Hidden secrets of the cancer genome: unlocking the impact of non-coding mutations in gene regulatory elements.

Discoveries in the field of genomics have revealed that non-coding genomic regions are not merely "junk DNA", but rather comprise critical elements involved in gene expression. These gene regulatory elements (GREs) include enhancers, insulators, silencers, and gene promoters. Notably, new evidence shows how mutations within these regions substantially influence gene expression programs, especially in the context of cancer. Advances in high-throughput sequencing technologies have accelerated the identification of somatic and germline single nucleotide mutations in non-coding genomic regions. This review provides an overview of somatic and germline non-coding single nucleotide alterations affecting transcription factor binding sites in GREs, specifically involved in cancer biology. It also summarizes the technologies available for exploring GREs and the challenges associated with studying and characterizing non-coding single nucleotide mutations. Understanding the role of GRE alterations in cancer is essential for improving diagnostic and prognostic capabilities in the precision medicine era, leading to enhanced patient-centered clinical outcomes.

Chromatin insulation orchestrates matrix metalloproteinase gene cluster expression reprogramming in aggressive breast cancer tumors.

BACKGROUND: Triple-negative breast cancer (TNBC) is an aggressive subtype that exhibits a high incidence of distant metastases and lacks targeted therapeutic options. Here we explored how the epigenome contributes to matrix metalloprotease (MMP) dysregulation impacting tumor invasion, which is the first step of the metastatic process. METHODS: We combined RNA expression and chromatin interaction data to identify insulator elements potentially associated with MMP gene expression and invasion. We employed CRISPR/Cas9 to disrupt the CCCTC-Binding Factor (CTCF) binding site on an insulator element downstream of the MMP8 gene (IE8) in two TNBC cellular models. We characterized these models by combining Hi-C, ATAC-seq, and RNA-seq with functional experiments to determine invasive ability. The potential of our findings to predict the progression of ductal carcinoma in situ (DCIS), was tested in data from clinical specimens. RESULTS: We explored the clinical relevance of an insulator element located within the Chr11q22.2 locus, downstream of the MMP8 gene (IE8). This regulatory element resulted in a topologically associating domain (TAD) boundary that isolated nine MMP genes into two anti-correlated expression clusters. This expression pattern was associated with worse relapse-free (HR = 1.57 [1.06 - 2.33]; p = 0.023) and overall (HR = 2.65 [1.31 - 5.37], p = 0.005) survival of TNBC patients. After CRISPR/Cas9-mediated disruption of IE8, cancer cells showed a switch in the MMP expression signature, specifically downregulating the pro-invasive MMP1 gene and upregulating the antitumorigenic MMP8 gene, resulting in reduced invasive ability and collagen degradation. We observed that the MMP expression pattern predicts DCIS that eventually progresses into invasive ductal carcinomas (AUC = 0.77, p < 0.01). CONCLUSION: Our study demonstrates how the activation of an IE near the MMP8 gene determines the regional transcriptional regulation of MMP genes with opposing functional activity, ultimately influencing the invasive properties of aggressive forms of breast cancer.

Machine Learning-Based Epigenetic Classifiers for Axillary Staging of Patients with ER-Positive Early-Stage Breast Cancer.

