Transcription factors TEAD2 and E2A globally repress acetyl-CoA synthesis to promote tumorigenesis.

Acetyl-coenzyme A (acetyl-CoA) plays an important role in metabolism, gene expression, signaling, and other cellular processes via transfer of its acetyl group to proteins and metabolites. However, the synthesis and usage of acetyl-CoA in disease states such as cancer are poorly characterized. Here, we investigated global acetyl-CoA synthesis and protein acetylation in a mouse model and patient samples of hepatocellular carcinoma (HCC). Unexpectedly, we found that acetyl-CoA levels are decreased in HCC due to transcriptional downregulation of all six acetyl-CoA biosynthesis pathways. This led to hypo-acetylation specifically of non-histone proteins, including many enzymes in metabolic pathways. Importantly, repression of acetyl-CoA synthesis promoted oncogenic dedifferentiation and proliferation. Mechanistically, acetyl-CoA synthesis was repressed by the transcription factors TEAD2 and E2A, previously unknown to control acetyl-CoA synthesis. Knockdown of TEAD2 and E2A restored acetyl-CoA levels and inhibited tumor growth. Our findings causally link transcriptional reprogramming of acetyl-CoA metabolism, dedifferentiation, and cancer.

Genomic characterization of metastatic breast cancers.

Metastasis is the main cause of death for patients with breast cancer. Many studies have characterized the genomic landscape of breast cancer during its early stages. However, there is evidence that genomic alterations are acquired during the evolution of cancers from their early to late stages, and that the genomic landscape of early cancers is not representative of that of lethal cancers1-7. Here we investigated the landscape of somatic alterations in 617 metastatic breast cancers. Nine driver genes (TP53, ESR1, GATA3, KMT2C, NCOR1, AKT1, NF1, RIC8A and RB1) were more frequently mutated in metastatic breast cancers that expressed hormone receptors (oestrogen and/or progesterone receptors; HR+) but did not have high levels of HER2 (HER2-; n = 381), when compared to early breast cancers from The Cancer Genome Atlas. In addition, 18 amplicons were more frequently observed in HR+/HER2- metastatic breast cancers. These cancers showed an increase in mutational signatures S2, S3, S10, S13 and S17. Among the gene alterations that were enriched in HR+/HER2- metastatic breast cancers, mutations in TP53, RB1 and NF1, together with S10, S13 and S17, were associated with poor outcome. Metastatic triple-negative breast cancers showed an increase in the frequency of somatic biallelic loss-of-function mutations in genes related to homologous recombination DNA repair, compared to early triple-negative breast cancers (7% versus 2%). Finally, metastatic breast cancers showed an increase in mutational burden and clonal diversity compared to early breast cancers. Thus, the genomic landscape of metastatic breast cancer is enriched in clinically relevant genomic alterations and is more complex than that of early breast cancer. The identification of genomic alterations associated with poor outcome will allow earlier and better selection of patients who require the use of treatments that are still in clinical trials. The genetic complexity observed in advanced breast cancer suggests that such treatments should be introduced as early as possible in the disease course.

Author Correction: Genomic characterization of metastatic breast cancers.

An Amendment to this paper has been published and can be accessed via a link at the top of the paper.

The protein histidine phosphatase LHPP is a tumour suppressor.

Histidine phosphorylation, the so-called hidden phosphoproteome, is a poorly characterized post-translational modification of proteins. Here we describe a role of histidine phosphorylation in tumorigenesis. Proteomic analysis of 12 tumours from an mTOR-driven hepatocellular carcinoma mouse model revealed that NME1 and NME2, the only known mammalian histidine kinases, were upregulated. Conversely, expression of the putative histidine phosphatase LHPP was downregulated specifically in the tumours. We demonstrate that LHPP is indeed a protein histidine phosphatase. Consistent with these observations, global histidine phosphorylation was significantly upregulated in the liver tumours. Sustained, hepatic expression of LHPP in the hepatocellular carcinoma mouse model reduced tumour burden and prevented the loss of liver function. Finally, in patients with hepatocellular carcinoma, low expression of LHPP correlated with increased tumour severity and reduced overall survival. Thus, LHPP is a protein histidine phosphatase and tumour suppressor, suggesting that deregulated histidine phosphorylation is oncogenic.

PipeIT2: A tumor-only somatic variant calling workflow for molecular diagnostic Ion Torrent sequencing data.

