Management of operable breast cancer in older women

Abstract

Older women are not being given the opportunity to benefit from the improvements in both local and systemic treatment for breast cancer. Mammographic screening call/recall system ceases at age 72, making access more difficult. Knowledge about breast cancer in those aged >75 is significantly reduced in terms of understanding symptoms and personal risk but studies have shown that intervention can improve this, at least in the short term. Although older women are more likely to have estrogen receptor positive tumours, nevertheless, more than one-third of women aged over 70 have grade III, aggressive breast cancers. Whenever possible, older women should be offered breast conserving therapy rather than mastectomy since this not only improves their quality of life but also reduces risk of subsequent mental health problems. Endocrine treatment alone should not be used other than in patients with severe co-morbidity and a life-expectancy of less than a year. As adjuvant treatment in those with estrogen receptor positive cancers, the choice between tamoxifen and an aromatase inhibitor will depend upon co-morbidity, side-effects and patient choice.

Introduction

Although 50% of the cases of breast cancer diagnosed in the UK are in women aged >65, unfortunately these comprise 60% of the deaths from the disease. In Europe there is a widening gap between survival after diagnosis of breast cancer in middle aged (55–69 years) and older women (70–84 years).¹ From a pool of 51 European populations diagnosed with all cancers between 1988 and 1999, five-year relative survival was estimated. There was a significant survival improvement over this period for all cancers combined and for every cancer site, but at a slower rate in the elderly. The gap between younger and older patients widened over the 11 years, particularly for breast and gynaecological cancers. After allowing for co-morbidity there is still relative under-treatment in the elderly together with obstacles preventing older women undergoing mammographic screening with the call/recall system ceasing at age 72.

Screening

While there is debate about the costbenefit of mammographic screening in the middle-aged population there appears to be indifference among some policy-makers concerning benefits in older women. Recently, van Schoor et al.² reported mortality statistics for 55,529 women screened in Nijmegen between 1975 and 2008. The odds ratio (OR) for mortality in screened versus unscreened cases was 0.72 for those screened between 1975 and 1991. The 95% confidence interval was 0.47–1.09, so that this did not achieve statistical significance. In contrast, among those screened between 1992 and 2008 the OR was 0.35 with a tight confidence interval of 0.19–0.64. This potential almost two-third reduction in breast cancer mortality is being withheld from many older women.

It will be argued by some that such benefits may not apply to older women. Galit et al.³ carried out a systematic review of all studies dealing with screening mammography in women >74 years. The studies were reviewed according to their outcomes and study design, focusing on breast cancer mortality and stage of breast cancer at diagnosis. In the three studies examining breast cancer screening and mortality, the risk of disease-specific mortality in the 75–84 years age-group was approximately doubled among unscreened women. There were four studies showing that older screened women had significantly smaller tumours of lower stage at diagnosis.

Vacek et al.⁴ examined a cohort of 19,779 Vermont women aged 70 and older and applied four different risk models to determine whether a high-risk group could be identified. After seven years follow-up, 821 developed breast cancer. The risk models were Gail, Tice modification, Barlow and Vermont. The c-statistics for associations between published individual risk factors and breast cancer risk were 0.54, 0.54, 0.55 and 0.55 (respectively indicating their lack of utility in assigning risk). The authors concluded that age-related attenuation of the effects of some risk factors makes the prediction of breast cancer in older women particularly difficult.

Breast awareness

This expression is often used, somewhat tritely, without questioning what it means, whether it can save lives and if it can be taught? In part, the concept concerns knowledge of symptoms that may presage malignancy and a reasonable estimation of personal risk. To assess the state of knowledge in the general population, Grunfeld et al.⁵ randomly selected 996 women from across the UK and asked them about their breast cancer risk, what they regarded as likely breast cancer symptoms, and their views on treatment and outcome. Given options for lifetime risk, 31% picked 1/1000, 35% chose 1/100 and only 23% selected the then correct option of 1/10. When asked about their risk in relation to the general population, 35% of the participants aged >65, 35% underestimated their risk of breast cancer believing themselves to be too old to develop the disease. Knowledge of the spectrum of symptoms of breast cancer was significantly reduced in those aged >75.

The same research group subsequently examined knowledge of symptoms and ability to detect a lump in 712 women aged 67–73.⁶ At this time, 50% of responders believed their lifetime cancer risk to be one in 100, and three quarters were unaware that age was the major risk factor for breast cancer. Approximately one-third were unconfident about their ability to detect a breast lump and one-fifth never bothered to check themselves.

