Molecular analyses of triple-negative breast cancer in the young and elderly

Abstract

Background: Breast cancer in young adults has been implicated with a worse outcome. Analyses of genomic traits associated with age have been heterogenous, likely because of an incomplete accounting for underlying molecular subtypes. We aimed to resolve whether triple-negative breast cancer (TNBC) in younger versus older patients represent similar or different molecular diseases in the context of genetic and transcriptional subtypes and immune cell infiltration.

Patients and methods: In total, 237 patients from a reported population-based south Swedish TNBC cohort profiled by RNA sequencing and whole-genome sequencing (WGS) were included. Patients were binned in 10-year intervals. Complimentary PD-L1 and CD20 immunohistochemistry and estimation of tumor-infiltrating lymphocytes (TILs) were performed. Cases were analyzed for differences in patient outcome, genomic, transcriptional, and immune landscape features versus age at diagnosis. Additionally, 560 public WGS breast cancer profiles were used for validation.

Results: Median age at diagnosis was 62 years (range 26-91). Age was not associated with invasive disease-free survival or overall survival after adjuvant chemotherapy. Among the BRCA1-deficient cases (82/237), 90% were diagnosed before the age of 70 and were predominantly of the basal-like subtype. In the full TNBC cohort, reported associations of patient age with changes in Ki67 expression, PIK3CA mutations, and a luminal androgen receptor subtype were confirmed. Within DNA repair deficiency or gene expression defined molecular subgroups, age-related alterations in, e.g., overall gene expression, immune cell marker gene expression, genetic mutational and rearrangement signatures, amount of copy number alterations, and tumor mutational burden did, however, not appear distinct. Similar non-significant associations for genetic alterations with age were obtained for other breast cancer subgroups in public WGS data. Consistent with age-related immunosenescence, TIL counts decreased linearly with patient age across different genetic TNBC subtypes.

Conclusions: Age-related alterations in TNBC, as well as breast cancer in general, need to be viewed in the context of underlying genomic phenotypes. Based on this notion, age at diagnosis alone does not appear to provide an additional layer of biological complexity above that of proposed genetic and transcriptional phenotypes of TNBC. Consequently, treatment decisions should be less influenced by age and more driven by tumor biology.

Trial registration: ClinicalTrials.gov NCT02306096.

Keywords: Age at diagnosis; Gene expression; Mutational signatures; Mutations; PD-L1; Patient outcome; TILs; Triple-negative breast cancer.

Conflict of interest statement

The authors declare that they have no competing interests with exception of Johan Hartman who has received speakers’ honoraria and travel support from Roche, advisory board fees from MSD, Novartis, and Roche, and institutional research grants from Cepheid and Novartis, and is the co-founder and shareholder of Stratipath AB.

Figures

Fig. 1

Study scheme. a Analyses performed in the SCAN-B TNBC cohort together with investigated main sample groups. b Analyses performed in the external Nik-Zainal et al. [3] cohort together with investigated main sample groups. In both panels, sample size numbers for patient groups refer to the largest set of patients available for at least one of the specified analyses. Specific sample size numbers are provided in the detailed results and the “Methods” section. References to the main figures and tables presenting results are provided for each analysis. HRD+: HRDetect-high, HRD-: HRDetect-low, TILs: tumor-infiltrating lymphocytes, TMB: tumor mutational burden, CNA: copy number alteration, Lum A: Luminal A, Lum B: Luminal B

Fig. 2

Patient age versus BRCA1 deficiency and gene expression subtypes in TNBC. a Cumulative proportion of patients with BRCA1 deficiency (BRCA1 hypermethylation or BRCA1-null tumors) and non-BRCA1-deficient patients versus age at diagnosis. Red triangles indicate age at diagnosis for 50%, 80%, and 90% of BRCA1-deficient patients. b Cumulative proportions of patients in HRDetect groups versus age at diagnosis. c Principal component analysis of gene expression data for 232 SCAN-B cases using 19,102 RefSeq genes and different molecular and clinicopathological factors, including age at diagnosis (years: Age) and stratified age groups (10-year intervals: Age groups). d Cumulative proportions of PAM50 subtypes versus age at diagnosis. e Cumulative proportions of IntClust 10 subtypes versus age at diagnosis. f Cumulative proportions of TNBCtype subtypes versus age at diagnosis. g Heatmap of 1179 genes differentially expressed between six 10-year interval age groups in 232 SCAN-B cases. Hierarchical clustering of cases (columns) and genes (rows) were performed using Pearson correlation as distance metric and ward.D linkage on mean-centered log2 transformed data with an offset of 0.1. The six top clusters were identified and labeled. MKI67: Ki67. Steroid response: Scores according to the steroid response metagene [40]. h From left to right: Estimations of tumor cell content from WGS (ASCAT method), epithelial, stromal, B cell lymphocyte, and endothelial cell proportions from CIBERSORTx versus the hierarchical clusters in g. For age group definitions, “[” equals ≥, “)” equals <, and “]” equals ≤ for the value specified next to it

