Adjuvant Chemotherapy for Breast Cancer in Older Women

“Age clearly matters. Every physician views a younger patient as potentially different from an older patient. But how different?”

Few cancers are as sensitive to chemotherapy or hormonal therapy as those of the breast. Moreover, breast cancer subtypes have been more extensively investigated than those of other organs. Both of these points are good news for women with breast cancer. A consequence of this is that breast cancer mortality has decreased since 1990 in many countries; in particular, mortality has decreased by more than 25% in the USA. Screening mammography and treatment have contributed to this reduction, although the credit due to each is not precisely known [1].

“The older a person is, the less remaining lifetime there is for any benefits of treatments to accrue and, therefore, any treatment that carries even a modicum of negative aspects is less appealing.”

The quantitative impact of a 25% reduction in breast cancer mortality is much greater than this figure belies. The effect of a therapy or an intervention is typically measured as a hazard ratio. To achieve a 25% reduction in breast cancer mortality requires a hazard ratio of less than 0.50 (i.e., a hazard reduction of >50%). Such a hazard reduction is due to improvements in overall care actually delivered to women with breast cancer. Therefore, progress in breast cancer care has been impressive, at least relative to most other cancers.

For caregivers this good news makes life interesting, but also challenging. How should women with the various subtypes of breast cancer be treated? For example, how should a woman with a node-positive, estrogen receptor-positive, HER2-negative breast cancer be treated? Hormonal therapy is the easy part. What about chemotherapy? Which chemotherapy? Do the answers to these questions depend on other tumor and clinical characteristics? The various genetic profiles (OncotypeDX™, Medical Solutions, Nottingham, UK [2,3] and MammaPrint®, Agendia, CA, USA [4], among others [5]) provide some help. However, these profiles do not address the question of which chemotherapy? In addition, they do not address the question of patient age.

Age clearly matters. Every physician views a younger patient as potentially different from an older patient. But how different? For example, for two cancers that appear the same in every way except for the patients' ages, should treatment be different? According to an Institute of Medicine (IOM) report [101], “The healthcare system today often fails to provide high-quality care to older adults.” The focus of the IOM report is that the medical workforce is not appropriately trained to provide care for older adults. A complication is that which care is the highest quality for older patients may not be known. Such is the case in breast cancer. This situation has led some researchers to call for treatment guidelines for older women with breast cancer [6], but there is no general agreement as to what those guidelines should be. This editorial gives my perspective regarding the evidence.

Any chemotherapy must be judged on the basis of its benefits and its harms. The latter are usually better understood than the former. Available evidence suggests that chemotherapy reduces the hazards of relapse and death to the same extent for older as for younger patients [7]. Toxicity is also similar in the two age groups, except that the rate of grade 4 hematological toxicity is somewhat greater in older patients [8].

“Any chemotherapy must be judged on the basis of its benefits and its harms. The latter are usually better understood than the former.”

How old is old? The older a person is, the less remaining lifetime there is for any benefits of treatments to accrue and, therefore, any treatment that carries even a modicum of negative aspects is less appealing. Every treatment has diminishing returns with increasing age. The age at which the harms outweigh the benefits depends on the patient. There is an age at which a cancer should not be treated (and consequently an age at which screening for cancer is more harmful than beneficial), but this age will depend on the patient – on her health status and on her own weighing of benefits and harms.

“…considering the dearth of randomized evidence for older women, I favor less use of chemotherapy than is typical among oncologists.”

For the purposes of this editorial, ‘older’ is considered to mean an age of at least 65 years. These women are under-represented in clinical trials [9] even though they represent approximately half of all women with breast cancer in the USA. Such under-representation leads to calls for replicating positive trials in older populations. This will not happen and should not happen. There are some differences in breast cancer by age, but after appropriately accounting for the heterogeneous nature of breast cancers, what is learned in younger populations also applies to older populations, with some caveats.

Breast cancer in older women is well understood to be less aggressive than in younger women. This makes physiological sense. Breast cancers that become evident at a young age, for example 35 years of age, are themselves young – the cancer cannot be older than the patient. Many of these cancers are likely to be 5–15 years of age. Approximately 27 doublings in size are required for a tumor to be detectable. Therefore, the doubling times in these ‘young cancers’ might be 2–6 months. While many breast cancers that become evident at the age of 65 years may also be 5–15 years old, some are older, perhaps older than 30 years. Any tumor that has existed in a woman's body for such a long time is by definition slowly growing, with a doubling time of 1 year or more. Such a tumor may not be lethal at all, and it may not be growing fast enough to kill a 65-year-old woman even if left untreated.

When a woman has breast cancer, we know more than her age. We know the stage of the disease, at least approximately, and we may know the tumor's histology, grade, hormone-receptor status, HER-2/neu status, Ki67 expression and and other factors. These characteristics provide some information regarding the lethality of the tumor, but considering the current state of knowledge, the information is far from perfect – unless the tumor has metastasized. The question is how to treat a tumor that has particular characteristics in a woman at a particular age. A thorough investigation of this question would be a huge undertaking and is beyond the scope of this editorial. At best, I will be able to address the tip of the iceberg and relate it to patient age.

An essential consideration for understanding the benefits of chemotherapy is the role of screening. Screening mammography is effective in finding many cancers early. This leads to a ‘stage shift’; screen-detected tumors tend to be found at an earlier stage of development. Some tumors detected by screening would never be found otherwise, giving rise to overdiagnosis [102]. The rate of overdiagnosis is greater in older women, partly because their remaining lifetime is less.