Precision oncology relies on the accurate identification of somatic mutations in cancer patients. While the sequencing of the tumoral tissue is frequently part of routine clinical care, the healthy counterparts are rarely sequenced. We previously published PipeIT, a somatic variant calling workflow specific for Ion Torrent sequencing data enclosed in a Singularity container. PipeIT combines user-friendly execution, reproducibility and reliable mutation identification, but relies on matched germline sequencing data to exclude germline variants. Expanding on the original PipeIT, here we describe PipeIT2 to address the clinical need to define somatic mutations in the absence of germline control. We show that PipeIT2 achieves a > 95% recall for variants with variant allele fraction >10%, reliably detects driver and actionable mutations and filters out most of the germline mutations and sequencing artifacts. With its performance, reproducibility, and ease of execution, PipeIT2 is a valuable addition to molecular diagnostics laboratories.

PI3K inhibition circumvents resistance to SHP2 blockade in metastatic triple-negative breast cancer.

The protein tyrosine phosphatase SHP2 activates oncogenic pathways downstream of most receptor tyrosine kinases (RTK) and has been implicated in various cancer types, including the highly aggressive subtype of triple-negative breast cancer (TNBC). Although allosteric inhibitors of SHP2 have been developed and are currently being evaluated in clinical trials, neither the mechanisms of the resistance to these agents, nor the means to circumvent such resistance have been clearly defined. The PI3K signaling pathway is also hyperactivated in breast cancer and contributes to resistance to anticancer therapies. When PI3K is inhibited, resistance also develops for example via activation of RTKs. We therefore assessed the effect of targeting PI3K and SHP2 alone or in combination in preclinical models of metastatic TNBC. In addition to the beneficial inhibitory effects of SHP2 alone, dual PI3K/SHP2 treatment decreased primary tumor growth synergistically, blocked the formation of lung metastases, and increased survival in preclinical models. Mechanistically, transcriptome and phospho-proteome analyses revealed that resistance to SHP2 inhibition is mediated by PDGFRß-evoked activation of PI3K signaling. Altogether, our data provide a rationale for co-targeting of SHP2 and PI3K in metastatic TNBC.

Multi-omics subtyping of hepatocellular carcinoma patients using a Bayesian network mixture model.

Comprehensive molecular characterization of cancer subtypes is essential for predicting clinical outcomes and searching for personalized treatments. We present bnClustOmics, a statistical model and computational tool for multi-omics unsupervised clustering, which serves a dual purpose: Clustering patient samples based on a Bayesian network mixture model and learning the networks of omics variables representing these clusters. The discovered networks encode interactions among all omics variables and provide a molecular characterization of each patient subgroup. We conducted simulation studies that demonstrated the advantages of our approach compared to other clustering methods in the case where the generative model is a mixture of Bayesian networks. We applied bnClustOmics to a hepatocellular carcinoma (HCC) dataset comprising genome (mutation and copy number), transcriptome, proteome, and phosphoproteome data. We identified three main HCC subtypes together with molecular characteristics, some of which are associated with survival even when adjusting for the clinical stage. Cluster-specific networks shed light on the links between genotypes and molecular phenotypes of samples within their respective clusters and suggest targets for personalized treatments.

Systematic identification of novel cancer genes through analysis of deep shRNA perturbation screens.

Systematic perturbation screens provide comprehensive resources for the elucidation of cancer driver genes. The perturbation of many genes in relatively few cell lines in such functional screens necessitates the development of specialized computational tools with sufficient statistical power. Here we developed APSiC (Analysis of Perturbation Screens for identifying novel Cancer genes) to identify genetic drivers and effectors in perturbation screens even with few samples. Applying APSiC to the shRNA screen Project DRIVE, APSiC identified well-known and novel putative mutational and amplified cancer genes across all cancer types and in specific cancer types. Additionally, APSiC discovered tumor-promoting and tumor-suppressive effectors, respectively, for individual cancer types, including genes involved in cell cycle control, Wnt/ß-catenin and hippo signalling pathways. We functionally demonstrated that LRRC4B, a putative novel tumor-suppressive effector, suppresses proliferation by delaying cell cycle and modulates apoptosis in breast cancer. We demonstrate APSiC is a robust statistical framework for discovery of novel cancer genes through analysis of large-scale perturbation screens. The analysis of DRIVE using APSiC is provided as a web portal and represents a valuable resource for the discovery of novel cancer genes.

Multi-omics data integration reveals novel drug targets in hepatocellular carcinoma.

