Skip to main content
Download PDF

Molecular profiling of a real-world breast cancer cohort with genetically inferred ancestries reveals actionable tumor biology differences between European ancestry and African ancestry patient populations

Breast Cancer Research volume 25, Article number: 58 (2023) Cite this article

Abstract

Background

Endocrine-resistant HR+/HER2- breast cancer (BC) and triple-negative BC (TNBC) are of interest for molecularly informed treatment due to their aggressive natures and limited treatment profiles. Patients of African Ancestry (AA) experience higher rates of TNBC and mortality than European Ancestry (EA) patients, despite lower overall BC incidence. Here, we compare the molecular landscapes of AA and EA patients with HR+/HER2- BC and TNBC in a real-world cohort to promote equity in precision oncology by illuminating the heterogeneity of potentially druggable genomic and transcriptomic pathways.

Methods

De-identified records from patients with TNBC or HR+/HER2- BC in the Tempus Database were randomly selected (N = 5000), with most having stage IV disease. Mutations, gene expression, and transcriptional signatures were evaluated from next-generation sequencing data. Genetic ancestry was estimated from DNA-seq. Differences in mutational prevalence, gene expression, and transcriptional signatures between AA and EA were compared. EA patients were used as the reference population for log fold-changes (logFC) in expression.

Results

After applying inclusion criteria, 3433 samples were evaluated (n = 623 AA and n = 2810 EA). Observed patterns of dysregulated pathways demonstrated significant heterogeneity among the two groups. Notably, PIK3CA mutations were significantly lower in AA HR+/HER2- tumors (AA = 34% vs. EA = 42%, P < 0.05) and the overall cohort (AA = 28% vs. EA = 37%, P = 2.08e−05). Conversely, KMT2C mutation was significantly more frequent in AA than EA TNBC (23% vs. 12%, P < 0.05) and HR+/HER2- (24% vs. 15%, P = 3e−03) tumors. Across all subtypes and stages, over 8000 genes were differentially expressed between the two ancestral groups including RPL10 (logFC = 2.26, P = 1.70e−162), HSPA1A (logFC = − 2.73, P = 2.43e−49), ATRX (logFC = − 1.93, P = 5.89e−83), and NUTM2F (logFC = 2.28, P = 3.22e−196). Ten differentially expressed gene sets were identified among stage IV HR+/HER2- tumors, of which four were considered relevant to BC treatment and were significantly enriched in EA: ERBB2_UP.V1_UP (P = 3.95e−06), LTE2_UP.V1_UP (P = 2.90e−05), HALLMARK_FATTY_ACID_METABOLISM (P = 0.0073), and HALLMARK_ANDROGEN_RESPONSE (P = 0.0074).

Conclusions

We observed significant differences in mutational spectra, gene expression, and relevant transcriptional signatures between patients with genetically determined African and European ancestries, particularly within the HR+/HER2- BC and TNBC subtypes. These findings could guide future development of treatment strategies by providing opportunities for biomarker-informed research and, ultimately, clinical decisions for precision oncology care in diverse populations.

Introduction

Breast cancer (BC) is the most frequently diagnosed cancer in the USA and the second most common cause of cancer-related death in women [1], although there have been significant advances in treatment strategies over the past few decades. In addition to cytotoxic chemotherapy, endocrine therapy, and radiation therapy, in the molecular era, BC patients often receive targeted treatments based on their unique cancer biology in both adjuvant and neoadjuvant settings. For instance, anti-Human Epidermal Growth Factor Receptor 2 (HER2) therapy combined with chemotherapy has become the standard of care in HER2-positive (HER2+) BCs [2, 3]. Hormone receptor-positive (HR+)/HER2- tumors have historically received endocrine therapy and, more recently, CDK4/6 inhibitors as targeted treatments [4, 5]. Some molecularly targeted therapies have also emerged for triple-negative breast cancer (TNBC), such as poly-ADP ribose polymerase inhibitors for BRCA-mutated tumors [6, 7] and checkpoint inhibitors for PD-L1+ tumors [9, 10]. Despite these advances, most patients with metastatic BC develop drug resistance or rapidly progress shortly after initiating guideline-informed systemic therapy. Increasingly, genomic testing is used to expand treatment options to overcome de novo or acquired resistance. Patients with HR+/HER2- tumors harboring a PIK3CA mutation, for example, experience a prolonged progression-free survival when treated with a phosphoinositide-3-kinase (PI3K) inhibitor in combination with endocrine therapy [8]. However, there is still much room for improvement in endocrine-resistant HR+/HER2- BC and TNBC survival due to their aggressive natures and relatively limited treatment options [9, 10].

TNBC is much less common than the HR+/HER2- subtype in the US population [11], but disproportionately affects patients of African ancestry (AA) compared to those of European ancestry (EA). Moreover, AA patients experience a 40% higher mortality rate than EA patients across all subtypes [11], despite lower incidence rates of BC overall. The complex biologic and social drivers causing these disparities are beginning to be revealed through rigorous studies. While socioeconomic factors like inadequate access to quality care partially contribute to higher mortality, outcome disparities remain even after adjusting for these factors [12]. Several studies point to biological differences in the molecular drivers and evolutionary trajectory of BC based on ancestry. For example, population-based studies have observed higher rates of germline BRCA1 mutations in AA compared to EA patients [13]. In addition to germline mutations, somatic mutational differences have also been reported. An analysis of tumor sequencing data in the Cancer Genome Atlas conducted by our group revealed a higher prevalence of TP53 mutations in AA patients and a lower prevalence of alterations in PIK3CA [14]. In this same dataset, TP53 mutations were found to be a positive predictor for recurrence. These data and similar reports demonstrate a significant contribution of genomic differences to the mortality gap between AA and EA patients with BC and point to differential drivers of disease that may be potential therapeutic targets [15].

There is a paucity of molecular and clinical data from underserved and understudied populations, and the few studies using molecular assessments have been underpowered to demonstrate prognostic value in AA patients with BC in the USA [16, 17]. Analyses of comprehensive, large-scale oncology databases are needed to close the knowledge gap and address the unmet clinical need of diverse patient populations. Here, we performed a retrospective analysis of genomic and transcriptomic breast tumor sequencing data from Tempus’ large database and compared results between AA and EA patients. Furthermore, we estimated genetic ancestry employing ancestry-informative markers (AIMs) tailored specifically to our genomic data [18], which can fill gaps in race/ethnicity metadata from electronic health records (EHRs), provide more meaningful biological insights, and increase data availability in under-researched ancestral populations. Combining ancestry estimations, EHR metadata, genomics, and transcriptomics in this real-world cohort may guide the future development of treatment strategies by providing data for biomarker-informed research and precision cancer care.

