Return to Table of Contents
Primary therapy for breast cancer has changed substantially in recent years. Breast-conserving surgery (BCS), followed by radiation therapy, was recommended by the 1990 National Institutes of Health (NIH) Consensus Conference (1) as a safe alternative to mastectomy for most women with stage I and II breast cancer. Since 1988, the California Cancer Registry (CCR) has collected statewide information on cancer-directed surgical procedures performed as part of the first course of treatment for women with breast cancer. This chapter focuses on patterns of surgical treatment, particularly BCS, in California from 1988 to 1992. Information is presented by stage at diagnosis, race/ethnicity, age at diagnosis, reporting region, and indicators of socioeconomic status in the patient's immediate neighborhood. Temporal trends of BCS utilization and radiotherapy are evaluated. Age-specific and race/ethnic differences in BCS utilization are further addressed.
The NIH Consensus Conference recommendation for BCS is based on the American Joint Committee on Cancer definition of stage I and II breast cancer (2), which includes tumors up to 5 cm in diameter, with or without extension to axillary lymph nodes on the same side. The classification for stage at diagnosis adopted by the CCR follows the SEER summary stage at diagnosis, which does not evaluate tumor size (3). Therefore, the classification "localized" in this chapter may include tumors larger than 5 cm in diameter, but excludes tumors with any evidence of lymph node involvement. Please refer to Chapter 7 for a more detailed definition of SEER summary stage at diagnosis.
In the following analyses, early stage breast cancer is defined as in situ or localized invasive tumors. Surgical procedures and radiotherapy codes follow the California Cancer Reporting System Standards (4). Breast-conserving surgery (BCS) includes partial or segmental mastectomy, quadrantectomy, tylectomy, wedge resection, nipple resection, lumpectomy, or excisional biopsy, with or without dissection of axillary lymph nodes. Mastectomy includes subcutaneous, total, simple, modified, radical, and non-specified breast removal procedures. Information on breast reconstruction following mastectomy was available only if such procedure was performed as part of the planned first course of treatment. Radiation includes administration of radioactive beams, implants, or radioisotopes. A trend towards earlier diagnosis of breast cancer during the five years of the existence of the CCR has been described in a previous chapter of this report, particularly in women over the age of 50. Because tumors diagnosed at an earlier stage tend to be smaller, and therefore more likely to be considered for BCS, temporal trends were also examined restricting the data to women with in situ or localized invasive tumors up to 2 cm in diameter. Temporal trends and possible associations between factors were examined with the chi-square test. A p-value smaller than 0.05 was considered statistically significant.
From 1988 to 1992, 97,032 cases of breast cancer with known age were diagnosed in California women (5). This chapter excludes 708 cases which were reported by autopsy or death certificate only. Detailed information on surgical procedures during the first course of treatment for these women is shown in Table 8.1. Overall, mastectomy was the most common procedure for women with breast cancer diagnosed from 1988 to 1992. A total of 61,336 (63.7%) women received a mastectomy, of which 5,234 (8.5%) had a planned breast reconstruction as part of the first course of treatment. Mastectomy was performed in 48.8% of in situ tumors, 63.8% of localized tumors, 79.4% of regional tumors, and 34.5% of remote stage (metastatic) breast cancer. BCS was performed in 30,450 (31.6%) women with breast cancer: 5,594 with in situ tumors, 18,243 with localized tumors, 4,217 with regional tumors where there was evidence of lymph node involvement but no direct extension to surrounding tissues, and 2,396 with more advanced or unstaged disease. These figures correspond to 48.9%, 35.4%, 18.2%, and 22.5% of women with in situ, localized, regional (lymph node extension only), and later stage breast cancer, respectively. Among women receiving BCS, lymph node dissection was performed in the majority of women with localized (74.9%) and regional (91.6%) breast cancer, but less often among women with in situ tumors (14.5%), for a total of approximately 63% of all cases receiving BCS.
During the five-year period covered in this report, there were significant race/ethnic differences in utilization of BCS (Figure 8.1 and Table 8.2). The proportion of Asian/Other women receiving BCS, with either in situ or localized tumors, was lower than in any other race/ethnicity (p-values < 0.001 when compared to non-Hispanic white, black, and Hispanic women). Among women with in situ tumors, no other race/ethnic differences in BCS utilization were statistically significant. Among women with localized tumors, Hispanics were less likely to be treated with BCS than non-Hispanic white or black women (p-values < 0.001). There were no significant differences in BCS utilization between non-Hispanic white women and black women.