BACKGROUND: Genetic aberrations in hepatocellular carcinoma (HCC) are well known, but the functional consequences of such aberrations remain poorly understood. RESULTS: Here, we explored the effect of defined genetic changes on the transcriptome, proteome and phosphoproteome in twelve tumors from an mTOR-driven hepatocellular carcinoma mouse model. Using Network-based Integration of multi-omiCS data (NetICS), we detected 74 'mediators' that relay via molecular interactions the effects of genetic and miRNA expression changes. The detected mediators account for the effects of oncogenic mTOR signaling on the transcriptome, proteome and phosphoproteome. We confirmed the dysregulation of the mediators YAP1, GRB2, SIRT1, HDAC4 and LIS1 in human HCC. CONCLUSIONS: This study suggests that targeting pathways such as YAP1 or GRB2 signaling and pathways regulating global histone acetylation could be beneficial in treating HCC with hyperactive mTOR signaling.

Nestin and CD34 expression in colorectal cancer predicts improved overall survival.

BACKGROUND: Nestin, a class VI intermediate filament protein of the cytoskeleton, and CD34, a transmembrane phosphoglycoprotein, are markers of progenitor cells. This study aimed to evaluate their expression and clinical significance in colorectal cancer. METHODS: A clinically annotated tissue microarray, including 599 patients with colorectal cancer, was analyzed by immunohistochemistry. Furthermore, nestin and CD34 correlations with HIF-1a and a panel of cytokines and chemokines were assessed using quantitative reverse transcription PCR and The Cancer Genome Atlas dataset. RESULTS: Expression of nestin and CD34 was observed only in the tumor stroma. Patients displaying high expression of nestin and CD34 demonstrated higher rates of T1 and T2 tumors (p = .020), lower vascular invasion (p < .001) and improved 5-year overall survival (65%; 95% CI = 55-73 vs 45%; 95% CI = 37-53) after adjusting for clinicopathological characteristics (HR: 0.67; 95% CI = 0.46-0.96). A moderate to strong correlation (r = 0.37-0.78, p < .03) of nestin and CD34 was demonstrated for the following markers; HIF-1α, CD4, CD8, FOXP3, IRF1, GATA3, CCL2, CCL3, CXCL12 and CCL21. CONCLUSIONS: Combined expression of nestin and CD34 expression is associated with better overall survival possibly by modulating a favorable immune response.

Infiltration by myeloperoxidase-positive neutrophils is an independent prognostic factor in breast cancer.

PURPOSE: Myeloperoxidase (MPO) is an enzyme secreted by neutrophil granulocytes as a result of phagocytosis during inflammation. In colorectal cancer, tumour infiltration by MPO expressing cells has been shown to be independently associated with a favourable prognosis. In this study, we explored the role of MPO-positive cell infiltration and its prognostic significance in invasive breast cancer. METHODS: We performed immunohistochemical staining for MPO on multiple tissue microarrays comprising a total of 928 human breast cancer samples with detailed clinical-pathological annotation and outcome data. RESULTS: MPO-positive cell infiltration (≥ 5 cells/tissue punch) was found in 150 (16%) of the 928 evaluable breast cancer cases. In univariate survival analyses, infiltration by MPO-positive cells was associated with a significantly better overall survival (p < 0.001). In subset univariate analyses, the infiltration by MPO-positive cells was associated with significantly better overall survival in the Luminal B/HER2-negative subtype (p = 0.005), the HER2 enriched subtype (p = 0.011), and the Triple Negative subtype (p < 0.001). In multivariate analysis, MPO expression proved to be an independent prognostic factor for improved overall survival (p < 0.001). CONCLUSIONS: This is the first study to show that infiltration of MPO-positive cells is an independent prognostic biomarker for improved overall survival in human breast cancer.

NGS-pipe: a flexible, easily extendable and highly configurable framework for NGS analysis.

Motivation: Next-generation sequencing is now an established method in genomics, and massive amounts of sequencing data are being generated on a regular basis. Analysis of the sequencing data is typically performed by lab-specific in-house solutions, but the agreement of results from different facilities is often small. General standards for quality control, reproducibility and documentation are missing. Results: We developed NGS-pipe, a flexible, transparent and easy-to-use framework for the design of pipelines to analyze whole-exome, whole-genome and transcriptome sequencing data. NGS-pipe facilitates the harmonization of genomic data analysis by supporting quality control, documentation, reproducibility, parallelization and easy adaptation to other NGS experiments. Availability and implementation: https://github.com/cbg-ethz/NGS-pipe. Contact: niko.beerenwinkel@bsse.ethz.ch.

Preoperative plasma fatty acid metabolites inform risk of prostate cancer progression and may be used for personalized patient stratification.

