Do site and type of metastasis in breast cancer show a changing pattern with increased age? A cross comparison of clinicopathological characteristics between age groups

An estimated 6.6%% of all breast cancers (BCs) occur in women younger than 40 years old [1]; moreover, up to 50% of all BCs are considered to occur in women over 65 years old [2, 3].

Generally, BC in the elderly population is considered to have a better clinical presentation regarding receptor status [human epidermal receptor2 (HER2) expression], grade, lymph node involvement, and tumor size [4, 5]; however, some studies have shown that older individuals are diagnosed at later stages [6]. Considering associated comorbidities and appropriate treatment of choice in the elderly population, BC’s are highly undertreated in this population, mainly due to the physician and patient’s choice [7].

On the other hand, BC’s in younger populations present another dilemma and current evidence support the idea that younger patients have overall worse prognosis, although the underlying etiology for this phenomena remains to be unclear [1, 8]. Different gene expression and BC characteristics have advocated researchers to consider the disease a separate entity among this population [9, 10].

To date, studies comparing the two populations have included limited data on patient- and disease-specific characteristics [5, 11, 12]. More importantly, multiple aspects of BC specifics remain to be addressed in these populations. Considering that previous literature has shown BC to be different between young and older patients, in here, we hypothesized that pattern of metastasis and, more importantly, location of metastasis maybe different between these groups of patients.

In an attempt to clarify differences in BC between different ages, in this study, we classified individuals in three age groups and determined clinicopathological and baseline differences between these groups in a comprehensive manner.

Overall, 564 individuals entered group 1, 4519 group 2, and 670 group 3. Comparison of baseline and clinical characteristics between age groups showed that the groups were significantly different regarding pathological sub-type, tumor necrosis, invasion status, ER receptor status, HER2 expression status, BC stage, BC grade, grade of nucleus, operation type, site of metastasis, lymph node management, chemotherapy, radiotherapy, hormone therapy, and type of hormone therapy (p < 0.05).

Those younger than 30 years old had higher rates of ER-negative individuals compared to older groups (31.2 vs. 25.5 and 23.1% for groups 2 and 3, respectively) (p = 0.001). HER2-positive individuals were significantly lower in the older age groups (19.2, 27.2, and 29.9% for groups 3, 2, and 1, respectively) (p < 0.001).

Individuals in the younger groups had statistically higher number of individuals with stage 3 BC (31.1, 25.6, and 19.7% for those in groups 1, 2, and 3) (p = 0.016) and higher rates of individuals with grade 3 BC (27.4 vs. 20.6 and 16.5% for groups 1, 2, and 3, respectively) (p = 0.001). This was the same with nucleus grade, as the youngest age group had the highest number of individuals with grade 3 nucleus (43.1 vs. 34.5 and 27.6% for groups 1, 2, and 3, respectively) (p < 0.001) (Table 1).

Individuals in the older groups had significantly lower rates of radiotherapy (58.4 vs. 81.6 and 89.7% for groups 3, 2, and 1, respectively) (p < 0.001). Those in the oldest group had the lowest rate of chemotherapy (76.7 vs. 97 and 97.4% for groups 2 and 1, respectively) (p < 0.001).

Comparison of recurrence rates between age groups showed that those in younger age groups had higher rates of individuals with less than 5 years recurrence compared to those in older age groups (16.3, 11.7, and 8.9%, for groups 1, 2, and 3, respectively) (p = 0.023) (Table 2).

In here, we compared breast cancer characteristics between three age groups to determine changes in pattern of metastasis and to compare clinicopathological characteristics between these groups.

One of the most interesting results in our study was that although BC in younger individual is considered rare [14], we had a very high rate of individuals in the lower than 30 years old group (almost 10%), which opens a wide window for future research among these patients. This is attributed to the younger age of first presentation of BC in the Iranian population which is significantly higher than that of the Western world [15, 16].