To determine whether brief interventions could promote early presentation in breast cancer, Linsell et al.⁷ carried out a three-way randomised trial. Participants were 867 women aged 67–70 years, attending for their final routine appointment on the UK Breast Screening Programme. They received a scripted 10-min interaction with a radiographer plus a booklet, a booklet alone or usual care. At one month, the intervention increased the proportion who were breast cancer aware compared with usual care (33% versus 4%). There was also a smaller benefit for the booklet alone (13% versus 4%). At one year, the effects of both the interaction plus booklet, and the booklet alone were largely sustained, although the interaction plus booklet was much more effective. This is encouraging evidence that breast cancer awareness may be learnt in older women.

Linsell et al.⁸ went on to devise a Breast Cancer Awareness Model (BCAM) in order to quantify awareness and also then validated the measure. The psychometric properties of the BCAM were measured in 1035 women attending the NHS Breast Screening Programme. Acceptability was assessed with a feedback questionnaire conducted on 292 women aged 67–73. The sensitivity after an intervention was assessed together with test–retest reliability in 167 women aged 43–82. Construct validity was examined using the ‘known-groups’ method. Over 90% of women found the readability of the BCAM was acceptable. The BCAM was sensitive to change with an increase in the proportion of women getting the full score for breast cancer awareness one month after an intervention promoting breast cancer awareness. Test–retest reliability of the BCAM was moderate to good for the majority of items. Cancer experts had higher levels of awareness than non-medical academics (50% versus 6%, P = 0.001), suggesting that there was good construct validity, but indicating a knowledge gap among non-medical academics.

Breast cancer biology and age

Daidone et al.⁹ reported the hormone receptor status in 14,007 women with operable breast cancers treated at National Cancer Institute Milan, between 1974 and 2001, and analysed the rate of positivity in relation to age as shown in Figure 1. This clearly indicates an increase in estrogen receptor positivity with age but, even in the oldest age group >84 years, almost 20% were estrogen receptor negative (ER −ve) and over one-third were progesterone receptor negative (PR−ve). On this basis age alone cannot be used as an indication for endocrine therapy which should be determined by receptor assay of the primary tumour.

Figure 1.

Estrogen and progesterone receptor status of breast cancers in relation to patient age (Daidone 2003)

Fisher et al.¹⁰ examined the histology of 1869 breast cancer cases treated at Guy's Hospital between 1983 and 1992, and subdivided this into four age groups: ≤39years (148), 40–49 (355), 50–69 (984) and ≥70 (382). It is of note that 21% of the cases were aged ≥70. Figure 2 shows that in the youngest age group, 65% had grade III cancers compared with 37% in those aged ≥70. Hence although there is a reduction in poorly differentiated cancers, with age there still remains a significant proportion with high-risk cancers.

Figure 2.

Histology of breast cancer in relation to age (Fisher 97)

Impact of mastectomy in older women

There is a common misconception that older women are not concerned about their body image so that mastectomy rather than breast conserving surgery is an acceptable option. This is not borne out by the available evidence. Figueiredo et al.¹¹ carried out a telephone study on a cohort of 563 women aged ≥67 years, diagnosed with operable breast cancer, 3, 12 and 24 months after surgery. All had tumours suitable for breast conservation. Body image was measured with the Cancer Rehabilitation Evaluation System-Short Form, and mental health assessed using a Medical Outcomes Study subscale. For 31% of those surveyed, body image was a major factor in treatment decisions and this is similar to that reported in younger women.¹² Those treated by breast conservation had significantly better body image two years after treatment. The poorest body image occurred in those preferring breast conservation but being subjected to mastectomy. Of great interest, body image predicted two-year mental health. It is therefore most important that preferences about appearance are sought from women since shared decision-making may improve mental health in older patients with breast cancer.

Avoidance of mastectomy

There are three reported prospective randomised trials that compared tamoxifen alone with surgery for women aged >70 with operable breast cancer and two that involved randomised patients using tamoxifen alone or tamoxifen plus surgery. The relapse rates are shown in Figure 3.^13–17 The consistent finding was that in a group of older women with both ER+ve and ER−ve cancers, there was an increased risk of both progression and recurrence of the disease in those given tamoxifen alone. In the Cancer Research Campaign trial the increase in relapse rate was also associated with a significant increase in breast cancer mortality.¹⁵ As a result the authors advised that the use of tamoxifen alone should be avoided unless the patient had such severe co-morbidity that their life-expectancy was <1 year.

Figure 3.

Relapse rates in randomized trials of tamoxifen versus surgery: St GH, St George's Hospital; Notts, Nottingham City Hospital, CRC, Cancer Research Campaign, GRETA, Group for Research on Endocrine Therapy in the Elderly, EORTC, European Organization for Research & Treatment of Cancer

In European Organization for Research & Treatment of Cancer (EORTC) trial 10850, patients aged >70 with operable breast cancer were treated by either wide excision and tamoxifen or modified radical mastectomy, as opposed to tamoxifen alone versus mastectomy in EORTC 10851.¹⁸ No selection was made for ER+ve cases. There were significantly more loco-regional relapses in the Wide Local Excision+T group but more distant metastases in the modified radical mastectomy (MRM) group, but with a similar overall survival in both groups. No trial has compared wide excision and tamoxifen with mastectomy plus tamoxifen. In terms of the safety of tamoxifen in older women, there were 30 deaths from cardiovascular disease in the wide excision plus tamoxifen group compared with 31 in the mastectomy group.