Methods

Cohort selection

A sample of de-identified records from patients with TNBC or HR+/HER2- BC in the Tempus Database was randomly selected (N = 5000). A subset of those records was included for analyses after applying relevant inclusion criteria and performing genomic ancestry estimates to identify AA and EA patients (n = 3433). Clinical and molecular data were abstracted from patient records as previously described [19]. To be included in the cohort, all patients required results from either the Tempus xT (648-gene targeted panel) or Tempus xE (whole-exome) next-generation sequencing (NGS) assay. Furthermore, only AA or EA patients with known BC subtypes, stages at diagnosis, treatment histories, ages, biopsy sites at the time of testing, and histology (lobular vs. ductal) were selected.

Classification of patients by genetic ancestry

Although a portion of the patient records contained race and ethnicity metadata obtained from order forms or by abstraction of clinical documents, we opted to estimate global genetic ancestry proportions in the cohort using NGS data. The intent of this approach was to maximize the amount of patient records available for analysis, as race and ethnicity data are sometimes scarce in clinical reports, but also to ensure consistency in comparisons between genetically relevant ancestries and molecular markers. Furthermore, genetic ancestry estimates generally complement self-reported race and/or ethnicity, as we have previously found the two metrics to be concordant [14]. From the representative sample of patient records described above (N = 5000), a supervised version of the ADMIXTURE algorithm [20] was applied to a predefined set of AIMs detected in the DNA sequencing data of each patient sample to estimate ancestry proportion likelihoods for five major continental regions: Africa, Americas, East Asia, Europe, and South Asia, as defined in the 1000 Genomes Project (Additional file 1: Table S1) [21]. When a matched normal tissue sample was available (n = 3358), the algorithm was run on both tumor and normal NGS data, finding highly concordant ancestry proportions with either specimen for each patient (Africa r = 0.999, Americas r = 0.9993, East Asia r = 0.9997, European r = 0.9998, South Asia r = 0.9977; Pearson’s r, Additional file 2: Fig. S1).

The ancestry inference algorithm uses 654 and 6711 AIMs overlapping the regions targeted by the Tempus xT and Tempus xE assays, respectively. Considering the selected AIMs and their allele frequencies, ancestry proportion likelihoods for the five aforementioned continental groups were calculated for each patient. Thresholds were then established to determine the genetic ancestry proportions required from each patient to be classified as AA or EA using a combination of admixture proportions reported in the literature for African-American and Hispanic/Latino groups in the USA [22], and by analyzing the ranges of genetic admixture present in patients with available race metadata from clinical records. Genetic ancestry was estimated as AA if likelihoods were > 20% African, < 10% American (Amerindian or Native American), and if the patient reached a total combined African plus European likelihood of > 70%. Conversely, ancestry was estimated as EA if likelihoods were > 80% European and < 10% American. Patient records with genetically determined ancestries other than AA or EA were excluded from subsequent analyses.

Molecular profiling and comparative analyses

Mutational prevalence from the Tempus xT or xE assay and transcriptional signatures from whole-exome capture RNA-seq were evaluated [23, 24]. Briefly, Tempus xT includes a targeted panel DNA assay evaluating 648 genes, whereas Tempus xE is a whole-exome DNA assay. Among the somatic single-nucleotide variants (SNVs), insertions/deletions (indels), copy number variants (CNVs), and select chromosomal rearrangements identified, variants were annotated as pathogenic or likely pathogenic based on evidence from the Tempus Database including findings from the literature and public databases of association with carcinogenesis or cancer, as previously described [23]. DNA analyses were restricted to the 648 genes included in both platforms, ultimately including 3425 samples from the overall cohort due to quality control (AA n = 621, EA n = 2804).

The expression of each gene was compared between AA and EA patients with the edgeR software (v3.34.1) in R (v4.1.0) [25], stratified by stage and subtype. Methodology was adapted to account for results from two different versions of the same RNA-seq assay that were included in the cohort. Although both versions were performed based on exome-capture with the same protocol, differences in probe design produced minor batch effects and were accounted for in statistical analyses. Evaluation of differential expression was conducted on samples measured with each assay separately. Then, computed P values per gene were combined using the wFisher method [26]. Additionally, log fold-changes (logFC) computed with each assay were combined using a weighted average by sample size for each assay. Results were filtered for genes that had the same logFC directionality between the two assays. There was no significant difference in the proportions of the two different versions of the RNA-seq assay between EA (53% of samples evaluated with version 1, 47% with version 2) and AA samples (49.6% of samples quantified with version 1, 50.4% with version 2).

Gene set enrichment scores were obtained for a series of relevant gene sets using a Python implementation of single-sample gene set enrichment analysis (ssGSEA) based on GSVA v1.46.0 [27]. ssGSEA calculates separate enrichment scores for each pairing of a sample and gene set. Each ssGSEA enrichment score represents the degree to which the genes in a particular gene set are coordinately up- or down-regulated within a sample. For this study, ssGSEA was run using two gene sets: the Molecular Signature Database (MSigDB) hallmark gene set [28], a curated collection of well-defined biological states or processes, and the oncogenic signature gene set [29], which includes signatures of cellular pathways often dysregulated in cancer. These enrichment scores were then compared between AA and EA patients using limma [30].

Statistical analysis

Differences between the mutational prevalence of the genes harboring the highest number of alterations across both AA and EA patients were evaluated by a Pearson's chi-squared test, stratified by subtype. When considering genes beyond the 15 most frequently mutated, P values were adjusted for multiple hypothesis testing. The significance of differential gene expression was evaluated by edgeR with Bonferroni correction applied to adjust for multiple testing [25], where a threshold of 0.01 for meta-analysis P values was applied resulting in a cutoff of P < 1.0e−06. Transcriptional signatures of AA and EA patients were compared and evaluated by limma with Bonferroni correction, where a threshold of 0.01 for P values was applied resulting in a cutoff of P < 4.2e−05. In the portion of the cohort with available race metadata from clinical records, genetic ancestry classifications were compared to metadata using a Pearson's chi-squared test.