Not surprisingly, cancers detected by mammographic screening have better prognoses. (This improved prognosis is frequently misunderstood to be a benefit of screening [10].) What may be surprising is that even for the same stage of disease and the same characteristics of the cancer, screen-detected tumors have markedly better prognoses than those detected symptomatically [11–13]. This is true whether the latter are interval cancers (detected between scheduled mammograms) or cancers found in women who have never been screened or who have not been screened in many years.

An explanation for this observation is ‘length bias’ [14] (screening preferentially finds slowly growing tumors), which is present at every age, but just as with overdiagnosis it is more pronounced in older women.

“The rate of overdiagnosis is greater in older women, partly because their remaining lifetime is less.”

Consider the categories of breast cancer in older women. One of the few randomized clinical trials in older women studied radiation therapy in women aged 70 years and older who had small, node-negative, estrogen receptor-positive tumors treated with tamoxifen [15]. This Cancer and Leukemia Group B (CALGB) trial demonstrated that radiation therapy reduced local or regional recurrence in these women by a modest 5% at 5 years. However, radiation therapy had no effect on distant recurrence or overall survival. (This supports the conclusion of the Early Breast Cancer Trialists' Cooperative Group that radiation therapy in older women has little benefit, and any benefit is outweighed by the harms [16].) For our purposes, the important conclusion of this trial is that only approximately 5% of deaths were caused by breast cancer. Any benefit of chemotherapy would be restricted to this tiny proportion of patients. Furthermore, for reasons discussed later in this editorial, it is unlikely that chemotherapy would benefit even these few patients.

“Screening in women over the age of 65 years may lead to overdiagnosis and unnecessary, and possibly harmful, treatment.”

Another CALGB trial considered adjuvant chemotherapy in breast cancer patients aged 65 years and older, irrespective of hormone-receptor status [17]. Randomization was to single-agent capecitabine versus standard polychemotherapy, either cyclophosphamide-methotrexate-fluorouracil or doxorubicin-cyclophosphamide. Standard therapy had substantially better relapse-free and overall survival compared with capecitabine. Moreover, this benefit was restricted to patients with hormone receptor-negative tumors. This conclusion is supported by other studies of elderly patients [18,19], and it is supported by studies in patients of any age demonstrating that the benefits of modern improvements in chemotherapy are largely carried by hormone receptor-negative tumors [20], especially by those that are also HER-2/neu positive [21–24].

Recommendations of oncologists regarding chemotherapy in elderly patients vary greatly. Some eschew chemotherapy to older women while, to others, this attitude is untenable [6,25]. Based on the empirical evidence, including that which has been extrapolated from younger patients, considering the dearth of randomized evidence for older women, I favor less use of chemotherapy than is typical among oncologists. My recommendations are shown in Figure 1.

Figure 1.

Recommendations for ‘chemotherapy’ or ‘no chemotherapy’ for primary breast cancer in patients aged 65 years and older.

As indicated in Figure 1, many factors not considered explicitly in the diagram affect decisions regarding chemotherapy. This is the case when the recommendation is ‘possibly chemotherapy’, but also when it is ‘chemotherapy’ or ‘no chemotherapy’. One factor is patient age within the older population. An age of 80 years is in one category while the age of 79 years is in another. Obviously, these two ages are not very different. On the other hand, the ages of 65 and 79 years are quite different even though they are in the same category. Another factor is the patient's comorbidities. Some 80-year-old women have life expectancies greater than some 65-year-old women. Individual patient attitudes toward chemotherapy and its toxicities vary considerably. Patients should be fully informed about them and about the (usually great) uncertainty regarding therapeutic benefit. A very important factor is the method of detection. Since tumors detected by mammographic screening have relatively good prognosis, chemotherapy stands to be less useful for patients with these tumors.

Not indicated in Figure 1 is the type of chemotherapy. This is also subject to much controversy, and not just for older patients [103].

“Despite enormous progress in breast cancer research, as yet we cannot discern lethal from nonlethal tumors. This is an important goal.”

Any recommendation regarding chemotherapy is temporary, including that in Figure 1. Breast cancer research is active and dynamic. Molecular markers and combinations of molecular markers are continually being discovered that may predict prognosis and also the benefits of therapy. In the not too distant future we may be able to tell which tumors are lethal and which are not. In addition, alternative treatment options are rapidly becoming available for women with breast cancer, both young and old. Figure 1 is tailored to the information available at present, tomorrow is another day.

Conclusion

The various subtypes of breast cancer have been investigated more extensively than for other cancers. Tumor sensitivity to chemotherapy and hormone therapy depends on subtype. Many clinical trials have been undertaken to determine appropriate screening and treatment protocols for perimenopausal and younger women. Few studies have evaluated breast cancer in women aged 65 years or older. Despite enormous progress in breast cancer research, as yet we cannot discern lethal from nonlethal tumors. This is an important goal, especially when determining treatment for older women who may not benefit from intervention for a slow-growing breast tumor. Screening in women over the age of 65 years may lead to overdiagnosis and unnecessary, and possibly harmful, treatment. No consensus exists on treatment guidelines for older women with breast cancer. I favor less use of chemotherapy in this population.