A comparison between young and old patients with triple-negative breast cancer: biology, survival and metastatic patterns

Much research has focused on triple-negative breast cancer (TNBC) in young breast cancer patients. TNBC constitutes a larger proportion among younger than older breast cancer patients [1, 2]. With an aging population and increasing breast cancer incidence with age, a significant number of TNBC cases arise in elderly. Older patients with breast cancer are underrepresented in randomized clinical trials [3]. Few studies have compared the prognosis of younger and older women with TNBC [4].

Breast cancer (BC) is the most common cancer in women worldwide with more than 2 million new cases and 627 000 deaths in 2018 [5]. TNBC accounts for about 15% [6] of breast cancer cases but is the cause of a disproportional number of breast cancer deaths. TNBC lacks the expression of the estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) [7]. TNBC tends to behave more aggressively and remains a poorly defined and heterogeneous subset of breast cancer with an adverse prognosis [7]. The risk of distant recurrence typically peaks between 1 and 3 years after diagnosis and declines rapidly thereafter [8]. Unlike ER-positive and HER2-positive breast cancer, there are no approved targeted adjuvant treatments available and chemotherapy remains the sole medical intervention with a proven effect on long-term survival [6]. There is a larger benefit to adjuvant/neoadjuvant chemotherapy in ER-negative compared with ER-positive breast cancer [9]. Up to 20% of TNBC cases harbor a BRCA mutation, particularly in BRCA1 and the risk of having a BRCA mutation is higher among young TNBC patients [10].

Is TNBC the same disease in older as in younger patients? We conducted this study to determine the biology, recurrence rate, metastatic patterns and survival from diagnosis of primary breast cancer, as well as from diagnose of distant metastasis in a population-based Swedish cohort of women with primary TNBC, with focus on the comparison between younger and older patients.

Previous data have shown improved survival in women with BC over time but this is not true to the same extent in patients > 75 years old [14]. The major findings of this investigation were that recurrence-free, distant disease-free and breast cancer-specific survival was statistically significantly worse in older patients with TNBC diagnosed at > 74 years compared to the younger group. To avoid selection bias with a more advanced tumor stage among elderly patients, the cut-off levels for the separation of younger and elderly were prospectively set to 40 and 74 years, respectively, as this is the age limits for the national mammogram screening program. In addition, when patients were split into five age groups, survival following treatment for primary TNBC was markedly worse for the older group, while the other four had a similar outcome.

We could not demonstrate that older patients had diagnose of BC at a more advanced stage. However, a larger proportion of 26% among the elderly were solely assessed clinically in the axilla. This is higher compared with a large registry based publication [15]. Although our patients are slightly older, only 12% of patients above 70 years were left without axillary surgery in their study. Thus, lymph node involvement among older might be underdiagnosed.

The incidence of TNBC is higher among younger patients. As BC is a much more common disease among older patients, there is still a substantial number of old patients with this diagnose. We show that the median age at diagnosis in this population-based cohort of TNBC was 60 years, which is close to the median age at diagnosis for all BC. A minor proportion in the older group of around 10% received chemotherapy, both in the adjuvant and palliative setting. In addition, postoperative radiotherapy was delivered to fewer older patients, but this is likely due to more mastectomies in this group and not to actively avoiding treatment. So far, chemotherapy is the sole systemic treatment with proved long-term efficacy for patients with TNBC, although newer compounds have been introduced in the palliative setting [16‐18]. Data from clinical trials where chemotherapy has been examined in the older population are few. Previous published studies have shown excess mortality among elderly BC patients and that chemotherapy is less commonly used than in younger individuals [18]. However, two large studies where patients in the US Surveillance, Epidemiology and End Results Program (SEER) database were investigated specifically addressed the use and effect of adjuvant chemotherapy in older patients, defined as > 65 years. The results concordantly showed a reduction in mortality by adjuvant chemotherapy in patients with hormone receptor-negative node-positive BC [19, 20], with consistent results in the subgroup of patients’ aged 70 years or older [18]. Patients studied received adjuvant chemotherapy during the 1990s, and both publications demonstrate an increase in chemotherapy use during the study periods [17, 18]. It is not possible to extract the TNBC patients’ as HER2 status not was determined during this period. With regard to type of chemotherapy, CMF-based regimens were the most common during the initial years, but this regimen was gradually replaced by anthracycline-containing regimens which also included taxanes for a proportion of patients [18]. Probably, these patients received more moderate adjuvant chemotherapy than todays’ recommendations, but still resulted in a patient gain.

A recently published large registry study from two regions in England determined the use of and effect of chemotherapy in > 10,000 patients above the age of 70 years from 2002 to 2012 [21]. Among patients aged 70–79 years, the study identified an increase in the use of chemotherapy from 2002 as well as a reduction in breast cancer mortality. Moreover, a prospective trial demonstrated that standard AC or CMF is better than Capecitabine in the adjuvant setting for elderly patients [22], while no advantage could be documented for anthracycline-containing chemotherapy compared with CMF [18]. Taken together, this data support that elderly TNBC patients in good general condition benefit from adjuvant chemotherapy and that less toxic regimens can be an efficient option.

