Undertreated Breast Cancer in the Elderly

The effect of undertreatment with adjuvant hormonal therapy, chemotherapy, or radiation was studied in elderly women with breast cancer. A prospectively maintained database was used to identify women undergoing potentially curative surgery between 1978 and 2012. The presentation, pathologic findings, treatment, and outcomes of 382 women over 70 were compared to the findings in 2065 younger patients. Subsequently, conventionally treated and undertreated elderly patients were identified and their characteristics and outcomes were compared. Both young and old patients presented most frequently with mammographic findings, but older patients presented more frequently with mammographic masses while younger patients presented more frequently with mammographic calcifications. Cancers of older patients were significantly more favorable than cancers in younger patients: smaller, with more infiltrating lobular, fewer ductal carcinoma in situ, and more frequently estrogen receptor positive and fewer were poorly differentiated. Elderly patients had less axillary sampling, fewer mastectomies, less adjuvant radiation therapy, and more hormonal therapy. Fifty-one percent of the 382 elderly patients were undertreated by conventional criteria. Undertreated patients were more frequently in situ, better differentiated, smaller, and more often estrogen receptor positive. Forty-four percent of the undertreated patients died during followup without disease recurrence. Despite undertreatment, local and distant disease-free survival was comparable to patients who were not undertreated.


Introduction
e population of elderly individuals in the United States is increasing. Between 2000 and 2010 the population of women aged 65 and over increased by 11.3% with those 70 and over increasing by 6.4% [1]. According to the Surveillance Epidemiology and End Results (SEERs) database, from 2000 to 2009 the median age for breast cancer diagnoses in the USA was 61 years of age. Approximately 41% were diagnosed above the age of 65, of which 21% were above the age of 75 [2]. As the USA population of women over 65 increases, breast cancer in older individuals has and will continue to become more prevalent.
e management of breast cancer in the elderly has been a topic of debate. ere is a lack of evidence on the optimal management of this group of patients secondary to low enrollment in randomized clinical trials [3,4]. As a result, treatment decisions have been largely based on studies in younger patients which may not be applicable to elderly patients with breast cancer. Breast cancers in elderly women compared to younger women are histologically less aggressive and have a good response to hormonal therapy. is favorable biologic pro�le impacts the decision as to whether an elderly patient should be subjected to adjuvant therapy. e consequences of these considerations are that elderly patients are oen undertreated when compared to younger patients [5][6][7], but the question that needs to be answered is are there any clinical rami�cations to the undertreatment of breast cancer in elderly women [6,8,9] ? Diab and colleagues demonstrated that the impact of breast cancer on the expected survival of these elderly patients decreases with age [9] and the risk of dying from comorbid conditions oen exceeds the risk of cancer recurrence and breast cancer mortality [10]. Although recommendations based on expert opinion are emerging, there is a paucity of level 1 evidence [11]. Determining the optimal treatment for an elderly patient depends largely on clinical judgement, weighing the patients' comorbid conditions with the biology of the tumor. Data was collected on age, clinical presentation, mammographic �ndings, diagnostic method, histopathologic �ndings, tumor differentiation, tumor size, estrogen receptor status, axillary node status, resection margins, number of pathologically examined nodes, surgical treatment, reexcision, adjuvant hormone treatment, and chemotherapy and radiation therapy. Followup information was acquired from hospital and office records, patients, and their families. e last date of followup and the date of local or distant recurrence were recorded. e local and distant disease free survival rates were then calculated from the date of de�nitive surgery. For estimates of local and distant disease recurrence rates, patients in whom a recurrence did not develop were censored at the last followup or death, whichever occurred �rst.
Patients over 71 years of age who were undertreated by conventional criteria were compared to their appropriately treated counterparts. Our criteria for undertreatment included (1) omission of axillary sampling in patients with invasive tumors; (2) lack of postoperative radiation therapy in patients treated with breast conserving surgery; (3) lack of hormonal treatment in estrogen receptor positive patients with invasive cancers; (4) lack of chemotherapy in nodepositive patients; (5) lack of chemotherapy in estrogen receptor negative patients with tumors larger than 2 cm. e data was analyzed using SPSS soware (SPSS Inc., Chicago, IL, USA) run on a Dell personal computer. e patients were divided into two groups by age (including the age of 71 and over or younger than age 71) and compared. e signi�cance of differences in categorica variables was evaluated using chi-square test, and the signi�cance of differences in continuous variables was evaluated using Student'stest. Cumulative 5-year local and distant disease free survival rates were calculated using Kaplan-Meier method [12]. Cox's proportional hazards regression model was used to evaluate the relative prognostic signi�cance of variables for both local and distant disease free survival [13].

