In their study, the operation consisted of a simple excision of the primary carcinoma (lumpectomy) without removal of the skin and the pectoral fascia. In our study, the operation was much more extensive, removing 2 to 3 cm of normal breast tissue around the tumor, the corresponding portion of overlying skin, and the underlying muscular fascia.

Regardless of such differences, the results of both these studies lead us to conclude that in young women, even an extensive surgical resection offers only incomplete protection against local recurrence and that postsurgical radiotherapy is essential. In contrast, in older women a more extensive operation such as quadrantectomy may be sufficient to prevent local recurrence without a need for radiotherapy, whereas this may not be true of a simple operation such as lumpectomy.