Litvak DA, Arora R. Treatment of Elderly Breast Cancer Patients in a Community Hospital Setting. Arch Surg. 2006;141(10):985-990. doi:10.1001/archsurg.141.10.985
Elderly breast cancer patients are underdiagnosed and undertreated in a community hospital setting.
Retrospective study of tumor registry patients.
A total of 354 breast cancer patients aged 70 years or older diagnosed between 1992 and 2002 were identified from the tumor registry. The data for 3 prospectively chosen age subgroups (aged 70-74, 75-79, and ≥80 years) were subsequently analyzed.
Main Outcome Measures
Data were collected regarding the method of detection of cancer, tumor characteristics and staging, type of treatment, and patient comorbidities. Comparisons between age subgroups were made using a likelihood ratio χ2 test.
Forty-six percent of all patients presented with palpable breast cancer. Mammography was frequently used as an adjunct to the physical examination but was helpful in detecting occult disease in only 54% of all patients and 38% of patients older than 80 years. Although more than 70% of patients were considered stage I or II, complete pathologic staging was only performed in 64% of patients. Approximately 50% of patients were treated with breast-conservation surgery; however, adjuvant radiation therapy was omitted in 55% of these patients. Additionally, only 29% of patients with positive lymph nodes received chemotherapy, and 67% patients who were estrogen receptor–positive received adjuvant hormonal therapy.
Older breast cancer patients are more likely to be diagnosed and staged clinically. Mammography is underused as a screening modality. While breast-conservation surgery was performed in about half of the patients, adjuvant radiation, chemotherapy, and hormonal therapies were frequently excluded.
As the proportion of people of advanced age in the United States has increased over time, so has the number of elderly patients with breast cancer. In fact, approximately 50% of breast cancer patients are older than 65 years and 35% of patients are older than 70 years.1,2 However, the most appropriate management of elderly patients with breast cancer remains difficult to define. As patients age, not only does the incidence of cancer increase, but also does the probability that patients will have significant medical comorbidities. The average patient older than 70 years with cancer will have 3 or 4 comorbidities and a greater than 30% chance of at least 1 of these conditions being severe.3 Assessing the risk to the patient of the breast cancer vs the risk of any concomitant medical comorbidity can be challenging to the most experienced clinician. Thus, it has been difficult for the medical community to reach a consensus on the best treatment of older cancer patients. Unfortunately, this lack of consensus may have let fit older breast cancer patients be denied standard therapies such as breast-conservation surgery (BCS), adjuvant radiation therapy, axillary lymph node dissection (ALND), or chemotherapy solely because of advanced age.3- 7 Moreover, patients older than 70 years generally have been intentionally excluded from clinical trials that have led to standard recommendations for the treatment of younger breast cancer patients. Collectively, studies suggest that older breast cancer patients are more likely to be undertreated compared with their younger counterparts.
There is a similar degree of uncertainty in what should be considered appropriate screening for breast cancer in older patients. There have been few retrospective, let alone prospective, studies looking at cancer screening in older patients. This has led to conflicting recommendations from agencies such as the United States Preventative Services Task Force and the American Cancer Society in terms of breast cancer screening for elderly women. Currently, these recommendations range from an upper age limit of 70 years to no upper age limit for screening mammography.8,9 This uncertainty stems from the difficulty in assessing the benefit of screening mammography (ie, the benefit of early detection of breast cancer) and the risk of dying from either breast cancer or other medical causes for older women. Again, studies suggest that breast cancer is underdiagnosed in older women and may result in a more unfavorable stage of disease at the time of diagnosis.9,10
The majority of cancer patients are treated locally in community hospitals. The studies evaluating the diagnosis and treatment of elderly cancer patients generally have used large national cancer databases, such as Surveillance, Epidemiology, and End Results programs, or have been from tertiary referral institutions.3,4,6,10- 13 Although these databases include information on patients treated in community hospitals, the treatment of elderly cancer patients in this setting remains unclear. Considering that 77% of the deaths due to breast cancer are in women older than 55 years,3 the improved treatment of older patients with breast cancer should be of paramount interest to the health field. The purpose of the present study was to determine whether elderly breast cancer patients are underdiagnosed and undertreated in a community hospital setting.