BACKGROUND: In the era of molecular stratification and effective multimodality therapies, surgical staging of the axilla is becoming less relevant for patients with estrogen receptor (ER)-positive early-stage breast cancer (EBC). Therefore, a nonsurgical method for accurately predicting lymph node disease is the next step in the de-escalation of axillary surgery. This study sought to identify epigenetic signatures in the primary tumor that accurately predict lymph node status. PATIENTS AND METHODS: We selected a cohort of patients in The Cancer Genome Atlas (TCGA) with ER-positive, HER2-negative invasive ductal carcinomas, and clinically-negative axillae (n = 127). Clinicopathological nomograms from the Memorial Sloan Kettering Cancer Center (MSKCC) and the MD Anderson Cancer Center (MDACC) were calculated. DNA methylation (DNAm) patterns from primary tumor specimens were compared between patients with pN0 and those with > pN0. The cohort was divided into training (n = 85) and validation (n = 42) sets. Random forest was employed to obtain the combinations of DNAm features with the highest accuracy for stratifying patients with > pN0. The most efficient combinations were selected according to the area under the curve (AUC). RESULTS: Clinicopathological models displayed a modest predictive potential for identifying > pN0 disease (MSKCC AUC 0.76, MDACC AUC 0.69, p = 0.15). Differentially methylated sites (DMS) between patients with pN0 and those with > pN0 were identified (n = 1656). DMS showed a similar performance to the MSKCC model (AUC = 0.76, p = 0.83). Machine learning approaches generated five epigenetic classifiers, which showed higher discriminative potential than the clinicopathological variables tested (AUC > 0.88, p < 0.05). CONCLUSIONS: Epigenetic classifiers based on primary tumor characteristics can efficiently stratify patients with no lymph node involvement from those with axillary lymph node disease, thereby providing an accurate method of staging the axilla.

Epigenetic Signatures Predict Pathologic Nodal Stage in Breast Cancer Patients with Estrogen Receptor-Positive, Clinically Node-Positive Disease.

BACKGROUND: Breast cancer patients with clinically positive nodes who undergo upfront surgery are often recommended for axillary lymph node dissection (ALND), yet more than half are found to have limited nodal disease (≤ 3 positive nodes, pN1) at surgery. In this study, we examined the efficiency of molecular classifiers in stratifying patients with clinically positive nodes to pN1 versus > pN1 disease. METHODS: We evaluated the clinical and epigenetic data of patients in The Cancer Genome Atlas with estrogen receptor-positive, human epidermal growth factor receptor 2-negative invasive ductal carcinoma who underwent ALND for node-positive disease. Patients were divided into control (pN1, ≤ 3 positive nodes) and case (> pN1, > 3 positive nodes) groups. Machine learning algorithms were trained on 50% of the cohort and validated on the remaining 50% to identify DNA methylation signatures that predict > pN1 disease. Clinical variables and epigenetic signatures were compared. RESULTS: Controls (n = 34) and case (n = 24) cohorts showed similar mean age (56.4 ± 12.2 vs. 57.6 ± 16.7 years; p = 0.77), number of nodes removed (16.1 ± 7.3 vs. 17.5 ± 6.2; p = 0.45), tumor grade (p = 0.76), presence of lymphovascular invasion (p = 0.18), extranodal extension (p = 0.17), tumor laterality (p = 0.89), and tumor location (p = 0.42). The mean number of positive nodes was significantly different (1.76 ± 0.82, pN1; 8.83 ± 5.36, > pN1; p < 0.001). Three epigenetic signatures (EpiSig14, EpiSig13, EpiSig10) based on DNA methylation patterns of the primary tumors demonstrated high accuracy in predicting > pN1 disease (area under the curve 0.98). CONCLUSIONS: Epigenetic signatures have an excellent diagnostic accuracy for stratifying nodal disease in patients with clinically positive nodes. Validation of this tool is warranted and may provide an accurate and cost-effective method of identifying patients with predicted low nodal burden who could be spared the morbidity of ALND.

Establishing Novel Molecular Subtypes of Appendiceal Cancer.