BACKGROUND: Little is known about the relationship between the metabolite profile of plasma from pre-operative prostate cancer (PCa) patients and the risk of PCa progression. In this study we investigated the association between pre-operative plasma metabolites and risk of biochemical-, local- and metastatic-recurrence, with the aim of improving patient stratification. METHODS: We conducted a case-control study within a cohort of PCa patients recruited between 1996 and 2015. The age-matched primary cases (n = 33) were stratified in low risk, high risk without progression and high risk with progression as defined by the National Comprehensive Cancer Network. These samples were compared to metastatic (n = 9) and healthy controls (n = 10). The pre-operative plasma from primary cases and the plasma from metastatic patients and controls were assessed with untargeted metabolomics by LC-MS. The association between risk of progression and metabolite abundance was calculated using multivariate Cox proportional-hazard regression and the relationship between metabolites and outcome was calculated using median cut-off normalized values of metabolite abundance by Log-Rank test using the Kaplan Meier method. RESULTS: Medium-chain acylcarnitines (C6-C12) were positively associated with the risk of PSA progression (p = 0.036, median cut-off) while long-chain acylcarnitines (C14-C16) were inversely associated with local (p = 0.034) and bone progression (p = 0.0033). In primary cases, medium-chain acylcarnitines were positively associated with suberic acid, which also correlated with the risk of PSA progression (p = 0.032, Log-Rank test). In the metastatic samples, this effect was consistent for hexanoylcarnitine, L.octanoylcarnitine and decanoylcarnitine. Medium-chain acylcarnitines and suberic acid displayed the same inverse association with tryptophan, while indoleacetic acid, a breakdown product of tryptophan metabolism was strongly associated with PSA (p = 0.0081, Log-Rank test) and lymph node progression (p = 0.025, Log-Rank test). These data were consistent with the increased expression of indoleamine 2,3 dioxygenase (IDO1) in metastatic versus primary samples (p = 0.014). Finally, functional experiments revealed a synergistic effect of long chain fatty acids in combination with dihydrotestosterone administration on the transcription of androgen responsive genes. CONCLUSIONS: This study strengthens the emerging link between fatty acid metabolism and PCa progression and suggests that measuring levels of medium- and long-chain acylcarnitines in pre-operative patient plasma may provide a basis for improving patient stratification.

Expression of RET is associated with Oestrogen receptor expression but lacks prognostic significance in breast cancer.

BACKGROUND: The Rearranged during Transfection (RET) protein is overexpressed in a subset of Estrogen Receptor (ER) positive breast cancer, with both signalling pathways functionally interacting. This cross-talk plays a pivotal role in the resistance of breast cancer cells to anti-endocrine therapies, and RET expression is assumed to correlate with poor prognosis based on findings in small patient cohorts. The aim of our study was to investigate the impact of RET expression on patient outcome in human breast cancer. METHODS: We performed an immunohistochemical analysis of RET protein expression on a tissue microarray encompassing 990 breast cancer patients and correlated its expression with clinicopathological parameters and survival data. RESULTS: Expression of RET was detected in 409 out of 990 cases (41.3%). RET and ER expression significantly correlated (p < 0.0001). The Luminal B HER2-positive subtype showed the highest expression rate (48.9%). In univariate and multivariate survival analyses, RET expression had no impact on overall survival. CONCLUSION: We confirmed the co-expression of RET and ER, but we did not find RET expression to be an independent prognostic factor in human breast cancer. Clinical trials with newly developed RET inhibitors are needed to evaluate if RET inhibition has a beneficial impact on patient survival in ER positive breast cancer.

Contralateral breast cancers: Independent cancers or metastases?

A cancer in the contralateral breast in a woman with a previous or synchronous breast cancer is typically considered to be an independent primary tumor. Emerging evidence suggests that in a small subset of these cases the second tumor represents a metastasis. We sought to investigate the issue using massively parallel sequencing targeting 254 genes recurrently mutated in breast cancer. We examined the tumor archives at Memorial Sloan Kettering Cancer Center for the period 1995-2006 to identify cases of contralateral breast cancer where surgery for both tumors was performed at the Center. We report results from 49 patients successfully analyzed by a targeted massively parallel sequencing assay. Somatic mutations and copy number alterations were defined by state-of-the-art algorithms. Clonal relatedness was evaluated by statistical tests specifically designed for this purpose. We found evidence that the tumors in contralateral breasts were clonally related in three cases (6%) on the basis of matching mutations at codons where somatic mutations are rare. Clinical data and the presence of similar patterns of gene copy number alterations were consistent with metastasis for all three cases. In three additional cases, there was a solitary matching mutation at a common PIK3CA locus. The results suggest that a subset of contralateral breast cancers represent metastases rather than independent primary tumors. Massively parallel sequencing analysis can provide important evidence to clarify the diagnosis. However, given the inter-tumor mutational heterogeneity in breast cancer, sufficiently large gene panels need to be employed to define clonality convincingly in all cases.

