Similar long-term results of breast-conservation treatment for Stage I and II invasive lobular carcinoma compared with invasive ductal carcinoma of the breast

The University of Pennsylvania experience

Authors


  • Presented in part at the 45th annual meeting of the American Society for Therapeutic Radiology and Oncology, Salt Lake City, Utah, October 19–23, 2003.

Abstract

BACKGROUND

The objective of the current study was to determine the long-term results of breast-conservation treatment in women with early-stage, invasive lobular carcinoma of the breast.

METHODS

Between 1977 and 1995, 1093 women with Stage I and II invasive ductal carcinoma of the breast and 55 women with invasive lobular carcinoma of the breast underwent lumpectomy, axillary lymph node dissection, and radiation treatment. Overall, 49% of the women received adjuvant systemic therapy (chemotherapy and/or hormones).

RESULTS

The median age was 52 years for patients in the invasive ductal group and 54 years for patients in the invasive lobular group. The median follow-up was 8.7 years and 10.2 years for patients in the invasive ductal and invasive lobular groups, respectively. A comparison of patients who had invasive lobular carcinoma with patients who had invasive ductal carcinoma showed no difference in the 10-year actuarial rates of overall survival (85% vs. 79%, respectively; P = 0.73), cause-specific survival (93% vs. 84%, respectively; P = 0.85), or freedom from distant metastases (81% vs. 80%, respectively; P = 0.76). The 10-year rates of local failure were 18% for patients with invasive lobular carcinoma and 12% for patients with invasive ductal carcinoma (P = 0.24), and the 10-year rates of contralateral breast carcinoma development for the 2 groups were 12% and 8%, respectively (P = 0.40).

CONCLUSIONS

Breast-conservation treatment yielded similar long-term results for women with early-stage, invasive lobular carcinoma and women with the more prevalent invasive ductal carcinoma. Cancer 2005. © 2005 American Cancer Society.

It has been shown that breast-conservation treatment (i.e., lumpectomy followed by breast irradiation) is as effective as mastectomy in the treatment of early-stage breast carcinoma.1–6 However, the data establishing the long-term effectiveness of breast-conservation treatment largely reflect the outcomes of women with infiltrating ductal carcinoma, which represents the large majority of invasive breast carcinomas. In contrast, the information addressing the long-term results of breast-conservation treatment in women with infiltrating lobular carcinoma is limited. The existing literature includes conflicting reports regarding outcomes after breast-conservation treatment for patients who have infiltrating lobular carcinoma compared with women who have infiltrating ductal carcinoma.7–17 Some reports suggest that infiltrating lobular carcinomas are associated with higher local recurrence rates than infiltrating ductal carcinomas and, thus, recommend against breast-conservation treatment for women who have infiltrating lobular carcinoma.7, 8

The objective of the current study was to evaluate the long-term outcomes in a cohort of women with Stage I–II infiltrating lobular carcinoma of the breast after the underwent breast-conserving surgery and received radiotherapy. Information is provided regarding survival, local control, and patterns of failure.

MATERIALS AND METHODS

The study population was derived from women with invasive carcinoma of the breast who received breast-conservation treatment at the Hospital of the University of Pennsylvania between 1977 and 1995. The study was limited to women who presented with unilateral, American Joint Committee on Cancer pathologic Stage I (T1N0M0) or Stage II (T1N1M0, T2N0M0, and T2N1M0) disease.18 Only women who had breast tumors in which the invasive component was either only invasive lobular carcinoma or only invasive ductal carcinoma were considered for analysis. Other eligibility criteria included having undergone a complete macroscopic excision of the primary tumor, an axillary lymph node dissection, and radiotherapy to the whole breast with a boost to the primary tumor bed to a total dose of at least 60 Gray (Gy). Women who had a prior or concurrent malignancy, including a contralateral breast carcinoma, were excluded. The study cohort consisted of 1148 women, including 1093 women with invasive ductal carcinoma and 55 women with invasive lobular carcinoma.