PURPOSE: Appendiceal cancer is a rare disease process with complex treatment strategies. The objective of this study was to identify mutation-based genetic subtypes that may differ from the current histological classification, compare the genetic make-up of primaries and metastases, and find novel targetable alterations. METHODS: The analyses involved the curation and normalization of gene mutation panels from appendiceal adenocarcinoma and mucinous adenocarcinoma (n = 196) stored in the AACR GENIE Database v6.0. Genes mutated in less than one patient and tumors profiled with incomplete mutation panels were excluded from the study. The optimal number of AC subtypes was established using the Nonnegative Matrix Factorization algorithm. Statistical comparisons of mutation frequencies were performed using Pearson's χ2 test. RESULTS: AC patients were stratified into five mutation subtypes, based on a final set of 41 cancer-related genes. AC0 had no mutations. The most frequently mutated genes varied between the subtypes were: AC1: KRAS (91.9%) and GNAS (77.4%); AC2: KRAS (52.5%), APC (32.5%), and GNAS (30%); AC3: KMT2D (38.7%), TP53 (38.7%), KRAS (35.5%), EP300 (22.6%); and AC4: TP53 (97.2%), KRAS (77.8%), and SMAD4 (36.1%). Additionally, AC3 was less likely to be mucinous (22.6% vs. 50.0-74.2%, p < 0.001) and had a higher mutation frequency (3.6 vs. 0-3.1, p < 0.001). There were no significant differences between primary tumors and metastases in the 41 assessed genes (p = 0.35). CONCLUSIONS: The characterization of these subtypes suggests a need for molecular approaches to complement anatomical and histopathological staging for AC. A prospective comparison of subtype prognosis and response to surgery and adjuvant treatment is needed to identify the clinical applications of the novel molecular subtypes.

Is the 21-Gene Recurrence Score on Core Needle Biopsy Equivalent to Surgical Specimen in Early-Stage Breast Cancer? A Comparison of Gene Expression Between Paired Core Needle Biopsy and Surgical Specimens.

BACKGROUND: Molecular testing on surgical specimens predicts disease recurrence and benefit of adjuvant chemotherapy in hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) early-stage breast cancer (EBC). Testing on core biopsies has become common practice despite limited evidence of concordance between core/surgical samples. In this study, we compared the gene expression of the 21 genes and the recurrence score (RS) between paired core/surgical specimens. METHODS: Eighty patients with HR+/HER2- EBC were evaluated from two publicly available gene expression datasets (GSE73235, GSE76728) with paired core/surgical specimens without neoadjuvant systemic therapy. The expression of the 21 genes was compared in paired samples. A microarray-based RS was calculated and a value ≥ 26 was defined as high-RS. The concordance rate and kappa statistic were used to evaluate the agreement between the RS of paired samples. RESULTS: Overall, there was no significant difference and a high correlation in the gene expression levels of the 21 genes between paired samples. However, CD68 and RPLP0 in GSE73235, AURKA, BAG1, and TFRC in GSE76728, and MYLBL2 and ACTB in both datasets exhibited weak to moderate correlation (r < 0.5). There was a high correlation of the microarray-based RS between paired samples in GSE76728 (r = 0.91, 95% confidence interval [CI] 0.81-0.96) and GSE73235 (r = 0.82, 95% CI 0.71-0.89). There were no changes in RS category in GSE76728, whereas 82% of patients remained in the same RS category in GSE73235 (κ = 0.64). CONCLUSIONS: Gene expression levels of the 21-gene RS showed a high correlation between paired specimens. Potential sampling and biological variability on a set of genes need to be considered to better estimate the RS from core needle biopsy.

Mitochondrial stress triggers a pro-survival response through epigenetic modifications of nuclear DNA.

Mitochondrial dysfunction represents an important cellular stressor and when intense and persistent cells must unleash an adaptive response to prevent their extinction. Furthermore, mitochondria can induce nuclear transcriptional changes and DNA methylation can modulate cellular responses to stress. We hypothesized that mitochondrial dysfunction could trigger an epigenetically mediated adaptive response through a distinct DNA methylation patterning. We studied cellular stress responses (i.e., apoptosis and autophagy) in mitochondrial dysfunction models. In addition, we explored nuclear DNA methylation in response to this stressor and its relevance in cell survival. Experiments in cultured human myoblasts revealed that intense mitochondrial dysfunction triggered a methylation-dependent pro-survival response. Assays done on mitochondrial disease patient tissues showed increased autophagy and enhanced DNA methylation of tumor suppressor genes and pathways involved in cell survival regulation. In conclusion, mitochondrial dysfunction leads to a "pro-survival" adaptive state that seems to be triggered by the differential methylation of nuclear genes.

The RhoJ-BAD signaling network: An Achilles' heel for BRAF mutant melanomas.