Information on women receiving BCS, mastectomy, or other procedure/no surgery, by stage at diagnosis, is presented by ten-year age categories in Table 8.3. These data are summarized by further categorizing age at diagnosis to under 65 years or 65 years and older. These two age groups were selected to reflect a woman's eligibility for Medicare coverage. Among women of each race/ethnicity with in situ tumors, both age groups were equally likely to receive BCS (Figure 8.2). However, among non-Hispanic white women with localized tumors, those under 65 were significantly more likely to receive BCS than women 65 years and older (p-value < 0.001). Age differences in BCS utilization for other race/ethnicities were not significant.
The proportion of women with in situ or localized breast cancer who received BCS has increased substantially in California over time. In 1988, BCS was elected by 44.8% and 28.3% of women with in situ and localized disease, respectively. By 1992 these percentages had risen to 55.1% and 43.1% (Figure 8.3). Trends were statistically significant for both in situ and localized tumors (p-values < 0.001), but the largest increase was observed for localized disease: in 1992, women with localized tumors were almost twice as likely to receive BCS than those diagnosed in 1988.
The proportion of women with in situ or localized tumors diagnosed from 1988 to 1992 who received BCS, categorized by race/ethnicity and year of diagnosis, is shown in Table 8.4. Overall, there was a substantial increase in BCS utilization for these women, particularly after 1989. BCS was elected by 31.1% of women with early stage breast cancer in 1988. and by 33.0% in 1989. This percentage rose to 33.0%, 37.4%, 41.0% and 45.4% in 1989, in 1990, 1991, and 1992, respectively.
Figure 8.4 shows BCS utilization by stage, year and age at diagnosis (under 65 years or 65 years and older). Trends of increased BCS utilization among women with either in situ or localized tumors were statistically significant in both age groups (p-values for both age groups < 0.001).
The proportion of women receiving BCS following the diagnosis of in situ tumors has increased in all race/ethnicities (Figure 8.5), but the trend was significant only among non-Hispanic white and Hispanic women. Among women diagnosed with localized tumors, the trend for BCS utilization is evident and statistically significant for all race/ethnicities.
Trends for BCS utilization among women with early stage breast cancer (in situ or localized tumors combined), in either age-group (< 65, 65) were statistically significant for all four race/ethnicities (Figure 8.6). Among those 65 years and older, the largest increase in BCS utilization was observed among Asian/Other women, who were almost three times more likely to receive BCS in 1992 than in 1988.
The recommended technique for BCS is the total removal of the primary tumor with clear resection margins (1). Women with small tumors are, therefore, more likely to be considered eligible for BCS. To better evaluate trends for surgical treatment, an analysis of BCS utilization among women with small ( 2 cm) in situ or localized tumors was performed. In addition, the analysis was restricted to the first occurrence of primary breast cancer for each woman. Tumor size was recorded for 74.6% of cases.
The analysis of data from 30,975 women with a first occurrence of primary in situ or localized breast cancer with tumor size up to 2 cm, showed similar findings to those discussed above. (Table 8.5). Trends for increased BCS utilization were significant for both in situ or localized tumors, under and over 65 years, and in all race/ethnicities. When BCS utilization by race/ethnicity and age at diagnosis was examined for women with either in situ or localized tumors, a trend towards BCS was also observed for women of all race/ethnicities and age groups, except for black women under 65 years, for whom such trend was not statistically significant. It is worth noting that, overall, the proportion of women of all races receiving BCS increased when only small tumors were taken into account. However, information on tumor size was missing in 25% of cases, and among those with known size, 32% of tumors were larger than 2 cm. As the number of cases in the restricted analysis was reduced by approximately two thirds, rates of BCS utilization among minority women showed considerable fluctuation, and trends stratified by race/ethnicity, age, and stage at diagnosis were accordingly less evident.
BCS increased among women with early stage breast cancer (in situ or localized tumors) in all ten cancer reporting regions of California from 1988 to 1992, but there was substantial regional variation in BCS utilization (Table 8.6, and Figure 8.7). BCS utilization was consistently higher in the San Francisco Bay Area than any other region in California (52.5% of women with early stage breast cancer in 1992). On the other hand, although the Central Valley had the lowest BCS utilization in the state during the entire period, there was nearly a three-fold increase from 13.7% in 1988 to 36% in 1992. Table 8.7 shows the percent of women treated with BCS from 1988 to 1992, by stage at diagnosis, in the ten cancer reporting regions in California.