Patients underwent treatment before the era of sentinel lymph node biopsy. In general, axillary lymph node dissection encompassed Level I or Levels I and II. The median number of lymph nodes obtained was 15 (mean, 16 lymph nodes; range, 1–75 lymph nodes).

Radiotherapy to the whole breast was delivered with tangential fields to a median dose of 46.0 Gy over 4.5–5.0 weeks (mean, 46.3 Gy; range, 44.0–52.7 Gy). The radiation energy was generally 6-MV photons, but higher energy photons were used as indicated for women with a large breast size or with a large separation between the medial and lateral chest wall. The dose to the primary tumor bed was boosted using electrons of varying energy in 1007 women (88%), using iridium implants in 130 women (11%), using orthovoltage in 8 women (< 1%), and using photons in 3 women (< 1%). The median boost dose was 18.0 Gy (mean, 16.6 Gy; range, 10.0–22.5 Gy). The clinical decision regarding the total dose was based on multiple factors, including the pathologic status of the final margins from the primary tumor excision, whether or not the patient had undergone a reexcisional biopsy, the presence versus absence of residual tumor at the time of reexcision, and the volume of the boost field.

Regional lymph node irradiation was delivered in 296 of 1148 women (26%) in the overall group. This included 271 of 315 women (86%) in the group with infiltrating ductal carcinoma who had lymph node involvement, 4 of 7 women (57%) in the group with infiltrating lobular carcinoma who had lymph node involvement, and 21 of 778 women (3%) in the group with infiltrating ductal carcinoma who were without lymph node involvement. No women with lymph node-negative lobular carcinoma received lymph node irradiation. The median dose of lymph node irradiation to the supraclavicular field was 46.0 Gy (mean, 45.6 Gy; range, 36.0–50.4 Gy); and, generally, it was prescribed to a depth of 3 cm. A posterior axillary boost field was used in 137 of 296 women (46%) who received lymph node irradiation to bring the midplane axillary dose to approximately 45.0–46.0 Gy. The internal mammary lymph node region was irradiated in 134 of 1148 women (12%), and the median dose was 46.0 Gy (mean, 46.4 Gy; range, 44.0–52.7 Gy).

Overall, 567 of 1148 women (49%) received adjuvant systemic therapy (chemotherapy and/or hormones). This included 298 of 315 women (95%) in the group with infiltrating ductal carcinoma who had lymph node involvement, compared with 4 of 7 women (57%) in the group with infiltrating lobular carcinoma who had lymph node involvement (P = 0.012). For patients with negative lymph node status, systemic therapy was given to 252 of 778 women (32%) with infiltrating ductal carcinoma and to 13 of 48 women (27%) with infiltrating lobular carcinoma. Adjuvant systemic therapy in the invasive ductal group (n = 1093 women) consisted of chemotherapy, hormones, both, or neither in 238 women (22%), 188 women (17%), 124 women (11%), and 543 women (50%), respectively; the corresponding values for the 55 women in the group with invasive lobular carcinoma were 2 women (4%), 11 women (20%), 4 women (7%), and 38 women (69%), respectively. The chemotherapy regimens employed consisted of cytoxan, methotrexate, and 5-fluorouracil in 271 women (24%); doxorubicin-based regimens in 76 women (7%); and other regimens in 21 women (2%). Tamoxifen was used in 214 of 327 women (65%) who received hormone therapy, and the form of hormone therapy used was not recorded in 107 women (33%).

A local failure was defined as a recurrence of carcinoma that developed within the treated breast at any time after treatment, regardless of regional or distant disease status. The location of local failure was scored according to the method reported by Recht et al.19 A regional failure was defined as a failure that occurred in the ipsilateral axillary, supraclavicular, infraclavicular, and/or internal mammary lymph node regions. Distant (metastatic) failure was defined as a failure that was beyond local or regional disease. A contralateral breast carcinoma was defined as the diagnosis of any invasive breast carcinoma or ductal carcinoma in situ within the opposite breast.

For overall survival, a patient was scored as a failure at the time of death, regardless of the cause of death. For cause-specific survival, a patient was scored as a failure at the time of death only if the cause of death was attributable to breast carcinoma. For analysis of distant failure, a patient was scored as a failure at the time of the first evidence of distant metastatic disease.