Genes and pathways that allow cells to cope with oncogene-induced stress represent selective cancer therapeutic targets that remain largely undiscovered. In this study, we identify a RhoJ signaling pathway that is a selective therapeutic target for BRAF mutant cells. RhoJ deletion in BRAF mutant melanocytes modulates the expression of the pro-apoptotic protein BAD as well as genes involved in cellular metabolism, impairing nevus formation, cellular transformation, and metastasis. Short-term treatment of nascent melanoma tumors with PAK inhibitors that block RhoJ signaling halts the growth of BRAF mutant melanoma tumors in vivo and induces apoptosis in melanoma cells in vitro via a BAD-dependent mechanism. As up to 50% of BRAF mutant human melanomas express high levels of RhoJ, these studies nominate the RhoJ-BAD signaling network as a therapeutic vulnerability for fledgling BRAF mutant human tumors.

B7H3 regulates differentiation and serves as a potential biomarker and theranostic target for human glioblastoma.

B7H3 (CD276), a co-stimulator molecule of the cell surface B7 protein superfamily, is expressed on glioblastomas (GBM). Recently, B7H3 functions beyond immune costimulation have been demonstrated. However, the mechanisms underlying B7H3 function are diverse and not well understood. GBM tumors contain undifferentiated self-renewing cells, which confound therapeutic attempts. We investigated the role of B7H3 in the regulation of GBM cell differentiation and the regulatory pathways involved. Analysis of public databases (TCGA, Rembrandt, and GEO NCBI) and RNA sequencing were performed to explore the role of B7H3 in GBM. Knockdown and overexpression of B7H3, were used to verify the downstream pathway in vitro. Further studies in vivo were performed to support the new finding. Bioinformatics analysis identified a correlation between the expression of B7H3, the expression of glioma self-renewing cell (GSC)-related genes, and MYC expression. These observations were verified by RNA-sequencing analyses in primary GBM cell lines. In vitro knockdown of B7H3-induced differentiation, associated with downregulation of SMAD6 (a TGF-ß pathway inhibitor) and enhancement of SMAD1 phosphorylation-induced SMAD4 expression. Importantly, activation of the TGF-ß pathway resulted in downregulation of MYC expression. In vivo assays conducted in a human GBM cell line xenograft mouse model demonstrated that B7H3 knockdown decreased MYC expression and inhibited tumor growth. B7H3 knockdown could regulate GBM differentiation by modulating MYC expression. So, B7H3 could serve as a potential theranostic target for the treatment of patients with GBM.

The metastatic microenvironment: Melanoma-microglia cross-talk promotes the malignant phenotype of melanoma cells.

Melanoma has the highest propensity to metastasize to the brain compared to other cancers, as brain metastases are found frequently high in patients who have prolonged survival with visceral metastasis. Once disseminated in the brain, melanoma cells communicate with brain resident cells that include astrocytes and microglia. Microglia cells are the resident macrophages of the brain and are the main immunological cells in the CNS involved in neuroinflammation. Data on the interactions between brain metastatic melanoma cells and microglia and on the role of microglia-mediated neuroinflammation in facilitating melanoma brain metastasis are lacking. To elucidate the role of microglia in melanoma brain metastasis progression, we examined the bidirectional interactions between microglia and melanoma cells in the tumor microenvironment. We identified the molecular and functional modifications occurring in brain-metastasizing melanoma cells and microglia cells after the treatment of each cell type with supernatants of the counter cell type. Both cells induced alteration in gene expression programs, cell signaling, and cytokine secretion in the counter cell type. Moreover, melanoma cells exerted significant morphological changes on microglia cells, enhanced proliferation, induced matrix metalloproteinase-2 (MMP-2) activation, and cell migration. Microglia cells induced phenotypic changes in melanoma cells increasing their malignant phenotype: increased melanoma proliferation, MMP-2 activity, cell migration, brain endothelial penetration, and tumor cells ability to grow as spheroids in 3D cultures. Our work provides a novel insight into the bidirectional interactions between melanoma and micoglia cells, suggesting the contribution of microglia to melanoma brain metastasis formation.

Clinical Implications of Transcriptomic Changes After Neoadjuvant Chemotherapy in Patients with Triple-Negative Breast Cancer.