Results
e 2,447 patients ranged in age from 22 to 96 years and 382 (16%) were of age 71 and above, considered elderly (Table 1). e 2,065 younger patients ranged in age from 22 to 70, with a median age of 53 and the patients over 70 years ranged in age from 71 to 96 years with a median age of 76. Most patients presented with a palpable mass (37%). Patients younger and older than 71 years were equally likely to have mammographic �ndings. Older patients presented more frequently with mammographic masses while younger patients presented more frequently with mammographic calci�cations. �oth the elderly and the younger patients were most commonly diagnosed by excisional biopsy followed by core needle biopsy and �ne-needle aspiration. �umerous signi�cant differences were observed between the elderly and younger patients in terms of their pathology (Table 2). Older patients had signi�cantly more in�ltrating lobular cancers and fewer cases of ductal carcinoma in situ than younger patients and signi�cantly fewer poorly differentiated cancers. e mean tumor size was signi�cantly smaller in the elderly but the T stage distribution among the elderly and younger patients was comparable. Estrogen receptor positivity was more frequent among the elderly.
Axillary node sampling, sentinel node excision, or axillary dissection was more frequent in younger patients with removal of more lymph nodes with the proportionately more nodes involved with tumor. In addition to less aggressive treatment of the axilla, elderly patients also received less aggressive surgical treatment of the breast: only 14% received mastectomies compared to 25% of younger patients. Adjuvant therapy with both radiation and chemotherapy was signi�cantly less frequent in the elderly while the use of Tamoxifen or an Aromatase inhibitor was more frequent. 81% of the 1,529 young patients treated with breast conservation received radiation therapy compared to 53% of the 327 elderly patients treated with breast conservation ( < ). Among patients with invasive cancers, 18% of the young patients received neoadjuvant chemotherapy and 33% adjuvant chemotherapy compared to 6% and 7% of the comparable elderly patients ( < ). e main form of systematic therapy for the elderly patients was hormonal: either Tamoxifen or Aromatase inhibitor. 67% of elderly patients were treated with hormonal therapy compared with 56% of younger patients ( < ). Despite these differences, the elderly and younger patients had similar 5-year local and distant recurrence-free survival (Table 3).
Undertreated elderly patients were identi�ed as described in Section 2. Undertreatment consisted of omission of radiation therapy in 154 of the 317 patients treated with breast conservation, omission of axillary node sampling in 61 of the 310 elderly patients with invasive cancers, omission of chemotherapy in 10 of 63 elderly patients with involved nodes, and omission of hormonal therapy in 59 of 321 elderly patients with estrogen receptor positive cancers. By these criteria many patients were undertreated with more than one modality. As a consequence, 190 (51%) of the elderly patients were undertreated with at least one modality. Undertreated elderly patients were signi�cantly older than their appropriately treated counterparts (77 versus 75, < 0.001). e cancers of the undertreated elderly were more frequently in situ, better differentiated, smaller, and more oen estrogen receptor positive (Table 4). Re�ecting the criteria used to identify undertreated patients, one-third did not receive axillary sampling for invasive cancers, twothirds did not receive radiation, almost half did not receive hormonal therapy, and a few received chemotherapy. Despite these differences in treatment, elderly undertreated patients generally fared as well as the appropriately treated elderly (Table 5). Equal numbers of patients in both groups developed local recurrences resulting in �ve-year cumulative local disease-free rates of 93% for the appropriately treated and 91% for the undertreated. 9% of the 167 appropriately treated elderly patients with invasive cancers developed distant disease compared to 4% of undertreated patients causing the cumulative �ve-year distant disease free rate to be 89% in appropriately treated patients compared to 93% in the undertreated one. It is important to note that 44% of the 190 undertreated elderly died without disease recurrence compared to 29% of the appropriately treated patients ( < 0.001). A Cox regression model was used to evaluate potential prognostic factors such as tumor pathology, differentiation, size, number of involved nodes, estrogen receptor status, and treatment with chemotherapy, hormonal therapy, and radiation, among the elderly patients (Table 6). Local disease-free survival was signi�cantly related to estrogen receptor status ( < 0.001) and pathology ( 0.0 ). Twenty-four percent of the 46 patients with estrogen receptor negative cancers developed local recurrence within �ve years compared to 3% of the 275 patients with estrogen receptor positive tumors. e cumulative �ve-year risk of local recurrence in patients with ductal carcinoma in situ was 4% (2/66) compared to 10% (3/45) in patients with invasive lobular cancers and 9% (18/266) in patients with invasive ductal cancers. Among patients with invasive cancers, tumor size ( 0.00 ), number of involved nodes ( < 0.001), and estrogen receptor status ( 0.00 ) were signi�cantly related to distant recurrence. Undertreatment was not signi�cantly related to local or distant recurrence in univariate or multivariate analysis.