Results

Cohort overview and genetically determined ancestries

From the initial cohort of patients with BC (N = 5000), 623 AA (12.5%) and 2810 EA (56%) patients were identified from NGS-based genetic ancestry inference. When considering the subset of patient records with available metadata of reported race (n = 2428), genetically determined ancestries were strongly correlated with the race metadata gathered from clinical settings (97% concordance in AA and 91% in EA, Pearson’s chi-squared test of independence P < 2.2e−16; Additional file 3: Table S2). Demographics and clinical characteristics of the final cohort comprising samples with genetically determined ancestries are presented in Table 1. The median ages at diagnosis were similar between AA and EA patients (53.74 and 56.01 years, respectively). The majority of samples sequenced were from cases with metastatic cancer, as stage IV was the most common stage reported in both groups with similar proportions for each ancestry (AA = 83.3% vs. EA = 84.4%, P = 0.53; Table 1). HR+/HER2- was the most prevalent subtype in both ancestry groups, although HR+/HER2- diagnoses were relatively more frequent in EA patients (AA = 35.5% vs. EA = 45.6%, P = 5.14e−06; Table 1). In contrast, TNBC was relatively more frequent in AA patients (AA = 23.0% vs. EA = 14.2%, P = 9.56e−08; Table 1), as expected based on overrepresentation of TNBC in AA populations [11]. Consistent with previous reports, AA patients had higher tumor grades than EA patients across the cohort (AA = 56.8% and EA = 42.6%, P = 5.59e−07; Table 1).

Table 1 Clinical characteristics of African and European ancestry patients
Full size table

Distinct patterns in the mutational landscapes of African and European ancestry breast cancers

We next evaluated the prevalence of cancer somatic variants (SNVs, indels, and CNVs) classified as pathogenic or likely pathogenic in each ancestral group to identify ancestry-specific differences in tumor DNA. Additional file 4: Table S3 contains the frequency of carriers for alterations in each of the genes compared, the P value from a direct comparison between the two ancestry groups, and a P value adjusted for multiple hypothesis testing, as well as the most frequent alterations for each ancestry group. In total, frequencies of alterations in 633 genes were compared between AA and EA patients across all subtypes, 625 genes were compared in HR+/HER2- disease, and 612 in TNBC. Considering the top 15 most frequently mutated genes across the entire cohort, TP53 was the most commonly mutated (AA = 59% vs. EA = 49%; Fig. 1A). When stratifying by subtype, though, the differences in TP53 frequencies were less pronounced than in the entire cohort (AA = 40% vs. EA = 37% for HR+/HER2- and AA = 90% vs. EA = 88% for TNBC; Fig. 1B and C, respectively).

Fig. 1
figure 1

Mutational Landscapes of African and European Ancestry Breast Cancer Patients (AA and EA, respectively). Oncoplots show the top 15 pathogenic mutations in A all patients (AA n = 621, EA n = 2804), B patients with HR+/HER2- disease (AA n = 220, EA n = 1279), and C patients with triple-negative breast cancer (TNBC) (AA n = 143, EA n = 400)

Full size image

Among HR+/HER2- disease, the most frequently mutated gene in both ancestral groups was PIK3CA, with a significantly higher rate in EA patients (AA = 34% vs. EA = 42%, P < 0.05; Fig. 1B). The rate of PTEN mutation was also significantly higher in EA than AA patients (AA = 6% vs. EA = 12%, P < 0.05; Fig. 1B). For all other significant differences identified within the HR+/HER2- subtype, AA tumors exhibited higher mutation frequencies, including in GATA3 (AA = 22% vs. EA = 14%), KMT2C (AA = 24% vs. EA = 15%), and FGFR1 (AA = 18% vs. EA = 12%) (P < 0.05 for all; Fig. 1B).

Nonsignificant trends were observed in TNBC patients, such as BRCA1 mutations being less prevalent in AA compared with EA patients (AA = 8% vs. EA = 12%, P = 0.13; Fig. 1C), and APOB being mutated at a higher rate (AA = 13% vs. EA = 8%, P = 0.09; Fig. 1C). Meanwhile, KMT2C mutations were significantly more prevalent in AA than EA patients with TNBC (AA = 23% vs. EA = 12%, P < 0.05; Fig. 1C).

Differential gene expression between African and European ancestry breast cancers

Using full-transcriptome RNA-sequencing data, we surveyed gene expression by BC subtype and stage, identifying over 8000 genes with a significantly different expression between the two ancestral groups when considering all subtypes and stages (genes for which P < 0.05; Additional file 5: Table S4). The P value and logFC (with EA expression as the reference point) for each differentially expressed gene are available in Additional file 5: Table S4. As expected, there were variations in the patterns of differentially expressed genes when stratifying the cohort by subtypes and stages (Fig. 2B and C).

Fig. 2
figure 2

Differential Gene Expression Between Patients of African and European Ancestry (AA and EA, respectively). The volcano plots show all differentially expressed genes between AA and EA patients for A all breast cancer subtypes (n = 3433), B HR+/HER2- samples (n = 1502), and C triple-negative samples (n = 543). Gene sets were plotted by log fold-changes (logFC, x-axes) and meta P values (y-axes). Significance cutoffs were defined for each group, and gene sets with significantly different enrichment between ancestral groups were highlighted in green

Full size image

Several genes included in clinical prognostic assays for HR+/HER2- cancer, such as the Oncotype DX Breast Recurrence Score test, were identified as differentially expressed between the two ancestral groups (Fig. 2B, Additional file 5: Table S4). BAG1 and BCL2 were expressed significantly higher in AA patients with HR+/HER2- disease (stages I-III and stage IV, respectively), which have both been associated with favorable outcomes [31, 32]. However, many genes related to poor survival were also expressed higher in AA patients compared to EA patients, specifically in the stage IV HR+/HER2- subset, including BIRC5 [33], CCNB1 [34], UBE2C [35], CENPA [36], and RRM2 [37] (Additional file 5: Table S4). Meanwhile, SCUBE2, a gene with high expression linked to favorable outcomes [38], was expressed significantly lower in AA compared to EA patients with stage IV HR+/HER2- tumors (Additional file 5: Table S4).

Although there were less well-established expression-based biomarkers for TNBC, many of the differentially expressed genes identified between ancestral groups could carry prognostic implications in BC or other cancers (Fig. 2C, Additional file 5: Table S4). Interestingly, some of the genes with the strongest logFCs were differentially expressed in most subtypes and stages, such as HSPA1A (“All Subtypes, All Stages” logFC = − 2.73, P = 2.43e−49) [39], RPL10 (“All Subtypes, All Stages” logFC = 2.26, P = 1.70e−162) [40], NUTM2F (“All Subtypes, All Stages” logFC = 2.28, P = 3.22e−196), and ATRX (“All Subtypes, All Stages” logFC = − 1.93, P = 5.89e−83) (Additional file 5: Table S4). Furthermore, genes with implications for therapeutic research, such as those associated with immunotherapy mechanisms, were found to be differentially expressed in TNBCs. For example, CD70 was significantly higher in AA compared with EA patients among those with stage IV TNBC (logFC = 1.40, P = 7.46e−07). Various HLA genes were also differentially expressed across the cohort. HLA-DRA, for instance, was significantly lower in AA patients of every subtype and stage combination except for the “All Subtypes, Stage IV” and “HR+/HER2-, Stages I-III” subsets (“All Subtypes, All Stages” logFC = − 1.47, P = 9.57E−16; Additional file 5: Table S4).