Radiation therapy following BCS is recommended for women with early stage invasive breast cancer (1). The indication for radiotherapy in women with in situ tumors seems to be less clear (6), and is the object of randomized clinical trials currently in progress. Statewide, 68% of 18,243 women with localized breast cancer receiving BCS were also treated with radiation. The proportion of women receiving radiotherapy following BCS was higher among younger patients (76.9% of those under 65 versus 57.8% of those 65 and older). Radiotherapy utilization has not increased from 1988 through 1992. During the period, however, there was considerable regional variation, with some regions showing an increase and others a decrease (Figure 8.8). In 1992, the Bay Area and Sacramento Regions had the highest proportions of women receiving radiotherapy following BCS (79.4% and 78.8%, respectively). Radiotherapy utilization following BCS in 1992 were lowest in the Central Valley (54.6%), and in Los Angeles County (59.2%).
Radiotherapy utilization among patients with localized breast cancer receiving BCS, has fluctuated substantially by race/ethnicity and age at diagnosis (Figure 8.9). With the exception of non-Hispanic white women 65 years and older (p-value < 0.001), there was no evidence of a significant increase in the administration of radiotherapy subsequent to BCS from 1988 to 1992. However, utilization of radiotherapy may be underreported.
The socioeconomic status of women with in situ or localized breast cancer was characterized based on the profile of the census block group (a more homogeneous subdivision of a census tract with an average of 1,000 residents). This methodology has limitations, in that census group information is applied to individuals. However, the census-based methodology has been widely used and was validated in a recent study as a reasonable approach to overcome the absence of socioeconomic data in most medical records (7).
Income was classified based on the distribution of median household income in all block groups within each of the ten reporting regions in California. The criteria used to characterize patients as living in a low, medium, or high income neighborhood were as follows: a block group was considered to have low income if its median household income was within the 25% lowest block group incomes in that reporting region, high income block groups were those with median income within the 25% highest block group incomes in that region, and medium income block groups were those with median incomes falling between these two extremes. A patient was considered to live in a neighborhood with less formal education if no more than 25% of persons over 25 years of age living in that block group had a high school diploma. Sufficient information to determine the census block group was available for 93.5% of women with early stage breast cancer. Between 1988 and 1992, 27.1% of women with early stage breast cancer lived in neighborhoods with less formal education. During the same period, 32.7%, 50.1%, and 17.2% of these women were classified as living in high, medium, and low income neighborhoods, respectively.
BCS utilization by high school education and median household income in the immediate neighborhood is presented in Table 8.8. Overall, 33.8% of women with early stage breast cancer from low income neighborhoods received BCS, compared to 37.2% and 41.6% among medium and high income neighborhoods, respectively. A significant association between neighborhood income and BCS utilization was noticed in all race/ethnicities except among black women, for whom BCS rates declined, although not significantly, among those living in high income areas (Figure 8.10). Asian/Other women had the lowest BCS utilization in all three income categories. In medium and low income neighborhoods, black women were more likely to receive BCS than any other race/ethnicity, although the difference between black and white women in medium income areas was only marginally significant. In high income neighborhoods, however, white women were more likely to be treated with BCS.
Highly significant temporal trends for increased utilization of BCS in women with early stage breast cancer (Figure 8.11) were detected in all three income categories (low, medium, or high; p-values < 0.001). BCS utilization among women in high income neighborhoods was significantly higher than among women in medium or low income neighborhoods, and was significantly higher in medium than in low income neighborhoods (all p-values <0.001).
Women living in neighborhoods with less formal education were significantly less likely to receive BCS than women from more educated neighborhoods (33.0% and 39.9%, respectively) (Table 8.8). Similar to what was observed concerning median household income, Asian/Other women were the least likely to receive BCS in both less educated and more educated neighborhoods (Figure 8.12). Black women living in less educated neighborhoods were significantly more likely to be treated with BCS than either Hispanic or white women. In the more educated neighborhoods, there were no significant differences in BCS utilization among black, Hispanic, or white women. Increasing utilization of BCS in women with early stage breast cancer was evident for both levels of high school education (Figure 8.13).
During the period examined in this report, 1988 to 1992, there were differences in BCS utilization among California women with breast cancer. Overall, BCS was less likely to be performed in Asian/Other women than in women of other race/ethnicity, regardless of age and stage at diagnosis. BCS was also less likely among women residing in low income or less educated neighborhoods. Among women with in situ tumors, age and race/ethnicity (except for Asian/Other) were not a significant factor for BCS utilization. Age and race/ethnic differences were more evident among women with localized tumors. Despite these variations, there was a clear trend in all reporting regions in California for increased utilization of BCS following the diagnosis of in situ or localized breast cancer. Trends were significant for all race/ethnicities, age groups (under and above 65), and socioeconomic status.
Return to Table of Contents
Return to Chapter 7
Go to Table 8.1
Go to Chapter 9