Breast cancer in adolescents and young adults has a specific biology and poor patient outcome compared with older patients

Abstract

Background: We aimed to clarify the features of adolescents and young adults (AYA: younger than 40 years old) breast cancer (BC) compared with other age groups in estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative BC, given the effects of age-related hormonal status.

Methods: The cohorts analyzed were divided into AYA (15-39 years old), perimenopausal (40-54 years old), menopausal (55-64 years old), and old (65+ years old). Clinicopathological and biological features were analyzed using gene set variation analysis and xCell algorithm using transcriptome profiles from large public databases of ER-positive/HER2-negative BC (METABRIC; n = 1353, SCAN-B; n = 2381).

Results: In the ER-positive/HER2-negative subtype, pathological lymph node positivity, and Nottingham grade 3 were higher among AYA (all P < 0.001). AYA patients had a trend toward worse disease-specific and overall survival, particularly compared with the perimenopausal group. Estrogen response late signaling decreased with age (all P ≤ 0.001 in both METABRIC and SCAN-B cohorts). AYA was associated with significantly higher BRCAness and DNA repair than the other groups (all P < 0.05 in both cohorts). AYA significantly enriched cell proliferation-related and procancerous gene sets [mTORC1, unfolded protein response, and phosphoinositide 3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/mTOR) signaling] when compared with the others (all P < 0.03 in both cohorts). Interestingly, these features have also been observed in tumors <2 cm. Infiltration of CD8⁺, regulatory, T helper type 2 cells, and M1 macrophages was higher, while M2 macrophages were lower in AYA (all P < 0.03 in both cohorts). Finally, ER-positive/HER2-negative BC in AYA patients has different features of gene mutations, including AHNAK2, GATA3, HERC2, and TG, which were observed at a higher rate in AYA, and KMT2C, which was observed at a lower rate in AYA, compared with other age groups.

Conclusions: ER-positive/HER2-negative BC in AYA was highly proliferative with high immune cell infiltration compared with the other age groups.

Keywords: AYA; BRCAness; adolescent and young adult generation; breast cancer; gene expression; generation gap; hallmark signaling.

Figure 1

Patient survival outcomes by the age groups of whole and estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer (BC). Kaplan–Meier curves with log-rank disease-specific survival (DSS) and overall survival (OS) among each age group [red, adolescents and young adults (AYA; 15-39 years old); blue, perimenopausal (peri; 40-54 years old); olive, menopausal (meno; 55-64 years old); and gray, old (65+ years old)] in the whole (number of cases: AYA/peri/meno/old = 116/532/512/743) and ER-positive/HER2-negative BC of the METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) cohort.

Figure 2

Comparison of the level of estrogen response, DNA repair, BRCAness, and other procancerous gene signaling among adolescents and young adults (AYA) and the other age groups in the estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative subtype. Boxplots of the score level of (A) gene set defining late response to estrogen; (B) DNA repair signaling; (C) BRCAness score; (D) cell proliferation-related gene sets, including G2M checkpoint, E2F targets, and MYC targets v1; and (E) other procancerous gene sets, including mTORC1, unfolded protein response (UPR), and PI3K/AKT/mTOR, by each age group [red, AYA; blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in both METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) and SCAN-B (GSE96058, number of cases: AYA/peri/meno/old = 52/604/585/1140) cohorts. The P-value was calculated to compare AYA with other age groups with ER-positive/HER2-negative BC using the Mann–Whitney U test. AKT, protein kinase B; mTOR, mammalian target of rapamycin; PI3K, phosphoinositide 3-kinase.

Figure 2

Comparison of the level of estrogen response, DNA repair, BRCAness, and other procancerous gene signaling among adolescents and young adults (AYA) and the other age groups in the estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative subtype. Boxplots of the score level of (A) gene set defining late response to estrogen; (B) DNA repair signaling; (C) BRCAness score; (D) cell proliferation-related gene sets, including G2M checkpoint, E2F targets, and MYC targets v1; and (E) other procancerous gene sets, including mTORC1, unfolded protein response (UPR), and PI3K/AKT/mTOR, by each age group [red, AYA; blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in both METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) and SCAN-B (GSE96058, number of cases: AYA/peri/meno/old = 52/604/585/1140) cohorts. The P-value was calculated to compare AYA with other age groups with ER-positive/HER2-negative BC using the Mann–Whitney U test. AKT, protein kinase B; mTOR, mammalian target of rapamycin; PI3K, phosphoinositide 3-kinase.