Transcriptional signature activation differences between African and European ancestry breast cancers

While individual gene expression differences were found between the ancestral groups, simultaneous up- or downregulation of related genes in a pathway can have broader implications for biological processes. To investigate, we measured the differential expression of predefined, biologically relevant gene sets between AA and EA patients using ssGSEA. Figure 3 contains volcano plots displaying a general overview of the differentially expressed gene sets observed. Of the 152 gene sets evaluated, 125 exhibited significantly different enrichment between AA and EA patients across the entire cohort (n = 3433, Additional file 5: Table S5) and when stratifying by stage (58 differentially enriched sets by ancestry in stages I-III, n = 542, Fig. 3A; 118 sets in stage IV, n = 2891, Fig. 3A). Differential enrichments were also observed between ancestral groups in the HR+/HER2- subtype (73 gene sets, n = 1502, Additional file 6: Table S5), many of which persisted when only assessing patients with stage IV disease (64 differentially enriched gene sets, n = 1287, Fig. 3B) but not in stages I-III (no differentially enriched gene sets, n = 215, Fig. 3B). Interestingly, no differences were observed when comparing gene set expression between ancestral groups within the TNBC subpopulation, even when stratifying by stage (Fig. 3C). A full list of gene sets with their associated logFCs and P values from ssGSEA is available in Additional file 6: Table S5.

Fig. 3
figure 3

Differential Enrichment of Hallmark and Oncogenic Signature Gene Sets Between African and European Ancestry Patients (AA and EA, respectively). The volcano plots show all differentially expressed gene sets between AA and EA patients for A all breast cancer subtypes (n = 3433), B HR+/HER2- samples (n = 1502), and C triple-negative samples (n = 543). Gene sets were plotted by log fold-changes (logFC, x-axes) and limma P values (y-axes). Significance cutoffs were defined for each group, and gene sets with significantly different enrichment between ancestral groups were highlighted in green

Full size image

From the gene sets in Additional file 6: Table S5, five differentially expressed sets were considered particularly relevant to BC biology: ERBB2_UP.V1_UP, LTE2_UP.V1_UP, HALLMARK_FATTY_ACID_METABOLISM, HALLMARK_ANDROGEN_RESPONSE, and HALLMARK_ESTROGEN_RESPONSE_LATE [28, 41]. Comparisons of ssGSEA results between ancestral groups for each of these BC-related gene sets are presented in Fig. 4, stratified by subtype (HR+/HER2- or TNBC) and stage (I–III or IV). Among stage IV patients with HR+/HER2- disease, ERBB2_UP.V1_UP (Fig. 4A, P = 3.95e−06), LTE2_UP.V1_UP (Fig. 4B, P = 2.90e−05), HALLMARK_ANDROGEN_RESPONSE (Fig. 4C, P = 0.0074), and HALLMARK_FATTY_ACID_METABOLISM (Fig. 4D, P = 0.0073) were significantly enriched in EA patients. Again, none of the above gene sets were differentially enriched in any TNBC groups or stage I-III HR+/HER2- patients. Furthermore, there were no significant differences in HALLMARK_ESTROGEN_RESPONSE_LATE enrichment between AA and EA patients within any subtype-stage combination (Fig. 4E).

Fig. 4
figure 4

Differential Enrichment of Breast Cancer-Related Gene Sets in African and European Ancestry Patients (AA and EA, respectively). Violin plots depicting five gene sets from the hallmark and oncogenic signature collections that were considered particularly relevant to breast cancer biology, including: A ERBB2_UP.V1_UP, B LTE2_UP.V1_UP, C HALLMARK_ANDROGEN_RESPONSE, D HALLMARK_FATTY_ACID_METABOLISM, and E HALLMARK_ESTROGEN_RESPONSE_LATE. Only patients with stage IV HR+/HER2- disease exhibited differential enrichment in these gene sets (A–D), with the exception of the estrogen response pathway (E), which had no significantly different enrichment between ancestral groups

Full size image

Discussion

The results from our large-scale study provide innovative insight by presenting genomic and transcriptomic differences between BC tumors from AA and EA patients, with findings stratified by clinical features in a real-world cohort. The findings indicate the utility of assessing molecular landscapes when conducting both basic research and clinical trials according to the specific ancestries, as well as reveal biologically different mechanisms between AA and EA patients. Importantly, the mechanisms identified carry implications for use of molecularly targeted therapy to broaden early access to clinical trials, including the use of combination therapies in patients with residual disease after neoadjuvant therapy to prevent metastases and late recurrences.

In this cohort, the frequency of TP53 mutation was higher in AA than EA, which is consistent with previous reports, but there were no significant differences between ancestries in the stratified subtypes [14, 42, 43]. While TP53 mutations were much more frequent in AA patients with TNBC, our results indicate that the higher frequency of TP53 mutation in AA tumors from the entire cohort is due to the higher incidence of TNBC in AA. We also observed a significantly lower mutation rate of PIK3CA in AA tumors, particularly in patients with HR+/HER2- tumors. Luminal-type tumors are more likely to have PIK3CA mutations, providing opportunities for molecularly targeted therapy to improve survival [44]. Several compounds targeting PIK3CA have been studied and the US Food and Drug Administration has approved the first PI3K inhibitor alpelisib for advanced luminal BC harboring the PIK3CA mutation [8]. While the frequency of PIK3CA mutation was lower, any AA patient with identifiable mutations would be expected to benefit from alpelisib, but lack of access to molecular diagnosis could continue to drive racial disparities in outcomes for these patients. This underscores the need to consider each patient as an individual and utilize NGS testing early to capture patients with unique tumor biology and druggable pathways who would benefit from molecularly targeted therapy and/or immunotherapy.