BACKGROUND: Pathological response to neoadjuvant chemotherapy (NAC) is critical in prognosis and selection of systemic treatments for patients with triple-negative breast cancer (TNBC). The aim of this study is to identify gene expression-based markers to predict response to NAC. PATIENTS AND METHODS: A survey of 43 publicly available gene expression datasets was performed. We identified a cohort of TNBC patients treated with NAC (n = 708). Gene expression data from different studies were renormalized, and the differences between pretreatment (pre-NAC), on-treatment (post-C1), and surgical (Sx) specimens were evaluated. Euclidean statistical distances were calculated to estimate changes in gene expression patterns induced by NAC. Hierarchical clustering and pathway enrichment analyses were used to characterize relationships between differentially expressed genes and affected gene pathways. Machine learning was employed to refine a gene expression signature with the potential to predict response to NAC. RESULTS: Forty nine genes consistently affected by NAC were involved in enhanced regulation of wound response, chemokine release, cell division, and decreased programmed cell death in residual invasive disease. The statistical distances between pre-NAC and post-C1 significantly predicted pathological complete response [area under the curve (AUC) = 0.75; p = 0.003; 95% confidence interval (CI) 0.58-0.92]. Finally, the expression of CCND1, a cyclin that forms complexes with CDK4/6 to promote the cell cycle, was the most informative feature in pre-NAC biopsies to predict response to NAC. CONCLUSIONS: The results of this study reveal significant transcriptomic changes induced by NAC and suggest that chemotherapy-induced gene expression changes observed early in therapy may be good predictors of response to NAC.

Clinicopathological Features of Triple-Negative Breast Cancer Epigenetic Subtypes.

BACKGROUND/OBJECTIVE: Triple-negative breast cancer (TNBC) is a heterogeneous collection of breast tumors with numerous differences including morphological characteristics, genetic makeup, immune-cell infiltration, and response to systemic therapy. DNA methylation profiling is a robust tool to accurately identify disease-specific subtypes. We aimed to generate an epigenetic subclassification of TNBC tumors (epitypes) with utility for clinical decision-making. METHODS: Genome-wide DNA methylation profiles from TNBC patients generated in the Cancer Genome Atlas project were used to build machine learning-based epigenetic classifiers. Clinical and demographic variables, as well as gene expression and gene mutation data from the same cohort, were integrated to further refine the TNBC epitypes. RESULTS: This analysis indicated the existence of four TNBC epitypes, named as Epi-CL-A, Epi-CL-B, Epi-CL-C, and Epi-CL-D. Patients with Epi-CL-B tumors showed significantly shorter disease-free survival and overall survival [log rank; P = 0.01; hazard ratio (HR) 3.89, 95% confidence interval (CI) 1.3-11.63 and P = 0.003; HR 5.29, 95% CI 1.55-18.18, respectively]. Significant gene expression and mutation differences among the TNBC epitypes suggested alternative pathway activation that could be used as ancillary therapeutic targets. These epigenetic subtypes showed complementarity with the recently described TNBC transcriptomic subtypes. CONCLUSIONS: TNBC epigenetic subtypes exhibit significant clinical and molecular differences. The links between genetic make-up, gene expression programs, and epigenetic subtypes open new avenues in the development of laboratory tests to more efficiently stratify TNBC patients, helping optimize tailored treatment approaches.

Epigenomic and Transcriptomic Characterization of Secondary Breast Cancers.