Data in patients above 80 are scarce. In the English study, a minute proportion of patients > 80 years (1%) received chemotherapy [21] in agreement with the findings from the SEER database publications [17, 18]. We show that despite a median age of 81 years in our older group, the vast majority, almost 70%, died from their breast cancer and not from concomitant severe diseases. This is not true for BC overall where elderly patients often die from other diagnoses [23]. Since TNBC is an aggressive disease and the risk of recurrence is high during the first years after diagnosis, an older woman in a good performance status with TNBC often has a higher risk of dying from TNBC than from other causes.

Survival following distant metastasis was poor in our study for both age groups. As expected and potentially as a result of lack of primary treatment, the distant disease-free interval was shorter for elderly compared with younger patients with 13 and 24 months, respectively. Older patients had a poorer survival following distant metastasis, and this was significant when comparing patients over and under 75 years, concordant with previous published data including all breast cancer subtypes [24]. Taken into account that adjuvant chemotherapy was not given to a majority of the older, these patients may have had a less resistant disease at recurrence. However, our data show that the vast majority (90%) of the older patients did not even start any palliative chemotherapy. We cannot determine if this was due to a more advanced stage with a poor general condition, a result of delayed diagnosis or a general hesitation to initiate palliative chemotherapy from physicians or patients’ choice.

The metastatic patterns were in concordance with previous trials [8, 25]. We found no clear difference in metastatic spread other than the more frequently registered brain and liver metastases in the younger group. TNBC is a subgroup with a high risk of brain metastasis, in our whole population 40%, despite the commonly short post-recurrence survival. This is concordant with recently published data from a French multicenter database with more than 4000 BC patients with brain metastasis [26]. More than 50% of the younger patients were diagnosed with brain metastasis. The corresponding number among older was less than 20%, and the difference is probably partly a result of more frequent use of brain CT and MRI scans among younger patients. Moreover, older patients have a shorter survival time following recurrence, and thus death may occur before brain metastases have been developed.

We investigated if there was any biological difference between the age groups. Even though a majority of older patients had grade III tumors with a high proliferation rate, both grade and Ki67 were higher among young patients with TNBC. This is in concordance with a previous larger study comparing biology and survival in TNBC < 70 and ≥ 70 years [27]. The authors conclude that the equal survival found in this study, although no patient ≥ 70 years received adjuvant chemotherapy and 47% among the younger did, was explained by less aggressive tumors among old patients. The age comparison in our study is not identical, but we found a significant worse survival among TNBC > 75 compared with ≤ 75. This is probably explained by a higher proportion of adjuvant/neoadjuvant chemotherapy given among patients ≤ 75 years (92% among young and 11% among old patients) in our study, which is more recent. A previous study by Dreyer et al. [28] found a relation between age and histological subtype in patients with TNBC. Apocrine and lobular breast cancer seemed to be more common among old patients. However, subtypes other than ductal are rare and similarly we were in our study unable to find a statistical difference in histology in the different age groups.

Since a larger proportion of the young TNBC patients harbor a BRCA mutation, this could affect the prognosis and sensitivity to chemotherapy with more highly proliferative tumors in the younger group. The effect on BRCA mutation on outcome in TNBC is however debatable and a larger prospective trial showed that survival after 2 years is more favorable among BRCA mutation carriers with TNBC compared with wild-type TNBC but not at 5 years [29]. We do not consider it plausible that this affects our results in a significant way.

Our study has strength and weaknesses. Unlike register-based studies, in this up-to-date population-based cohort study with real-world data, we followed our patients for sufficient amount of time with medical record review and only 10 out of 524 patients were lost to follow-up. This provides strength to our results. To our very best knowledge, this is the first study that compared the biology, metastatic patterns and outcome of elderly versus younger patients with TNBC. The number of patients is small and this is especially obvious in the comparison between younger and older in the metastatic setting. Moreover, we lack information on BRCA status, further subgrouping of TNBC by use of gene expression profiling [30], as well as data on comorbidities and functional status.

In summary, older patients with TNBC in our study have a poor prognosis and as much as 68% of deaths in the older group were caused by TNBC. This can only be explained by the low proportion of patients in the older group that received adjuvant chemotherapy. Public health has improved in Sweden and life expectancy for women is now 84 years [31]. Chemotherapy should be considered in fit older patients with TNBC. However, there are still older patients with TNBC who are not suitable for adjuvant chemotherapy. General guidelines often do not take higher age into account. There may also be a hesitation from oncologists, patients and relatives when standard adjuvant chemotherapy is considered since older patients are more sensitive to adverse events [32]. This supports the use of the guidelines how to treat elderly with breast cancer [33] that includes geriatric assessment which we believe, at least in our region, is an underused but valuable tool.