We used a tumor registry database that was maintained per the recommendations of the Committee on Cancer of the American College of Surgeons at a community hospital. This database was used to identify 354 breast cancer patients older than 70 years diagnosed between 1992 and 2002. Three prospectively chosen subgroups of these patients (aged 70-74, 75-79, and ≥80 years) were subsequently analyzed. The following information for each subgroup was retrieved from the registry: size, grade, and histologic type of tumor; estrogen receptor (ER) status; type of surgical procedure; type of adjuvant therapy (chemotherapy, radiotherapy, hormonal therapy); mammogram; presence of a palpable mass; patient comorbidity; complete staging (ie, pathologic staging of tumor and lymph nodes). We used a likelihood ratio χ2 test to compare patient characteristics and treatments between patient subgroups. A P value of less than .05 was considered significant. Approval by the requisite institutional review boards was obtained for this study.
The characteristics of the 354 patients older than 70 years diagnosed with breast cancer between 1992 and 2002 identified from the tumor registry are outlined in Table 1. The median tumor size, stage of disease, histologic type, and grade were the same in each of the 3 age groups studied. The majority of patients were stage I or II according to standards by the American Joint Committee on Cancer. However, complete staging (eg, lymph node evaluation) was omitted in 36% of all patients and was more commonly omitted in women older than 80 years (56%; P<.001). Sentinel lymph node biopsy was used in only 5 patients (1.4%; data not shown). Forty-five percent of tumors expressed ERs, but ER status was unknown in 38%. Seventy-two percent of all patients underwent mammography, but only 60% of patients older than 80 years did so (P = .004). More importantly, mammography was largely used as an adjunct to the physical examination; mammography detected occult disease in only 64% of all patients and 38% of patients aged 80 years and older (P<.001). Thus, 46% of patients presented with a palpable mass, and this rate was significantly higher in the oldest age group (62%). Breast-conservation surgery (ie, lumpectomy) was performed in 47% of all patients and 48% of patients aged 80 years or older. Adjuvant therapies such as radiation and hormonal therapy were used in 32% and 31% of all patients, respectively. Only 12% of all patients received adjuvant chemotherapy. In the older age groups, only 5% to 10% of patients received chemotherapy, which was significantly less than the 19% of patients aged 70 to 74 years (P = .007). Medical comorbidity was identified in 16% of patients while 47% of patients had no other comorbidities; comorbidity data were missing in 36% of patients. As expected, there was a significant increase in the percentage of patients with comorbid medical problems in patients older than 74 years compared with patients aged 70 to 74 (P = .001). Nonetheless, the majority of patients of any age group (with data in this field) had no comorbid disease.
The type of therapy, if any, adjuvant to either mastectomy or lumpectomy for all patients was analyzed from the tumor registry database (Table 2). Radiation followed mastectomy 18% to 23% of the time in the 3 groups while hormonal therapy was used as an adjunct to mastectomy 28% to 32% of the time. Mastectomy was less likely to be followed by any form of adjuvant treatment in women older than 75 years (43%) than in women younger than 75 years (74%) (P = .01). The oldest patients (8%) were less likely to receive chemotherapy than the other age groups after mastectomy (P = .02). Similar trends were seen in older women treated with lumpectomy. The oldest women were less likely to be treated with any form of adjuvant therapy (58% vs 73%-84%; P<.001), especially chemotherapy, compared with the 2 other groups of women (0% vs 10%-23%; P = .001).
Curiously, women aged 75 to 79 years were more likely to receive radiation following lumpectomy (61%) compared with women aged 70 to 74 years (37%) or those aged 80 years or older (40%) (P<.001). Taken together, radiation was omitted from BCS in 55% of patients.
We also noted omissions of treatment in patients with indications for adjuvant treatment, such as positive lymph nodes, ER-positive tumors, and patients with positive surgical margins after resection (Table 3). All together, 70 patients (20%) had positive lymph nodes. Of these patients, 29% received chemotherapy. Moreover, only 17% of patients aged 80 years or older received chemotherapy in the presence of lymph node disease, significantly less than the 70- to 74-year age group (P = .006). Eleven percent of all patients had positive final surgical margins after resection. Only 54% of the oldest patients received radiation therapy for this, significantly less than patients aged 70 to 74 years (83%) or aged 75 to 79 years (67%) (P<.001). Sixty-eight percent of patients older than 70 years who were ER-positive received hormonal therapy. Unlike the 2 other types of adjuvant therapy, hormonal therapy was more likely to be received by the oldest patients (84%) compared with patients aged 70 to 74 years (55%) or 75 to 79 years (68%) (P<.001). Collectively, these data suggest that hormonal therapy might be substituted for either radiation therapy or chemotherapy in the treatment of certain older patients.