Within this study, quality of life (QOL) was assessed with a 36 item QOL questionnaire completed 2–12 months after randomization in 136 patients of whom 65 were treated by mastectomy and 71 by wide excision and tamoxifen.¹⁹ No significant difference in the duration of survival between the two treatment arms was observed with patients included in the QOL substudy. There was no difference between mastectomy and wide excision cases in terms of fatigue, emotional functioning, fear of recurrence, social support, physical functioning and leisure time activities, but conservatively treated patients reported fewer arm problems and better body image.

The outcome in 71 patients aged ≥70 years, randomised to wide excision and tamoxifen at Guy's Hospital has been recently updated.²⁰ These cases had been randomised within EORTC 10850 and a pilot trial at Guy's. After 15 years median follow-up, 29 patients (41%) had a loco-regional relapse, of whom five had synchronous metastatic disease. Most tumours recurred in the index quadrant, 16(67%). Salvage by mastectomy was feasible in 21/24 patients (88%). Three factors significantly predicted loco-regional recurrence: lympho-vascular invasion (HR [95% CI]: 11.88 [4.52, 31.22], P < 0.001), estrogen receptor, ER−ve status (HR [95% CI]: 3.61 [1.33, 9.79], P = 0.012), and tumour necrosis (HR [95% CI]: 2.95 [1.13, 7.70], P = 0.027). Age at diagnosis, clinical nodal status, progesterone receptor, PR status, site in breast and final margin status were not associated with loco-regional relapse.

Based on these results, older patients who are suitable for breast conserving surgery followed by endocrine therapy alone should have ER+ve tumours, without lymphovascular invasion or sentinel node involvement. Tumour necrosis would also appear to be a contraindication to this approach. The value of additional radiotherapy in terms of local control and overall survival in older patients will need assessment in a large multicentre trial.

Aromatase inhibitors

The first randomized trial to indicate the superiority of an aromatase inhibitor anastrozole compared with tamoxifen was ATAC (Arimidex, Tamoxifen Alone or in Combination). Recently, a subset banalysis has been conducted on 1662 (27%) patients who were aged ≥70 years at entry to the study.²¹ Older women were less likely to have breast conserving surgery, receive adjuvant radiotherapy or chemotherapy. They were at increased risk of recurrence and had a four-fold increased risk of death without recurrence.

In BIG 1-98 trial, 4922 patients were randomized to five years of adjuvant tamoxifen or letrozole and of these, 3127 were aged <65 years, 1500 were 65–74 years and 295 aged ≥75 years at entry. After a median follow-up of 40.4 months the disease-free survival was significantly better in those randomized to letrozole and this was seen in all three age groups, Those aged >75 were less likely to complete adjuvant therapy irrespective of whether this was tamoxifen or letrozole and although fractures were more frequent in the letrozole arm, rates did not differ with age.

In postmenopausal women with ER+ve breast cancer who have taken 2–3 years of adjuvant tamoxifen, switching to the steroidal aromatase inhibitor exemestane improves relapse-free survival compared with continuing tamoxifen.²² In the Tamoxifen Exemestane Adjuvant Multinational trial exemestane alone was compared with tamoxifen followed by exemestane.²³ A total of 4904 patients were assigned to exemestane and 4875 to sequential. At five years, relapse free survival was similar in both groups 85% sequential and 86% exemestane alone. Sequential treatment caused more gynaecological symptoms (20% versus 11%) and endometrial change (4% versus<1%). With exemestane alone, musculoskeletal adverse events were more frequent (50% versus 44%). In the absence of a clear survival benefit it is likely that the choice of tamoxifen or an aromatase inhibitor will depend upon the patient's co-morbidity and profile of side-effects.

Conclusions

To achieve a reduction in mortality from breast cancer in older women a multifaceted approach is necessary with campaigns to improve knowledge of the disease and foster breast awareness together with easier access to mammographic screening. Furthermore, once diagnosed with breast cancer older patients should be offered optimal local therapy and when appropriate, adjuvant therapy that will minimize side-effects and maximise their likelihood of remaining disease free.

DECLARATIONS

Competing interests

None declared

Funding

None declared

Ethical approval

Not applicable

Guarantor

ISF

Contributorship

ISF is the sole contributor

Acknowledgements

None declared