The median follow-up for all 1148 women was 8.8 years (mean, 9.0 years; range, 0.0–24.0 years). The median follow-up for the 905 surviving women was 9.3 years (mean, 9.5 years; range, 0.0–24.0 years). The median follow-up was 10.2 years for the 55 women with invasive lobular carcinoma and 8.7 years for the 1093 women with invasive ductal carcinoma (P = 0.14).

Actuarial curves were calculated using the Kaplan–Meier method20 and were compared using the log-rank test. To identify factors associated with outcome and control for potential confounding, both stratified analyses and Cox proportional hazards models were applied for the multivariate analyses. The log-rank test was used for statistical comparisons between curves.21 The statistical assumptions associated with those procedures were confirmed by graph and statistical testing.

RESULTS

Table 1 shows the patient characteristics by histologic type. The median patient age at diagnosis was 54 years (mean, 55 years; range, 33–74 years) in the lobular carcinoma group and 52 years (mean, 53 years; range, 21–89 years) in the ductal carcinoma group. Compared with invasive ductal carcinoma, invasive lobular carcinoma was associated with an increased incidence of tumors detected on physical examination (42% vs. 24%, respectively; P = 0.03), an increased incidence of pathologically negative lymph node disease (87% vs. 71%, respectively; P = 0.03), a decreased use of adjuvant systemic chemotherapy (11% vs. 33%, respectively; P = 0.001), and an increased use of reexcision (73% vs. 50%, respectively; P = 0.001). A trend suggesting a slightly higher proportion of tumors with positive estrogen receptor (ER) status was noted in the lobular carcinoma group compared with the ductal carcinoma group (60% vs. 55%, respectively; P = 0.08). No other significant differences were identified between the two groups (all P ≥ 0.29).

Table 1. Patient and Tumor Characteristics
CharacteristicInvasive lobular carcinomaInvasive ductal carcinomaP value
No. of patients%No. of patients%
  1. cGy: centigray.

Age at diagnosis    0.29
 <39 yrs5916415 
 40–49 yrs112029227 
 50–59 yrs213829627 
 60–69 yrs112023321 
 ≥ 70 yrs71310810 
Race    0.65
 White488791083 
 African-American71317516 
 Other0081 
Method of detection    0.03
 Physical examination234226524 
 Mammography112026925 
 Both213855050 
 Unknown0091 
Pathologic tumor classification    0.49
 T1437880974 
 T2122228426 
Pathologic lymph node status    0.03
 N0488777871 
 N1 (no. of positive lymph nodes)    
 1–361123522
 ≥ 412807 
Estrogen receptor status    0.08
 Positive336060455 
 Negative5923021 
 Not done/unknown173125924 
Progesterone receptor status    0.36
 Positive264747143 
 Negative91627125 
 Not done/unknown203735132 
Chemotherapy    0.001
 Yes61136233 
 No498973167 
Hormone therapy    0.45
 Yes152731229 
 No407378171 
Any systemic therapy (lymph node positive patients)    0.012
 Yes45729895 
 No343175 
Reexcison    0.001
 Yes407354750 
 No152754650 
Final margins of resection    0.78
 Negative285158754 
 Positive81512011 
 Close (≤ 2 mm)71311611 
 Unknown122227025 
Total dose to tumor bed (cGy)    0.45
 6000–6200234245341 
 6201–6400173142539 
 6401–6600142519318 
 > 660012222 

Fourteen of 55 patients (25%) in the lobular carcinoma group and 229 of 1093 patients (21%) in the ductal carcinoma group died during their follow-up. There were 12 distant failures (22%) among the patients with lobular tumors and 196 distant failures (18%) among the patients with ductal tumors. Two patients (4%) with lobular tumors and 33 patients (3%) with ductal tumors failed in regional lymph nodes. Both of the regional failures in the lobular group were associated with simultaneous local recurrences and occurred during the first 5 years of follow-up. At presentation, both patients had negative lymph node status, and one patient had tumor involvement of the final margins of resection. Six patients (11%) with lobular tumors and 78 patients (7%) with ductal tumors developed contralateral breast carcinoma. There were 9 local failures (16%) among the patients with lobular tumors, including 5 failures that were true recurrences or marginal misses, 1 recurrence that occurred elsewhere in the breast, 2 recurrences that were located in the skin, and 1 recurrence with an unknown location. There were 118 local failures (11%) among the patients with ductal tumors, including 78 that were true recurrences or marginal misses, 31 recurrences that occurred elsewhere in the breast, 5 recurrences that were located in the skin, and 4 recurrences with unknown locations.