BACKGROUND: Molecular alterations impact tumor prognosis and response to treatment. This study was designed to identify transcriptomic and epigenomic signatures of breast cancer (BC) tumors from patients with any prior malignancy. METHODS: RNA-sequencing and genome-wide DNA methylation profiles from BCs were generated in the Cancer Genome Atlas project. Patients with secondary breast cancer (SBC) were separated by histological subtype and matched to primary breast cancer controls to create two independent cohorts of invasive ductal (IDC, n = 36) and invasive lobular (ILC, n = 40) carcinoma. Differentially expressed genes, as well as differentially methylated genomic regions, were integrated to identify epigenetically regulated abnormal gene pathways in SBCs. RESULTS: Differentially expressed genes were identified in IDC SBCs (n = 727) and in ILC SBCs (n = 261; Wilcoxon's test; P < 0.05). In IDC SBCs, 105 genes were upregulated and hypomethylated, including an estrogen receptor gene, and 73 genes were downregulated and hypermethylated, including genes involved in antigen presentation and interferon response pathways (HLA-E, IRF8, and RELA). In ILC SBCs, however, only 17 genes were synchronously hypomethylated and upregulated, whereas 46 genes hypermethylated and downregulated. Interestingly, the SBC gene expression signatures closely corresponded with each histological subtype with only 1.51% of genes overlapping between the two histological subtypes. CONCLUSIONS: Differential gene expression and DNA methylation signatures are seen in both IDC and ILC SBCs, including genes that are relevant to tumor growth and proliferation. Differences in gene expression signatures corresponding with each histological subtype emphasize the importance of disease subtype-specific evaluations of molecular alterations.

Epigenome-wide DNA methylation landscape of melanoma progression to brain metastasis reveals aberrations on homeobox D cluster associated with prognosis.

Melanoma brain metastasis (MBM) represents a frequent complication of cutaneous melanoma. Despite aggressive multi-modality therapy, patients with MBM often have a survival rate of <1 year. Alteration in DNA methylation is a major hallmark of tumor progression and metastasis; however, it remains largely unexplored in MBM. In this study, we generated a comprehensive DNA methylation landscape through the use of genome-wide copy number, DNA methylation and gene expression data integrative analysis of melanoma progression to MBM. A progressive genome-wide demethylation in low CpG density and an increase in methylation level of CpG islands according to melanoma progression were observed. MBM-specific partially methylated domains (PMDs) affecting key brain developmental processes were identified. Differentially methylated CpG sites between MBM and lymph node metastasis (LNM) from patients with good prognosis were identified. Among the most significantly affected genes were the HOX family members. DNA methylation of HOXD9 gene promoter affected transcript and protein expression and was significantly higher in MBM than that in early stages. A MBM-specific PMD was identified in this region. Low methylation level of this region was associated with active HOXD9 expression, open chromatin and histone modifications associated with active transcription. Demethylating agent induced HOXD9 expression in melanoma cell lines. The clinical relevance of this finding was verified in an independent large cohort of melanomas (n = 145). Patients with HOXD9 hypermethylation in LNM had poorer disease-free and overall survival. This epigenome-wide study identified novel methylated genes with functional and clinical implications for MBM patients.

Epigenetic Profiles of Triple-Negative Breast Cancers of African American and White Females.