Our data suggest that older breast cancer patients are underdiagnosed, understaged, and undertreated in a community hospital setting. These patients are likely to be diagnosed with breast cancer by presenting with palpable disease. Mammography was used 72% of the time in these patients but largely as an adjunct to the clinical examination. Only 54% of all patients and 38% of patients aged 80 years or older had occult disease detectable only by mammography. Most women aged 70 years and older had stage I or II disease, but complete pathologic staging was only performed in 64%. The oldest group of women was completely staged only 44% of the time.While BCS was performed in 47% patients, adjuvant radiation therapy was used in only 45% of these women. Women aged 80 years and older were least likely to receive any form of therapy, especially chemotherapy (5%), adjuvant to mastectomy or BCS. Moreover, the rate of undertreatment identified in our study is not easily explained by the presence of medical comorbidity that might otherwise preclude the safe use of some treatments. In fact, 74% of all patients (with data) had no significant medical comorbidity (Table 1).
Older women with breast cancer are less likely to have BCS and ALND.
Several studies evaluating regional and/or national cancer databases have shown that the rate of BCS among women older than 65 years is less than 40%.3,6,10,11 The rate of BCS appears to decrease by 11% to 19% with each decade of life.11 By the age of 80 years, only 14% to 25% of women are treated with BCS.3,13,14 In our present study, we noted that 47% of all women aged 70 years or older had BCS and 48% of women aged 80 or older had BCS. Studies have also shown age-related decreases in the use of ALND. In the studies by Bland et al6 and Yancik et al,3 evaluating more than 70 000 women with breast cancer, there was an age-dependent decrease in the use of ALND with up to 60% of women aged 80 years or older having ALND omitted from their care.3,6 Older women were approximately 13 times more likely to have ALND omitted from their care compared with women younger than 65 years.3,6 In our present study, we noted that ALND was omitted from 36% of women older than 70 years and from 56% of women older than 80 years.
Older breast cancer patients are also less likely to have radiation therapy, chemotherapy, or hormonal therapy adjuvant to surgery. In a recent study of more than 16 000 breast cancer patients, multivariate analysis determined that women younger than 70 years of age were twice as likely as older women to receive radiation therapy after BCS.11 In our study, 45% of women older than 70 years received radiation after lumpectomy; the oldest group of women had radiation only 40% of the time after BCS. Yancik et al3 similarly have shown the underuse of radiation therapy after BCS in the treatment of older patients; in this study, up to approximately 11% of older patients received less than minimal therapy based on the National Institutes of Health Consensus Statement for early (stage I or II) breast cancer.3
Chemotherapy and hormonal therapy also have been underused in the treatment of older breast cancer patients. In a study of 407 women with breast cancer aged 80 years or older, chemotherapy was only used in 1.4% of patients following BCS or mastectomy.14 In this present study, we noted that 12% of all women older than 70 years and 5% of women older than 80 years received chemotherapy. Moreover, we noted that only 29% of patients with lymph node metastases were treated with chemotherapy. These rates of chemotherapy use are somewhat lower than the 39% to 53% rates reported by Wanebo et al10 for women older than 65 years with stage I through III breast cancer from the Rhode Island tumor registry. Hormonal therapy, such as tamoxifen, was used in 31% of our patients and 55%, 68%, and 84% of ER-positive women aged 70 to 74 years, 75 to 79, and 80 years or older, respectively. The rate of adjuvant hormonal therapy noted in our study is similar to that described by Gajdos et al15 in their study of 206 women older than 70 years (57%) but less than the 65% to 90% rate determined from the Rhode Island tumor registry for women aged 65 years or older.10 We acknowledge that our data may be limited somewhat by the accuracy of tumor registry rates of ambulatory services such as radiation therapy, chemotherapy, and hormone treatment, which has been questioned recently.16