We observed trends suggesting differences in the rates of somatic BRCA1 and BRCA2 mutations. Previous studies have reported elevated frequency of BRCA1 and BRCA2 germline mutations among AA patients [45,46,47]; however, the somatic mutation differences by race have not been well documented up until now. While the differences were not significant, we observed a potential racial difference of somatic BRCA1 and BRCA2 mutations in both HR+/HER2- and TNBC subtypes. AA patients were more likely to have BRCA2 mutations within the HR+/HER2- subtype, whereas EA patients were more likely to have BRCA1 mutations within the TNBC subtype. These findings and previous studies of germline alterations might also have implications for DNA repair-targeted therapies and/or immunotherapies [48].

KMT2C, a member of the myeloid/lymphoid or mixed-lineage leukemia family that encodes a histone methyltransferase, is one of the most frequently mutated genes in HR+ BC [49]. The deletion of KMT2C has been reported to be associated with resistance to endocrine therapy and worse prognosis [50]. We observed that the frequency of KMT2C mutation was significantly higher in AA than EA among TNBC patients (23% vs. 12%) as well as patients with HR+/HER2- tumors (24% vs. 15%), suggesting that the loss of KMT2C function might disproportionately affect the survival outcome of AA patients in both subtypes. Previous reports focusing on mutational differences between races also indicated that GATA3, which has a critical role in the development of luminal type BC, is mutated in around 10% of both AA and EA patients [42, 43]. In contrast, 22% of AA patients in our cohort had tumors with mutated GATA3 in HR+/HER2- subtypes, which might contribute to the racial disparity of luminal BC prognosis.

From the full-transcriptome RNA-sequencing data, we identified significantly different expression of over 8000 independent genes between the two ancestral groups. Various genes were exclusively expressed according to subtype (HR+/HER2- or TNBC) or disease stage (stage I-III or stage IV). Here, it is worth mentioning that RPL10, which is responsible for DNA replication stress and promoting proliferation and oncogenesis [40], was expressed higher in AA patients throughout almost every subtype and stage. Whereas the higher expression of RPL10 is reported to be associated with poor prognosis mainly in hematologic malignancy [51, 52], its relation to prognosis in breast malignancy has not been fully investigated. This transcriptomic change in AA patients regardless of subtype or stage might explain the prognostic disparity of AA patients and could be a potential therapeutic target in BC. On the other hand, HSPA1A, a member of the heat shock protein 70 family, was significantly lower in AA patients in both HR+/HER2- BC and TNBC. The clinical significance of HSPA1A status is unknown in malignant tumors, including BC [53]. ATRX, which has a critical role in chromatin remodeling, was also expressed lower in AA patients in both HR+/HER2- BC and TNBC. Previous studies revealed that ATRX loss is associated with an increase in cancer aggressiveness [54]. Among a large number of genes with little genomic and clinical annotation, NUTM2F was significantly overexpressed in AA patients throughout subtypes and stages. Our comprehensive birds-eye view analysis has identified many potential genes to consider in future basic and clinical research.

When breaking down the enrichment analysis of hallmark gene sets into the subtypes and stages, 125 exhibited significantly different enrichment between AA and EA patients across the entire cohort. No significant differences were observed when comparing gene set expression between ancestral groups within the TNBC subpopulation, even when stratifying by stage. Although the gene sets evaluated here are informative markers of oncogenic pathways, there are other collections of curated gene sets available for analysis, such as the KEGG pathways, and future studies could include these to provide a more thorough investigation of differences between AA and EA patients. Nevertheless, notable differences in many mutations and individual gene expressions between the two ancestral groups in TNBC were demonstrated as an initial assessment.

Meanwhile, among the stage IV HR+/HER2- group, 10 differentially expressed gene sets were identified. As a translation from research to clinic, we revealed four differentially expressed gene sets relevant to BC treatment: ERBB2_UP.V1_UP, LTE2_UP.V1_UP, HALLMARK_FATTY_ACID_METABOLISM, and HALLMARK_ANDROGEN_RESPONSE [28, 41]. In the stage IV HR+/HER2- group, these pathways could contribute to worse prognosis of AA patients and, accordingly, would be worth investigating in further prognostic studies consisting of AA and EA patients.

Our findings could have implications for prognosis, response to therapy, and enrollment of diverse populations in clinical trials and precision oncology studies. If genomic and transcriptomic differences are not considered, applying the results of clinical trials on the studied population (e.g., EA) to another population (e.g., AA) could lead to suboptimal treatment decisions. The molecular distinctions identified here indicate that the efficacy of novel targeted therapy combinations may differ by patient ancestry, and thus highlight an opportunity to optimize neoadjuvant and adjuvant clinical trial design. Additionally, the availability of this data on a population level and for each patient should ultimately result in more clinical trial opportunities including early access to trials for more patients.

Some potential limitations of this study are the broad analyses and heterogeneity of the cohort, as the patients included in the study were highly selected and derived from various institutions and spanned multiple subtypes and stages. Another is the lack of treatment data, as there is likely a mix of treatment-naïve and treatment-refractory patients included in this study, and outcomes data. Nevertheless, in previous studies from our group, we examined NGS data using the Tempus xT assay in patients undergoing neoadjuvant chemotherapy from the Chicago Multiethnic Breast Cancer Cohort and found similar patterns [55]. Furthermore, the patient characteristics were not entirely balanced between AA and EA, considering there were higher-grade tumors and more TNBCs in the AA patient group. Although our large sample size and stratifications by stage and subtype may have ameliorated the impact of differences in patients’ background between the two groups, our analyses did not directly address socioeconomic differences that could play roles in treatment outcomes. The incorporation of those additional factors is beyond the scope of this study and could be included in future studies.

Overall, these data show important differences in BC mutational spectrums, gene expression, and relevant transcriptional pathways between patients with genetically determined African and European ancestries, particularly within the HR+/HER2- BC and TNBC subtypes. To serve diverse populations of patients diagnosed with BC, promote equitable access to clinical trials, and accelerate the development of clinical decision tools for precision care, future studies should focus on geography and genetic ancestry when conducting biomarker-informed, early-phase clinical trials.

Availability of data and materials

All relevant data generated during this study are included in this published article and its supplementary information files. Raw data are not available due to proprietary restrictions.

Abbreviations

BC:

Breast cancer

TNBC:

Triple-negative breast cancer

AA:

African ancestry

EA:

European ancestry

LogFC:

Log fold-change

HER2:

Human epidermal growth factor receptor-2

HR:

Hormone receptor

AIM:

Ancestry-informative marker

EHR:

Electronic health record

NGS:

Next-generation sequencing

ssGSEA:

Single-sample gene set enrichment analysis

References

  1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2021. CA A Cancer J Clin. 2021. https://doi.org/10.3322/caac.21654.