Importance: Triple-negative breast cancer (TNBC) is the most aggressive breast cancer subtype and appears to have disproportionately higher incidence and worse outcomes among younger African American females. Objective: To investigate whether epigenetic differences exist in TNBCs of younger African American females that may explain clinical disparities seen in this patient group. Design, Setting, and Participants: This cross-sectional study used clinical, demographic, DNA methylation (HumanMethylation450; Illumina), and gene expression (RNA sequencing) data for US patient populations from publicly available data repositories (The Cancer Genome Atlas [TCGA], 2006-2012, and Gene Expression Omnibus [GEO], 2004-2013) accessed on April 13, 2021. White and African American females with TNBC identified in TCGA (69 patients) and a validation cohort of 210 African American patients from GEO (GSE142102) were included. Patients without available race or age data were excluded. Data were analyzed from September 2022 through April 2023. Main Outcomes and Measures: DNA methylation and gene expression profiles of TNBC tumors by race (self-reported) and age were assessed. Age was considered a dichotomous variable using age 50 years as the cutoff (younger [<50 years] vs older [≥50 years]). Results: A total of 69 female patients (34 African American [49.3%] and 35 White [50.7%]; mean [SD; range] age, 55.7 [11.6; 29-82] years) with TNBC were included in the DNA methylation analysis; these patients and 210 patients in the validation cohort were included in the gene expression analysis (279 patients). There were 1115 differentially methylated sites among younger African American females. The DNA methylation landscape on TNBC tumors in this population had increased odds of enrichment of hormone (odds ratio [OR], 1.82; 95% CI, 1.21 to 2.67; P = .003), muscle (OR, 1.85; 95% CI, 1.44 to 2.36; P < .001), and proliferation (OR, 3.14; 95% CI, 2.71 to 3.64; P < .001) pathways vs other groups (older African American females and all White females). Alterations in regulators of these molecular features in TNBCs of younger African American females were identified involving hormone modulation (downregulation of androgen receptor: fold change [FC] = -2.93; 95% CI, -4.76 to -2.11; P < .001) and upregulation of estrogen-related receptor α (FC = 0.86; 95% CI, 0.34 to 1.38; P = .002), muscle metabolism (upregulation of FOXC1: FC = 1.33; 95% CI, 0.62 to 2.03; P < .001), and proliferation mediators (upregulation of NOTCH1: FC = 0.71; 95% CI, 0.23 to 1.19; P = .004 and MYC (FC = 0.81; 95% CI, 0.18 to 1.45; P = .01). Conclusions and Relevance: These findings suggest that TNBC of younger African American females may represent a distinct epigenetic entity and offer novel insight into molecular alterations associated with TNBCs of this population. Understanding these epigenetic differences may lead to the development of more effective therapies for younger African American females, who have the highest incidence and worst outcomes from TNBC of any patient group.

Construction and validation of a gene expression classifier to predict immunotherapy response in primary triple-negative breast cancer.

BACKGROUND: Immune checkpoint inhibitors (ICI) improve clinical outcomes in triple-negative breast cancer (TNBC) patients. However, a subset of patients does not respond to treatment. Biomarkers that show ICI predictive potential in other solid tumors, such as levels of PD-L1 and the tumor mutational burden, among others, show a modest predictive performance in patients with TNBC. METHODS: We built machine learning models based on pre-ICI treatment gene expression profiles to construct gene expression classifiers to identify primary TNBC ICI-responder patients. This study involved 188 ICI-naïve and 721 specimens treated with ICI plus chemotherapy, including TNBC tumors, HR+/HER2- breast tumors, and other solid non-breast tumors. RESULTS: The 37-gene TNBC ICI predictive (TNBC-ICI) classifier performs well in predicting pathological complete response (pCR) to ICI plus chemotherapy on an independent TNBC validation cohort (AUC = 0.86). The TNBC-ICI classifier shows better performance than other molecular signatures, including PD-1 (PDCD1) and PD-L1 (CD274) gene expression (AUC = 0.67). Integrating TNBC-ICI with molecular signatures does not improve the efficiency of the classifier (AUC = 0.75). TNBC-ICI displays a modest accuracy in predicting ICI response in two different cohorts of patients with HR + /HER2- breast cancer (AUC = 0.72 to pembrolizumab and AUC = 0.75 to durvalumab). Evaluation of six cohorts of patients with non-breast solid tumors treated with ICI plus chemotherapy shows overall poor performance (median AUC = 0.67). CONCLUSION: TNBC-ICI predicts pCR to ICI plus chemotherapy in patients with primary TNBC. The study provides a guide to implementing the TNBC-ICI classifier in clinical studies. Further validations will consolidate a novel predictive panel to improve the treatment decision-making for patients with TNBC.

Triple-Negative Breast Cancer (TNBC) is an aggressive type of breast cancer, responsible for a substantial burden of breast cancer-related deaths. In recent years, immunotherapy, a therapy that triggers the patient's immune system to attack the tumor, has arisen as a promising treatment in various cancers, including TNBC. However, a subset of patients with TNBC does not respond to this treatment. Here, we employed advanced computational techniques to predict response to immunotherapy plus chemotherapy in patients with primary TNBC. Our method is more accurate than using other existing markers, such as PD-L1, but is not very accurate in patients with non-TNBC breast cancers or non-breast cancers. This method could potentially be used to better select patients for immunotherapy, upfront, avoiding the side effects and costs of treating patients in which immunotherapy might not work.