While the aforementioned studies suggest that older breast cancer patients are more likely to have certain treatments omitted from their care, the appropriate treatment of this group of cancer patients remains difficult to define. Some recent studies clearly have shown that the undertreatment of older breast cancer patients correlates with decreased survival. When ALND with or without radiation or chemotherapy was omitted from the treatment of older women with early stage breast cancer, there was an absolute decrease of 10% to 28% in the 10-year survival rate.6,10 The omission of adjuvant therapy to BCS was associated with an absolute decrease of 27% in the 5-year survival rate of women aged 80 years or older.14 Undertreatment has been correlated with a greater than 3-fold increased risk of death due to breast cancer in older patients.17 Nonetheless, certain other studies have demonstrated that the undertreatment of elderly breast cancer patients does not adversely affect outcome.15 Moreover, certain procedures such as ALND may be associated with increased morbidity and decreased quality of life in older patients.18 Although the complications of ALND may be minimized by using sentinel lymph node biopsy,19 the question has been raised as to whether certain treatments may be omitted safely in selected groups of older patients with breast cancer. For example, because of the low rate of tumor recurrence after BCS in women with stage I (node-negative, tumor size <2 cm), ER-positive breast cancer, tamoxifen without radiation therapy after BCS has been advocated by some for older women.20- 22 On the other hand, the study by Fyles et al23 still demonstrated a benefit of adding radiation to tamoxifen even in this very select group of breast cancer patients. The appropriate treatment of older breast cancer patients requires a thorough analysis of a number of factors and continues to be debated.24
Our study and others suggest that elderly breast cancer patients underuse mammography for screening purposes and are more likely to present with more advanced breast tumors than their younger counterparts.9,10,14 In the study by Grube et al,12 older women were less likely to present with a palpable breast mass; however, this finding seems to be due to the more generous use of mammography in this particular group of older women. In fact, the regular use of mammography is associated with a significant decrease in the tumor size in women aged 69 years and older.9 A 1-mm increase in tumor size at diagnosis has been correlated with a 0.8% decrease in survival from breast cancer.9 Collectively, the results of these studies argue for the regular use of mammography in elderly women. Along these lines, the study by Walter and Covinsky8 showed that there was at least the same expected benefit of screening a healthy 75- to 80-year-old woman for breast cancer as a woman aged 50 years in the lowest 25th percentile of life expectancy. Nonetheless, because of the absence of large studies evaluating cancer screening in older patients, it is still not clear that any age-related disparity in the use of mammography ultimately affects survival.
Several studies, including the present one, have shown that older breast cancer patients as a group are understaged, underdiagnosed, and undertreated compared with their younger counterparts. Although it is not entirely clear that this undertreatment leads to worse outcome, the evidence suggests that this is the case. Moreover, it appears that the undertreatment of elderly breast cancer patients occurs in the university or teaching hospital setting as well as in the community hospital setting.11 In fact, in the study by Morrow et al,11 breast cancer patients were more likely to have radiation therapy omitted after lumpectomy at a teaching hospital than at a community hospital with a comprehensive cancer program. While the management of older breast cancer patients may be complicated by the existence of comorbid conditions, many older patients appear to be denied treatment based solely on their age. Physician attitude toward treating older patients, in fact, actually might play a significant role in perpetuating the undertreatment of the elderly with cancer.25,26 Contrary to many physicians' beliefs, the data suggest that fit older patients derive the same benefits from treatment as do their younger counterparts.24,27,28 The cornerstone of treatment of older breast cancer patients is an adequate geriatric assessment that helps estimate life expectancy and predict tolerance of treatment.28 Treatment strategies then ought to be individualized based on this assessment. Similarly, realistic screening programs for breast cancer should start with this same sort of assessment. Patients considered to have a life expectancy of greater than 5 years would be expected to derive some benefit from regular mammography.1,8 Lastly, perhaps as a result of the inclusion of a greater number of older patients in clinical studies, the medical community might be able to establish better guidelines for the evaluation and care of elderly breast cancer patients in the future.
Correspondence: David A. Litvak, MD, Kaiser Permanente Medical Center, Orange County, Department of Surgery, 411 N Lakeview Ave, Anaheim, CA 92807 (email@example.com).
Accepted for Publication: August 1, 2005.
Author Contributions:Study concept and design: Litvak and Arora. Acquisition of data: Litvak and Arora. Analysis and interpretation of data: Litvak and Arora. Drafting of the manuscript: Litvak and Arora. Critical revision of the manuscript for important intellectual content: Litvak. Statistical analysis: Litvak and Arora. Obtained funding: Arora. Administrative, technical, and material support: Arora. Study supervision: Litvak.