    Article  Google Scholar 

  2. Schettini F, Pascual T, Conte B, Chic N, Brasó-Maristany F, Galván P, et al. HER2-enriched subtype and pathological complete response in HER2-positive breast cancer: a systematic review and meta-analysis. Cancer Treat Rev. 2020. https://doi.org/10.1016/j.ctrv.2020.101965.

    Article  PubMed  PubMed Central  Google Scholar 

  3. von Minckwitz G, von Minckwitz G, Procter M, de Azambuja E, Zardavas D, Benyunes M, et al. Adjuvant pertuzumab and trastuzumab in early HER2-positive breast cancer. N Engl J Med. 2017. https://doi.org/10.1056/nejmoa1703643.

    Article  Google Scholar 

  4. Finn RS, Martin M, Rugo HS, Jones S, Im S-A, Gelmon K, et al. Palbociclib and letrozole in advanced breast cancer. N Engl J Med. 2016;375:1925–36.

    Article  CAS  PubMed  Google Scholar 

  5. Goetz MP, Toi M, Campone M, Sohn J, Paluch-Shimon S, Huober J, et al. MONARCH 3: abemaciclib as initial therapy for advanced breast cancer. J Clin Oncol. 2017;35:3638–46.

    Article  CAS  PubMed  Google Scholar 

  6. Litton JK, Rugo HS, Ettl J, Hurvitz SA, Gonçalves A, Lee K-H, et al. Talazoparib in patients with advanced breast cancer and a germline BRCA mutation. N Engl J Med. 2018;379:753–63.

    Article  CAS  PubMed  Google Scholar 

  7. Robson M, Im S-A, Senkus E, Xu B, Domchek SM, Masuda N, et al. Olaparib for metastatic breast cancer in patients with a germline BRCA mutation. N Engl J Med. 2017;377:523–33.

    Article  CAS  PubMed  Google Scholar 

  8. André F, Ciruelos E, Rubovszky G, Campone M, Loibl S, Rugo HS, et al. Alpelisib for PIK3CA-mutated, hormone receptor-positive advanced breast cancer. N Engl J Med. 2019;380:1929–40.

    Article  PubMed  Google Scholar 

  9. Bianchini G, Balko JM, Mayer IA, Sanders ME, Gianni L. Triple-negative breast cancer: challenges and opportunities of a heterogeneous disease. Nat Rev Clin Oncol. 2016;13:674–90.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Yin L, Duan J-J, Bian X-W, Yu S-C. Triple-negative breast cancer molecular subtyping and treatment progress. Breast Cancer Res. 2020;22:61.

    Article  PubMed  PubMed Central  Google Scholar 

  11. DeSantis CE, Ma J, Gaudet MM, Newman LA, Miller KD, Goding Sauer A, et al. Breast cancer statistics, 2019. CA Cancer J Clin. 2019;69:438–51.

    Article  PubMed  Google Scholar 

  12. Prakash O, Hossain F, Danos D, Lassak A, Scribner R, Miele L. Racial disparities in triple negative breast cancer: a review of the role of biologic and non-biologic factors. Front Public Health. 2020;8: 576964.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Daly B, Olopade OI. A perfect storm: how tumor biology, genomics, and health care delivery patterns collide to create a racial survival disparity in breast cancer and proposed interventions for change. CA Cancer J Clin. 2015. https://doi.org/10.3322/caac.212716.

    Article  PubMed  Google Scholar 

  14. Huo D, Hu H, Rhie SK, Gamazon ER, Cherniack AD, Liu J, et al. Comparison of breast cancer molecular features and survival by African and European ancestry in the cancer genome atlas. JAMA Oncol. 2017;3:1654–62.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Goel N, Kim DY, Guo JA, Zhao D, Mahal BA, Alshalalfa M. Racial differences in genomic profiles of breast cancer. JAMA Netw Open. 2022;5: e220573.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Balogun OD, Olopade OI. Addressing health disparities in cancer with genomics. Nat Rev Genet. 2021;22:621–2.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Ibraheem A, Olopade OI, Huo D. Propensity score analysis of the prognostic value of genomic assays for breast cancer in diverse populations using the National Cancer Data Base. Cancer Wiley. 2020;126:4013–22.

    Article  CAS  Google Scholar 

  18. Yahya P, Sulong S, Harun A, Wangkumhang P, Wilantho A, Ngamphiw C, et al. Ancestry-informative marker (AIM) SNP panel for the Malay population. Int J Legal Med. 2020;134:123–34.

    Article  PubMed  Google Scholar 

  19. Fernandes LE, Epstein CG, Bobe AM, Bell JSK, Stumpe MC, Salazar ME, et al. Real-world evidence of diagnostic testing and treatment patterns in US patients with breast cancer with implications for treatment biomarkers from RNA sequencing data. Clin Breast Cancer. 2021. https://doi.org/10.1016/j.clbc.2020.11.012.

    Article  PubMed  Google Scholar 

  20. Alexander DH, Novembre J, Lange K. Fast model-based estimation of ancestry in unrelated individuals. Genome Res. 2009;19:1655–64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Consortium T 1000 GP, The 1000 genomes project consortium. An integrated map of genetic variation from 1,092 human genomes [Internet]. Nature. 2012. pp. 56–65. https://doi.org/10.1038/nature11632.

  22. Bryc K, Durand EY, Michael Macpherson J, Reich D, Mountain JL. The Genetic Ancestry of African Americans, Latinos, and European Americans across the United States. Am J Hum Genet. 2015. https://doi.org/10.1016/j.ajhg.2014.11.010.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Beaubier N, Bontrager M, Huether R, Igartua C, Lau D, Tell R, et al. Integrated genomic profiling expands clinical options for patients with cancer. Nat Biotechnol. 2019;37:1351–60.

    Article  CAS  PubMed  Google Scholar 

  24. Leibowitz BD, Dougherty BV, Bell JSK, Kapilivsky J, Michuda J, Sedgewick AJ, et al. Validation of genomic and transcriptomic models of homologous recombination deficiency in a real-world pan-cancer cohort. BMC Cancer. 2022;22:587.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40.

    Article  CAS  PubMed  Google Scholar 

  26. Yoon S, Baik B, Park T, Nam D. Powerful p-value combination methods to detect incomplete association. Sci Rep. 2021. https://doi.org/10.1038/s41598-021-86465-y.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102:15545–50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Liberzon A, Birger C, Thorvaldsdóttir H, Ghandi M, Mesirov JP, Tamayo P. The molecular signatures database hallmark gene set collection. Cell Syst. 2015. https://doi.org/10.1016/j.cels.2015.12.004.