The curves illustrating the rates of local failure as a function of time for both groups are shown in Figure 1. The 5-year and 10-year local failure rates in the invasive lobular carcinoma group were 14% and 18%, respectively, compared with 6% and 12% in the invasive ductal carcinoma group (P = 0.24). After salvage treatment for a local recurrence, 73 of 118 patients (62%) with invasive ductal carcinoma were alive without evidence of disease, and 40 of 118 patients (34%) had died either of disease or of unknown causes. Among the patients with invasive lobular carcinoma, after salvage therapy for local recurrence, 4 of 9 patients (44%) were alive without evidence of disease, and 4 of 9 patients (44%) had died either of disease or of unknown causes, and this difference in salvage rates was not statistically significant (P = 0.60).

Figure 1.

Local failure curves for patients with invasive lobular carcinoma and invasive ductal carcinoma.

The curves illustrating the rates of contralateral breast carcinoma as a function of time for both of the groups are shown in Figure 2. The 5-year and 10-year rates of contralateral breast carcinoma in the invasive lobular carcinoma group were 4% and 12%, respectively, compared with 3% and 8% in the invasive ductal carcinoma group (P = 0.40).

Figure 2.

Contralateral breast carcinoma curves for patients with invasive lobular carcinoma and invasive ductal carcinoma.

Table 2 lists 5-year and 10-year actuarial outcome data for each of the 2 histological groups as well as the results of statistical comparisons between the groups. In the invasive lobular carcinoma group, the 5-year and 10-year overall survival rates were 92% and 85%, and the corresponding cause-specific survival rates were 98% and 93%, respectively. In the invasive ductal carcinoma group, the 5-year and 10-year overall survival rates were 90% and 79%, and the corresponding cause-specific survival rates were 92% and 84%, respectively. There were no statistically significant differences between the groups for the analyzed outcome parameters (all P ≥ 0.23).

Table 2. Actuarial Outcome Data
OutcomeInvasive lobular carcinomaInvasive ductal carcinomaP-value
Five-yr rate % (95% CI)Ten-yr rate % (95% CI)Five-yr rateTen-yr rate
  1. 95% CI: 95% confidence interval.

Survival     
 Overall survival92 (80–97)85 (70–92)90 (88–92)79 (76–81)0.73
 Cause-specific survival98 (87–100)93 (79–98)92 (90–94)84 (81–87)0.85
Freedom from distant metastases94 (83–98)81 (64–90)87 (85–89)80 (77–83)0.76
Local failure14 (7–28)18 (9–33)6 (4–7)12 (10–14)0.24
Contralateral breast carcinoma4 (1–14)12 (5–26)3 (2–5)8 (6–10)0.40

Due to the statistical difference in lymph node status and chemotherapy use between the patients with invasive ductal carcinoma and invasive lobular carcinoma and the overall small number of tumors with positive lymph node status the lobular carcinoma group, to confirm that the survival outcomes were comparable, a separate analysis of the patients with negative lymph node status was performed, adjusting for chemotherapy use. This analysis also showed no statistical difference in outcomes between the invasive ductal carcinoma group and the invasive lobular carcinoma group (data not shown; all P ≥ 0.34).