In a recent multi-centered study by Sabiani et al. [5], pathological features of BC were compared among 5815 patients in age groups of ≤ 35 years old, 35–40 years old, 40–45 years old, and between 45 and 50 years old. They found ≤ 35-year-old patients to have higher rates of mastectomy (29.3 vs. 24.5, 21.1, and 21.1%, respectively); moreover, younger individuals had lower rates of SLNB (50.2 vs. 59.5, 65.9, and 69.8%, respectively), higher rates of AND (73.2 vs. 67.4, 63, and 57.2%, respectively), higher rates of larger than 5-cm tumors (6.1 vs. 5.2, 4.8, and 4.8%, respectively), higher rates of grade 3 BC (43.5 vs. 34.2, 25.5, and 19.6%, respectively), higher rates of lymphovascular invasion (50.6 vs. 40.9, 24.2, and 29.5%, respectively), more HER2 positive individuals (22.1 vs. 16.3, 10.9, and 10.2%, respectively), and lower rates of negative LN (51.4 vs. 56, 58.6, and 62.1%, respectively); received less hormone therapy (52.3 vs. 60.6, 67.5, and 72.6%, respectively); had more chemotherapy (81.2 vs. 68.9, 57.5, and 50.2%, respectively); and had higher rates of distant metastasis (21.8 vs. 12.6, 9.1, and 7.7%, respectively) compared to the older groups, respectively. In our study, we found that those lower than 30 years old had higher rates of HER2 expression; lower ER- and PR-positive individuals; higher stage and grade; higher rates of BCS, SLNB, and radiotherapy; lower rates of hormone therapy; and higher lymphovascular invasion compared to 30–60-year-old and > 60-year-old BC patients. The most important factor for the difference between the two studies relates to the different age groups (different cut-off points) which were studied, although some variables such as HER2 expression, stage and grade, and hormone therapy show a similar trend in both studies.

In a study by Anders et al. [4], clinicopathological and prognostic determinants were compared between patients younger than 45 years old and older than 65 years old from four different cohort databases. They found that younger individuals had lower incidence of ER positivity, higher grades of tumor, larger tumor sizes, and higher rates of LN positivity. Their findings were very similar to our results when comparing those younger than 30 years old and those older than 60 years old.

In a cohort of 390 individuals [11], BC patient between 65 and 75 years old and individuals older than 75 years old were evaluated. They found that older individuals had larger mean tumor sizes (2.3 vs. 1.7 cm); moreover, regarding sub-types of cancer, similar to our study, in which we found mucinous sub-type to have an increasing pattern by increased age, they also found mucinous sub-type of cancer to be more frequent in their older population (8 vs. 5%). Similarly, they found a higher rate of BCS in their younger population (83 vs. 66%; p < 0.001); they also found lower rates of chemotherapy (4 vs. 31%; p < 0.001) and radiotherapy (44 vs. 67%; p < 0.001) among older patients. Other studies have also evaluated elderly patients, in a retrospective cohort of 317 patients from South Korea [17]; authors did not find any difference between age groups of older than 65 years old and younger than 65 years old regarding stage, nuclear grade and histological grade. However, they did find those younger than 65 years old to have higher rates of lymphovascular invasion, adjuvant chemotherapy, radiotherapy, and endocrine therapy.

As the main objective of our study, we found that pattern of metastasis changes with increased age. Younger patients had higher rates of metastasis to the brain and mixed areas (more than one site of metastasis), while those older than 60 years old had higher rates of metastasis to the lung. This brings up the question whether age is an independent predictor of organ of metastasis or is site of metastasis the result of other clinicopathological determinants which differ between age groups. Among the reasons which may be attributing to the different pattern of recurrence between the age groups may be that number of individuals with triple negative BC was higher in the younger age groups, as one study showed that those with triple negative BC show higher rates of brain metastasis [18]. Although this may be attributing to these phenomena, further research is needed to evaluate the cause.

This finding may be significant in clinical practice, if proven by further research, those in younger age groups, considering their higher rates of metastasis to the brain, may require specific metastasis work-ups such as brain MRI and brain CT to rule-out metastasis to the brain which may be different from older individuals. On the other hand, older individuals with BC may need work-ups such as lung CT to rule-out metastasis to the lung, considering their higher rates of metastasis to this organ. This would significantly aid in determining prognosis and overall management pattern of BCs for clinicians.

This study was not without limitation. In here, although we included data from the largest breast cancer registry in Iran, our sample may not be representative of the whole region. We used a different cut-off for age groups in our study to compare individuals with BC and although in BC, younger than 40 is considered to be young, in the Iranian population, multiple studies have shown age of first presentation to be much younger than that of the mean global estimate [15, 16, 19]. So in this study, we categorized patients as those younger than 30 years old, 30–60, and older than 60 years old. More importantly, the different classification cut-offs do not compromise the main goal of the study which was to evaluate the overall pattern of recurrence with increased age.

Our study showed that multiple breast cancer parameters including mucinous sub-type of breast cancers, tumor necrosis, PR positivity, ER positivity, mastectomy rates, AND rates, and use of laterazole showed a higher rate among individuals in older age groups. On the other hand, parameters including lymphovascular invasion, HER2 positivity, stage and grade of breast cancer, mastectomy rates, SLNB and both SLNB and AND rates, chemotherapy and radiotherapy use, hormone therapy, and use of tamoxifen showed lower rates in older populations.

In conclusion, we found that multiple clinicopathological indexes differ in three age groups of younger than 30 years old, between 30 and 60 year olds, and older than 60 year old patients with BC. Our results showed that pattern of metastasis changes with increased age, which calls for future studies to determine the exact effect of age on mechanism of metastasis.