    Article  PubMed  PubMed Central  Google Scholar 

  29. GSEA [Internet]. [cited 2022 Jun 14]. Available from: https://www.gsea-msigdb.org/gsea/msigdb/collections.jsp.

  30. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43: e47.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Papadakis ES, Reeves T, Robson NH, Maishman T, Packham G, Cutress RI. BAG-1 as a biomarker in early breast cancer prognosis: a systematic review with meta-analyses. Br J Cancer. 2017. https://doi.org/10.1038/bjc.2017.130.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Eom YH, Kim HS, Lee A, Song BJ, Chae BJ. BCL2 as a subtype-specific prognostic marker for breast cancer. J Breast Cancer. 2016. https://doi.org/10.4048/jbc.2016.19.3.252.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Xu L, Yu W, Xiao H, Lin K. BIRC5 is a prognostic biomarker associated with tumor immune cell infiltration. Sci Rep. 2021. https://doi.org/10.1038/s41598-020-79736-7.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Ding K, Li W, Zou Z, Zou X, Wang C. CCNB1 is a prognostic biomarker for ER breast cancer. Med Hypotheses. 2014. https://doi.org/10.1016/j.mehy.2014.06.013.

    Article  PubMed  Google Scholar 

  35. Dastsooz H, Cereda M, Donna D, Oliviero S. A comprehensive bioinformatics analysis of UBE2C in cancers. Int J Mol Sci. 2019. https://doi.org/10.3390/ijms20092228.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Sun X, Clermont P-L, Jiao W, Helgason CD, Gout PW, Wang Y, et al. Elevated expression of the centromere protein-A(CENP-A)-encoding gene as a prognostic and predictive biomarker in human cancers. Int J Cancer. 2016. https://doi.org/10.1002/ijc.30133.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Putluri N, Maity S, Kommagani R, Creighton CJ, Putluri V, Chen F, et al. Pathway-centric integrative analysis identifies RRM2 as a prognostic marker in breast cancer associated with poor survival and tamoxifen resistance. Neoplasia. 2014. https://doi.org/10.1016/j.neo.2014.05.007.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Cheng C-J, Lin Y-C, Tsai M-T, Chen C-S, Hsieh M-C, Chen C-L, et al. SCUBE2 suppresses breast tumor cell proliferation and confers a favorable prognosis in invasive breast cancer. Cancer Res. 2009. https://doi.org/10.1158/0008-5472.can-08-3615.

    Article  PubMed  PubMed Central  Google Scholar 

  39. McCallister C, Siracusa MC, Shirazi F, Chalkia D, Nikolaidis N. Functional diversification and specialization of cytosolic 70-kDa heat shock proteins. Sci Rep. 2015. https://doi.org/10.1038/srep09363.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Keersmaecker KD, De Keersmaecker K, Atak ZK, Li N, Vicente C, Patchett S, et al. Exome sequencing identifies mutation in CNOT3 and ribosomal genes RPL5 and RPL10 in T-cell acute lymphoblastic leukemia. Nat Genet. 2013. https://doi.org/10.1038/ng.2508.

    Article  PubMed  Google Scholar 

  41. Creighton CJ, Hilger AM, Murthy S, Rae JM, Chinnaiyan AM, El-Ashry D. Activation of mitogen-activated protein kinase in estrogen receptor α–positive breast cancer cells in vitro induces an in vivo molecular phenotype of estrogen receptor α–negative human breast tumors. Cancer Res. 2006. https://doi.org/10.1158/0008-5472.can-05-4363.

    Article  PubMed  Google Scholar 

  42. Ademuyiwa FO, Tao Y, Luo J, Weilbaecher K, Ma CX. Differences in the mutational landscape of triple-negative breast cancer in African Americans and Caucasians. Breast Cancer Res Treat. 2017;161:491–9.

    Article  CAS  PubMed  Google Scholar 

  43. Keenan T, Moy B, Mroz EA, Ross K, Niemierko A, Rocco JW, et al. Comparison of the genomic landscape between primary breast cancer in African American versus white women and the association of racial differences with tumor recurrence. J Clin Oncol. 2015. https://doi.org/10.1200/jco.2015.62.2126.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Nixon MJ, Formisano L, Mayer IA, Valeria-Estrada M, González-Ericsson PI, Isakoff SJ, et al. PIK3CA and MAP3K1 alterations imply luminal A status and are associated with clinical benefit from pan-PI3K inhibitor buparlisib and letrozole in ER metastatic breast cancer. npj Breast Cancer. 2019. https://doi.org/10.1038/s41523-019-0126-6.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Zheng Y, Walsh T, Gulsuner S, Casadei S, Lee MK, Ogundiran TO, et al. Inherited breast cancer in Nigerian women. J Clin Oncol. 2018;36:2820–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Adedokun B, Zheng Y, Ndom P, Gakwaya A, Makumbi T, Zhou AY, et al. Prevalence of inherited mutations in breast cancer predisposition genes among women in Uganda and Cameroon. Cancer Epidemiol Biomark Prev. 2020;29:359–67.

    Article  Google Scholar 

  47. Hu C, Hart SN, Gnanaolivu R, Huang H, Lee KY, Na J, et al. A population-based study of genes previously implicated in breast cancer. N Engl J Med. 2021;384:440–51.

    Article  PubMed  PubMed Central  Google Scholar 

  48. Alexandrov L. Signatures of Mutational Processes in Human Cancer. University of Cambridge; 2014.

  49. Rao RC, Dou Y. Hijacked in cancer: the KMT2 (MLL) family of methyltransferases. Nat Rev Cancer. 2015. https://doi.org/10.1038/nrc3929.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Liu X, Qiu R, Xu M, Meng M, Zhao S, Ji J, et al. KMT2C is a potential biomarker of prognosis and chemotherapy sensitivity in breast cancer. Breast Cancer Res Treat. 2021. https://doi.org/10.1007/s10549-021-06325-1.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Girardi T, Vereecke S, Sulima SO, Khan Y, Fancello L, Briggs JW, et al. The T-cell leukemia-associated ribosomal RPL10 R98S mutation enhances JAK-STAT signaling. Leukemia. 2018;32:809–19.

    Article  CAS  PubMed  Google Scholar 

  52. Kampen KR, Sulima SO, Verbelen B, Girardi T, Vereecke S, Rinaldi G, et al. The ribosomal RPL10 R98S mutation drives IRES-dependent BCL-2 translation in T-ALL. Leukemia. 2019. https://doi.org/10.1038/s41375-018-0176-z.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Vostakolaei MA, Hatami-Baroogh L, Babaei G, Molavi O, Kordi S, Abdolalizadeh J. Hsp70 in cancer: a double agent in the battle between survival and death. J Cell Physiol. 2021. https://doi.org/10.1002/jcp.30132.