DISCUSSION

Results from the analyses of the current study generated several important observations regarding outcomes for patients with early-stage invasive lobular carcinoma. First, invasive lobular carcinoma of the breast differs significantly from invasive ductal carcinoma in several aspects of its presentation. Second, despite these differences and their impact on initial management, breast-conservation treatment for invasive lobular carcinomas yields similar rates of local-regional control compared with invasive ductal carcinoma. More specifically, the survival and local control afforded by breast-conservation treatment in patients with invasive lobular tumors do not differ statistically from those achieved in patients with invasive ductal tumors. Finally, the risk of developing contralateral breast carcinoma among patients with invasive lobular carcinomas did not differ significantly from the corresponding risk among patients with tumors of invasive ductal histology.

In this study, we observed that invasive lobular tumors were less likely to be detected solely on the basis of mammographic abnormalities and were more likely to be detected with physical examination findings compared with invasive ductal tumors (Table 1). Other reports have noted this pattern in the presentation of invasive lobular carcinomas. In one study, invasive lobular tumors were more likely either to be occult mammographically or to be associated with subtle mammographic abnormalities compared with their ductal counterparts.9 Two other studies found that invasive lobular carcinomas were more difficult to visualize mammographically than ductal carcinomas.10, 22 These observations may stem from the fact that invasive lobular carcinomas spread through the breast in relatively small clusters of cells without eliciting a prominent desmoplastic reaction from the surrounding tissues.23

The current series showed that a significantly higher proportion of patients in the lobular carcinoma group underwent reexcision of the primary tumor compared with patients in the ductal carcinoma group (Table 1). Although the status of the final margins of resections did not differ significantly between the two groups, the higher reexcision rate among patients with lobular tumors suggests that the process to achieve negative margins in these patients is more difficult. These findings are consistent with the results from other studies and may be explained in part by the difficulty with mammographic visualization.9, 10, 22 These findings suggest that, if reexcision is used appropriately, based on factors such as careful assessment of margin status, histologic findings, and details of the initial surgical procedure, then the higher rate of positive margins reported in other studies can be circumvented.9, 10, 22, 24

In this study, breast-conservation treatment in patients with early-stage, invasive lobular carcinoma yielded high rates of survival and local-regional control after treatment (Table 2). These results are comparable to the results from other series that reported the long-term results of breast-conservation treatment for invasive carcinomas, including invasive lobular carcinoma, of similar disease stage.11, 25–30 Overall survival and cause-specific survival did not differ significantly between the two histologic groups, whether all patients or only the lymph node-negative patients were compared. Multivariate analyses of the factors associated with an increased risk of either local recurrence or distant metastases, controlled for lymph node status and adjusted for chemotherapy use, showed no statistical difference in the hazard risk for either local recurrence or distant disease-free survival between the two histologic subtypes (all P ≥ 0.34). A small difference in local recurrence or distant metastatic rates cannot be excluded due to the small total numbers of patients and the small overall numbers of events in the invasive lobular carcinoma group, as well as the diminishing numbers of patients at risk at longer time points, problems that are intrinsic to the outcomes analyses of uncommon subtypes of invasive breast carcinoma. In the current study, a higher absolute rate and an increased hazard for local recurrence were observed in the invasive lobular carcinoma group, although the difference was not statistically significant. However, this study represents one of the largest series to date to report long-term outcomes in patients with invasive lobular breast carcinoma. The detection of small differences in local recurrence or survival outcomes, if any, will require the accumulation of additional patients and longer follow-up.

The lack of a statistical difference in terms of survival between invasive lobular and invasive ductal tumors has been observed by other investigators.9–17 However, one study reported that invasive lobular tumors may be associated with better survival rates than invasive ductal tumors.31 The possible survival advantage favoring lobular tumors may be explained in part by the fact that patients who have these tumors present less often with positive lymph nodes and more frequently have positive ER status compared with their counterparts with invasive ductal tumors (Table 1). These observations are consistent with other reports in the literature.11, 15, 22, 31

Regarding the probability of local control after treatment, patients with invasive lobular carcinoma were similar to those with invasive ductal carcinoma (Table 2, Fig. 1). The lack of a difference in local control between both histologic groups as well as the predominance of local failures at or near the primary tumor bed has been described previously.9–17 However, some investigators have noted an increased risk of local failure after breast-conservation treatment for patients with invasive lobular carcinoma.7, 8 Local failures were observed most commonly within (true recurrence) or just outside (marginal miss) the primary tumor bed. The observation that local failures occurred most commonly within or at the margins of the initial tumor bed suggests that whole-breast irradiation is capable of eradicating the more widely distributed, multicentric, microscopic tumor foci that have been described in some studies of invasive lobular carcinomas.7, 32–36