Undertreatment with radiation in elderly patients that underwent breast conservation was associated with increased risk of local recurrence. Five-year local disease-free survival of the unirradiated patients was 90% compared to 96% for the irradiated patients ( 0. 0). e cumulative �ve-year distant disease free survival of patients receiving chemotherapy was 73% compared to 93% for patients not receiving chemotherapy ( 0.00 ). is difference is attributable to the larger, more poorly differentiated cancers with more positive nodes among patients receiving chemotherapy. Omission of hormonal therapy in estrogen receptor positive patients resulted in a lower distant disease free survival: 91% of estrogen receptor patients treated without hormonal therapy were without distant metastases at �ve years compared to 94% of patient using hormonal therapy.

Discussion
is study found that elderly patients with breast cancer present with palpable masses and mammographic �ndings similar to younger patients, although mammographic masses were more frequent in the elderly and mammographic calci�cations were more frequent among the young patients. Cancers of the elderly tended to be less oen in situ than in younger patients but invasive cancers were generally smaller, better differentiated, more frequently estrogen receptor positive, and with less nodal involvement. Older patients were treated less aggressively than younger patients. ey received fewer mastectomies, less radiation aer breast conservation, and very seldom did they receive chemotherapy even for node-positive cases. Elderly patients received hormonal therapy as frequently as younger patients. Despite oen being undertreated, elderly patients experienced outcomes comparable to younger patients presumably because their cancers were smaller, better differentiated, and with fewer involved nodes.
More than one-half of our elderly patients were also undertreated according to current breast cancer treatment guidelines: omission of axillary sampling in patients with Previous studies of elderly patients with breast cancer have not universally observed that cancers in the elderly are biologically more favorable and less advanced than those seen in younger patients. is is in part due to differences in the populations studied. Generally when one compares the cancers of patients over 70 to patients between 50 and 70, differences are not striking [14,15]. However, if one includes all patients younger than 70, the more favorable biology becomes more apparent [16]. In addition, many studies included elderly patients who were not treated with surgery for a variety of reasons including comorbidity, advanced disease, and patient refusal [17][18][19][20]. All of the patients in the current study were potentially curable at presentation; all had surgery, and no stage IV patients are included. A universal �nding in all the studies is the increasing frequency of estrogen receptor positivity with increasing age. is usually results in the increased use of hormonal therapies in the elderly.
Undertreatment of the elderly is also a universal �nding. In fact several authors have found that undertreatment, that is, lack of adherence to guidelines, is frequent at all ages [14]. e controversy that exists is whether undertreatment of patients, particularly the elderly, results in adverse outcomes. ere is no question that radiation therapy reduces local recurrence rates aer breast conservation for invasive and in situ disease regardless of the patient's age. However, a reduction of 3% in local recurrence does not signi�cantly bene�t an 80-year-old woman with a life expectancy of ten years who has only a 50% chance of experiencing the bene�t of radiation therapy [21]. Another consideration is that patients who are not irradiated and develop local recurrences may be candidates for relumpectomy with or without radiation therapy, whereas patients who develop local recurrences aer treatment with radiation should undergo mastectomy.
Previous studies noted that elderly patients with invasive cancers experience higher mortality when axillary dissection is omitted [22]. Among these studies, a few measured breast cancer speci�c survival. It is likely that patients not undergoing axillary dissection have higher comorbidities causing the higher mortality, not that the omission of axillary surgery caused the higher mortality. e recently completed trial randomizing patients with involved sentinel nodes to completion axillary dissection versus no additional surgery showed no bene�t for completion axillary dissection [23].
Finally, with respect to chemotherapy, a few elderly patients are willing to participate in randomized trials with chemotherapy arms and a few are willing to accept chemotherapy even with relatively advanced disease [3,4,24,25]. Only 36 of our elderly patients were estrogen receptor negative and 13 of these had nodal involvement. All received chemotherapy and an additional 11 patients with node negative estrogen receptor negative larger cancers received chemotherapy. Because of the small numbers of patients and the association of chemotherapy with advanced estrogen receptor negative disease, patients receiving chemotherapy fared worse than patients not receiving chemotherapy.
is study has several limitations. It is a retrospective single surgeon database review and thus carries the inherent limitations of an observational study. is includes a potential physician bias and bias as a result of confounding by indication. It must be mentioned, however, that in today's world of cancer treatment, care is individualized and the patient ultimately determines what treatment she is to receive. A larger multicenter, prospective randomized trial of adherence to guidelines for the treatment of breast cancer in elderly patients would be needed to overcome these biases. is trial, however, is unlikely to occur and probably does not need to. Breast cancer in elderly patients has a favorable biological pro�le and therefore treatment does not need to fall under the con�nes of traditional guidelines. Moreover, coupled with comorbid conditions that are frequently encountered as people age, optimal treatment should be determined largely by clinical judgement on a case by case basis. It is known that elderly patients are undertreated but this study did not �nd that the omission of conventional surgery or adjuvant therapies adversely affected outcome among patients over 71 years of age. ��n��c� �f �n�eres�s e authors do not have any con�ict of interests to disclose.