Figure 2

Comparison of the level of estrogen response, DNA repair, BRCAness, and other procancerous gene signaling among adolescents and young adults (AYA) and the other age groups in the estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative subtype. Boxplots of the score level of (A) gene set defining late response to estrogen; (B) DNA repair signaling; (C) BRCAness score; (D) cell proliferation-related gene sets, including G2M checkpoint, E2F targets, and MYC targets v1; and (E) other procancerous gene sets, including mTORC1, unfolded protein response (UPR), and PI3K/AKT/mTOR, by each age group [red, AYA; blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in both METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) and SCAN-B (GSE96058, number of cases: AYA/peri/meno/old = 52/604/585/1140) cohorts. The P-value was calculated to compare AYA with other age groups with ER-positive/HER2-negative BC using the Mann–Whitney U test. AKT, protein kinase B; mTOR, mammalian target of rapamycin; PI3K, phosphoinositide 3-kinase.

Figure 3

Immune cell infiltrations in estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer (BC) by the age groups. Boxplots of infiltration fraction of (A) anticancerous immune cells; CD8⁺ T cells, CD4⁺ memory T cells, and M1 macrophages, and (B) procancerous immune cells; regulatory T cells, T helper type 2 (Th2) cells, and M2 macrophages by each age group [red, adolescents and young adults (AYA); blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in both METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) and SCAN-B (GSE96058, number of cases: AYA/peri/meno/old = 52/604/585/1140) cohorts. The Kruskal–Wallis test was used for the analysis.

Figure 4

Comparison of the level of cell proliferation-related and procancerous signaling among adolescents and young adults (AYA) and the other age groups in T-categories 1 and 2 of estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer (BC). Boxplots of the score level of (A) cell proliferation-related gene sets, including G2M checkpoint, E2F targets, and MYC targets v1, and (B) other procancerous gene sets, including mTORC1, unfolded protein response (UPR), and PI3K/AKT/mTOR, by each age group [red, AYA; blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in both METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) and SCAN-B (GSE96058, number of cases: AYA/peri/meno/old = 52/604/585/1140) cohorts. The P-value was calculated to compare AYA with each of the other groups in T-categories 1 and 2 of ER-positive/HER2-negative BC using the Mann–Whitney U test. AKT, protein kinase B; mTOR, mammalian target of rapamycin; PI3K, phosphoinositide 3-kinase.

Figure 4

Comparison of the level of cell proliferation-related and procancerous signaling among adolescents and young adults (AYA) and the other age groups in T-categories 1 and 2 of estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer (BC). Boxplots of the score level of (A) cell proliferation-related gene sets, including G2M checkpoint, E2F targets, and MYC targets v1, and (B) other procancerous gene sets, including mTORC1, unfolded protein response (UPR), and PI3K/AKT/mTOR, by each age group [red, AYA; blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in both METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) and SCAN-B (GSE96058, number of cases: AYA/peri/meno/old = 52/604/585/1140) cohorts. The P-value was calculated to compare AYA with each of the other groups in T-categories 1 and 2 of ER-positive/HER2-negative BC using the Mann–Whitney U test. AKT, protein kinase B; mTOR, mammalian target of rapamycin; PI3K, phosphoinositide 3-kinase.

Figure 5

Comparison of the mutation rate among adolescents and young adults (AYA) and the other age groups in estrogen receptor (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer (BC). Barplots of the mutation rate among each group of 20 genes with high mutation rates in breast cancer [red, AYA; blue, perimenopausal (peri); olive, menopausal (meno); and gray, old] in the METABRIC (number of cases: AYA/peri/meno/old = 37/336/367/613) cohort. The P-value was calculated to compare AYA with each of the other groups of ER-positive/HER2-negative BC using Fisher’s test. ∗ indicate P value smaller than 0.05.