Fig. 3

Immune cell landscape of TNBC with respect to age at diagnosis. a CIBERSORTx estimated B cell proportions per sample versus stratified age groups in all cases (left), HRDetect-high, HRDetect-low, and PAM50 basal-like cases (right). Top axes indicate group sizes. Two-sided p values calculated using Kruskal-Wallis test. Linear regression modeling showing p value and slope coefficient (k) when using B cell proportion and continuous age in the model. b Heatmap of 102 immune cell marker genes in 232 SCAN-B cases using Pearson correlation and ward.D linkage. Hierarchical clustering of cases (columns) and genes (rows) was performed using Pearson correlation as distance metric and ward.D linkage on mean-centered log2 transformed data with an offset of 0.1. c TIL percentage estimated from whole-slide hemotoxylin and eosin-stained sections versus stratified age groups in all cases (left), HRDetect-high, HRDetect-low, and PAM50 basal-like cases (right). Top axes indicate group sizes. Two-sided p values calculated using Kruskal-Wallis test. Linear regression modeling showing p value and slope coefficient (k) when using TIL percentage and continuous age in the model. For age group definitions, “[” equals ≥, “)” equals <, and “]” equals ≤ for the value specified next to it. In panels a and c, separate results of a sensitivity analysis for trend are reported in red due to very small sample numbers across the full stratified age range

Fig. 4

Copy number alterations versus age at diagnosis in TNBC. a Copy number landscape of HRDetect-high patients < 50 years at diagnosis versus > 70 years at diagnosis. b Difference in amplification frequency of CCND1, CCNE1, EGFR, and MCL1 with age groups when analyzed in the total SCAN-B cohort. Two-sided p values calculated using chi-square test for trends in proportions. c Difference in mutation frequency of PIK3CA and TP53 with age groups when analyzed in the total SCAN-B cohort. Two-sided p values calculated using chi-square test for trends in proportions. d Proportions of amplified cases for CCND1, CCNE1, EGFR, and MCL1 according to HRDetect classification. e Proportions of mutated cases for PIK3CA and TP53 according to HRDetect (left), PAM50 (center), and TNBCtype (right) classifications. For age group definitions, “[” equals ≥, “)” equals <, and “]” equals ≤ for the value specified next to it

Fig. 5

Composite view of molecular and genetic phenotypes versus age at diagnosis in TNBC. a Integrative view of gene expression subtypes (PAM50 basal-like, LAR, IntClust 4, 10), HRD classification, BRCA1, BRCA2, PALB2, RAD51C, MCL1, CCND1, CCNE1, EGFR, PIK3CA, and TP53 alterations, mutational signatures (S1-S26), rearrangement signatures (RS1-RS6), patterns of insertion, and deletions versus age groups stratified by an underlying BRCA1 deficiency, BRCA2 deficiency, and HRDetect classification. BRCA1-null, BRCA2-null, and PALB2-null imply biallelic loss of the gene based on WGS. b Illustration of signature patterns in BRCA1-deficient tumors (light gray) and HRDetect-low/intermediate cases (white) stratified by age for proportion of insertions (left), rearrangement signature 6 (RS6, center), and mutational substitution signature 3 (S3, right, refitted by SigFit). RS6 is characterized by clustered rearrangements typically found in cases with driver amplifications, while S3 is associated with BRCA1/2 deficiency [41]. Top axes indicate group sizes. Top axes indicate group sizes. Two-sided p values are calculated by Kruskal-Wallis test per group. For age group definitions, “[” equals ≥, “)” equals <, and “]” equals ≤ for the value specified next to it