    Article  PubMed  Google Scholar 

  54. Ren W, Medeiros N, Warneford-Thomson R, Wulfridge P, Yan Q, Bian J, et al. Disruption of ATRX-RNA interactions uncovers roles in ATRX localization and PRC2 function. Nat Commun. 2020. https://doi.org/10.1038/s41467-020-15902-9.

    Article  PubMed  PubMed Central  Google Scholar 

  55. Zhao F, Miyashita M, Hattori M, et al. Racial disparities in pathological complete response among patients receiving neoadjuvant chemotherapy for early stage breast cancer. JAMA Network Open. 2023 (In Press).

Download references

Acknowledgements

We would like to thank Alexandria M. Bobe for administrative support. We would also like to thank Elisabeth Sveen along with all laboratory and computational science personnel at the University of Chicago and Tempus Labs for their contributions to the data acquisition.

Funding

This work was partially supported by Susan G. Komen for the CURE SAC110026 and SAC210203. Additionally, O.I.O. is an American Cancer Society Professor.

Author information

Author notes

  1. Kevin P. White

    Present address: National University Singapore, Queenstown, Singapore

  2. Co-first authors: Minoru Miyashita and Joshua S. K. Bell

Authors and Affiliations

  1. The University of Chicago, Chicago, IL, USA

    Minoru Miyashita, Yonglan Zheng, Toshio F. Yoshimatsu, Galina Khramtsova, Fang Liu, Fangyuan Zhao, Frederick M. Howard, Rita Nanda, Dezheng Huo & Olufunmilayo I. Olopade

  2. Tempus Inc, Chicago, IL, USA

    Joshua S. K. Bell, Stephane Wenric, Ezgi Karaesmen, Brooke Rhead, Matthew Kase, Kristiyana Kaneva, Francisco M. De La Vega, Nike Beaubier & Kevin P. White

Authors
  1. Minoru Miyashita
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Joshua S. K. Bell
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Stephane Wenric
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ezgi Karaesmen
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Brooke Rhead
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Matthew Kase
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Kristiyana Kaneva
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Francisco M. De La Vega
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Yonglan Zheng
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Toshio F. Yoshimatsu
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Galina Khramtsova
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Fang Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  13. Fangyuan Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

  14. Frederick M. Howard
    View author publications

    You can also search for this author in PubMed Google Scholar

  15. Rita Nanda
    View author publications

    You can also search for this author in PubMed Google Scholar

  16. Nike Beaubier
    View author publications

    You can also search for this author in PubMed Google Scholar

  17. Kevin P. White
    View author publications

    You can also search for this author in PubMed Google Scholar

  18. Dezheng Huo
    View author publications

    You can also search for this author in PubMed Google Scholar

  19. Olufunmilayo I. Olopade
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

MM, JSKB, EK, SW, BR, KK, FDLV, KPW, NB, and OIO contributed to conception/design. JSKB, EK, SW, BR, FDLV, and GK analyzed the data. MM, SW, BR, MK, YZ, FDLV, and OIO interpreted the data. MK, KK, MM, JSKB, EK, SW, and BR drafted the work. MM, JSKB, MK, KK, FDLV, FMH, YZ, TFY, DH, and OIO revised the study critically. All authors agree to be accountable for all aspects of the work.

Corresponding author

Correspondence to Olufunmilayo I. Olopade.

Ethics declarations

Ethics approval and consent to participate

All analyses were performed using de-identified data; IRB exemption Pro00042950 was obtained from Advarra on April 15, 2020.

Consent for publication

Not applicable.

Competing interests

E.K., S.W., J.S.K.B., B.R., M.K., K.K., N.B., and F.D.L.V. are employees of Tempus Labs. R.N. serves on the Advisory Board for Astrazeneca, BeyondSpring, Cardinal Health, Fujifilm, Immunomedics/Gilead Infinity, iTeos, MacroGenics, Merck, OBI, Oncosec, and Seattle Genetics, and R.N.’s associated institution has received research funding from Arvinas, AstraZeneca, Celgene, Corcept Therapeutics, Genentech/Roche, Immunomedics, Merck, OBI Pharma, Odonate Therapeutics, OncoSec, Pfizer, Seattle Genetics, and Taiho. K.P.W. is a Scientific Advisor and stockholder of Tempus Labs. K.P.W. is chairman and stockholder of Provaxus, Inc. O.I.O. is co-founder at CancerIQ, serves as Scientific Advisor at Tempus, is on the Board of 54gene and received research support from Genentech/Roche and Color Genomics. All other authors declare no competing interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Table S1:

The predefined set of ancestry-informative markers (AIMs) used to estimate ancestry proportion likelihoods for five major continental regions: Africa, Americas, East Asia, Europe, and South Asia, as defined in the 1000 Genomes Project. The chromosome, position, reference, and alternate allele columns are shown for each of the 654 AIMs.

Additional file 2: Fig. S1.

Concordance of ancestry proportion likelihood estimates between tumor and normal tissue samples. Overall, samples were highly concordant for estimates of Africa, Americas, East Asia, European, and South Asiaancestries.

Additional file 3: Table S2:

The subset of patient records with available metadata of reported race, and the number of patient records within each reported race that were inferred as European or African_European_admixed ancestry from next-generation sequencing data.

Additional file 4: Table S3:

Results from the analysis of cancer somatic variants classified as pathogenic or likely pathogenic in each ancestral group from tumor DNA. Contains the frequency of carriers for alterations in each of the genes compared, the P value from a direct comparison between the two ancestry groups, and a P value adjusted for multiple hypothesis testing, as well as the most frequent alterations for each ancestry group.

Additional file 5: Table S4:

All genes identified in differential expression analyses by RNA sequencing between African ancestry (AA) and European ancestry (EA) patients. The P value and log fold-change (logFC) with EA expression as the reference point are provided for each differentially expressed gene.

Additional file 6: Table S5:

Gene sets with significantly different enrichment between AA and EA patients across the entire cohort.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Miyashita, M., Bell, J.S.K., Wenric, S. et al. Molecular profiling of a real-world breast cancer cohort with genetically inferred ancestries reveals actionable tumor biology differences between European ancestry and African ancestry patient populations. Breast Cancer Res 25, 58 (2023). https://doi.org/10.1186/s13058-023-01627-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13058-023-01627-2

Keywords

Breast Cancer Research

ISSN: 1465-542X

Contact us