There has been significant progress in the screening, assessment, and treatment of breast carcinoma since the majority of the patients in this study were treated. Although radiation treatment practices did not change substantially during the study period and do reflect conventional breast tangential-field therapy, other areas of management have evolved significantly. Imaging modalities to delineate the extent of tumor better include digital as well as film screening mammography with magnification views, directed ultrasound, and magnetic resonance imaging when other imaging results are equivocal or are difficult to interpret. These improvements in radiographic assessment help guide the surgical excision with increased precision. Current practices minimize the number of patients with positive or unknown margin status who receive breast-conservation therapy at all. Meticulous pathologic assessment of margins and the use of reexcision for all patients who have positive margins or, in some patients, close margins are performed routinely. In the current study cohort of women with invasive lobular or ductal carcinoma, 15% and 11%, respectively, had positive margins, and 22% and 25%, respectively, had no assessment of the margins, representing a substantial proportion of patients whose local recurrence risk would be reduced using current radiologic, pathologic, and surgical techniques. In the treatment of invasive lobular carcinomas, utilizing the standards of meticulous margin assessment, the use of reexcision for close or positive margins and adequate radiation doses produced local control rates comparable to those achieved in the treatment of invasive ductal carcinomas; thus, breast-conservation therapy is a therapeutic choice for appropriately selected patients.

Systemic therapy is recommended currently for the vast majority of patients with lymph node-positive breast disease, regardless of invasive histology, and for many premenopausal patients. In the current series, whereas 95% of patients with lymph node-positive, invasive ductal carcinoma received systemic therapy, only 57% patients (4 of 7 patients) with lymph node-positive, invasive lobular carcinoma received chemotherapy, although the impact on outcomes of this omission is difficult to assess given the very small numbers of patients with lymph node-positive tumors in this group. In this series, among the invasive ductal carcinoma group, 19% of lymph node-negative, premenopausal women received chemotherapy, and 6% of pre-menopausal patients (1 of 16 patients) in the invasive lobular carcinoma group also received chemotherapy, a rate that is below current standards. Indications for the use of systemic treatment have changed over the years of the study in the invasive ductal group, and its use has increased over time. Due to too few numbers of patients per year, it was not possible to reach any conclusions regarding these patterns in the invasive lobular group. Systemic hormone therapy is recommended current for all women with ER-positive tumors, but the current series showed that, over the study period, ER status was not assessed in 31% of women in the lobular group and in 24% of women in the ductal group. Among the patients who had tumors with known ER-positive status, only 40% in both groups received any systemic hormone therapy. Both the assessment of ER status and the use of hormone therapy increased during the study period. The increasing use of systemic therapy is likely to result in further improvements in the local control and survival rates observed in both groups in the current study.

Although some reports have suggested that the incidence of subsequent contralateral breast carcinoma is higher in patients with lobular tumors, this observation has not been confirmed uniformly.13, 32, 34, 37–39 In the current study, the risk of developing contralateral breast carcinoma among patients with invasive lobular carcinoma of the breast did not differ significantly from the corresponding risk among patients with invasive ductal carcinoma (Table 2, Fig. 2). These observations suggest that continued, long-term surveillance of the contralateral breast is warranted for women who have either type of invasive carcinoma.

In conclusion, results of the current study demonstrated that breast-conservation treatment in patients with early-stage, invasive, lobular breast carcinoma provides high rates of survival and local-regional control that do not differ statistically from the results achieved in patients with invasive ductal tumors. Due to the small numbers of patients in the invasive lobular group and the very small numbers of events, a small difference in the local recurrence rates cannot be excluded. The results observed in the current study support the routine use of breast-conservation treatment for patients with invasive lobular carcinoma of the breast.

Ancillary