The Role of Long Non-Coding RNAs in Hepatocarcinogenesis.

Whole-transcriptome analyses have revealed that a large proportion of the human genome is transcribed in non-protein-coding transcripts, designated as long non-coding RNAs (lncRNAs). Rather than being "transcriptional noise", increasing evidence indicates that lncRNAs are key players in the regulation of many biological processes, including transcription, post-translational modification and inhibition and chromatin remodeling. Indeed, lncRNAs are widely dysregulated in human cancers, including hepatocellular carcinoma (HCC). Functional studies are beginning to provide insights into the role of oncogenic and tumor suppressive lncRNAs in the regulation of cell proliferation and motility, as well as oncogenic and metastatic potential in HCC. A better understanding of the molecular mechanisms and the complex network of interactions in which lncRNAs are involved could reveal novel diagnostic and prognostic biomarkers. Crucially, it may provide novel therapeutic opportunities to add to the currently limited number of therapeutic options for HCC patients. In this review, we summarize the current status of the field, with a focus on the best characterized dysregulated lncRNAs in HCC.

Leiomyoma with bizarre nuclei: a morphological, immunohistochemical and molecular analysis of 31 cases.

Leiomyomas associated with hereditary leiomyomatosis and renal cell carcinoma syndrome and leiomyomas with bizarre nuclei often show overlapping morphological features, in particular cells with prominent eosinophilic nucleoli, perinucleolar halos, and eosinophilic cytoplasmic inclusions. Although hereditary leiomyomatosis and renal cell carcinoma syndrome is defined by fumarate hydratase (FH) germline mutations, resulting in S-(2-succino)-cysteine (2SC) formation, it is unknown whether leiomyomas with bizarre nuclei show similar alterations. In this study, we evaluated the morphology and FH/2SC immunoprofile of 31 leiomyomas with bizarre nuclei. DNA from tumor and normal tissues from 24 cases was subjected to massively parallel sequencing targeting 410 key cancer genes. Somatic genetic alterations were detected using state-of-the-art bioinformatics algorithms. No patient reported a personal history of renal neoplasia or cutaneous leiomyomas, but one had a family history of renal cell carcinoma while another had a family history of uterine leiomyomas. Aberrant FH/2SC expression was noted in 17 tumors (16 FH-negative/2SC-positive, 1 FH-positive/2SC-positive). On univariate analysis, staghorn vessels, eosinophilic cytoplasmic inclusions, diffuse distribution of prominent eosinophilic nucleoli with perinucleolar halos, and an 'alveolar pattern of edema' were associated with an abnormal immunoprofile, but only staghorn vessels remained significant on multivariate analysis. Massively parallel sequencing analysis (n=24) revealed that 13/14 tumors with aberrant FH/2SC immunoprofile harbored somatic FH somatic genetic alterations, including homozygous deletions (n=9), missense mutations coupled with loss of heterozygosity (n=3), and a splice site mutation (n=1), whereas no somatic FH mutations/deletions were found in tumors with normal immunoprofile (n=10; P<0.0001). Leiomyomas with bizarre nuclei with normal FH/2SC staining pattern more frequently harbored TP53 and/or RB1 alterations than those with aberrant FH/2SC immunoprofile (60 vs 14%; P=0.032). These data demonstrate that leiomyomas with bizarre nuclei are morphologically and genetically heterogeneous and that hereditary leiomyomatosis and renal cell carcinoma syndrome-related morphological features, abnormal FH/2SC staining, and somatic FH mutations/deletions can be seen in a subset of sporadic tumors.

Uterine adenosarcomas are mesenchymal neoplasms.

Uterine adenosarcomas (UAs) are biphasic lesions composed of a malignant mesenchymal (ie stromal) component and an epithelial component. UAs are generally low-grade and have a favourable prognosis, but may display sarcomatous overgrowth (SO), which is associated with a worse outcome. We hypothesized that, akin to breast fibroepithelial lesions, UAs are mesenchymal neoplasms in which clonal somatic genetic alterations are restricted to the mesenchymal component. To characterize the somatic genetic alterations in UAs and to test this hypothesis, we subjected 20 UAs to a combination of whole-exome (n = 6), targeted capture (n = 13) massively parallel sequencing (MPS) and/or RNA sequencing (n = 6). Only three genes, FGFR2, KMT2C and DICER1, were recurrently mutated, all in 2/19 cases; however, 26% (5/19) and 21% (4/19) of UAs harboured MDM2/CDK4/HMGA2 and TERT gene amplification, respectively, and two cases harboured fusion genes involving NCOA family members. Using a combination of laser-capture microdissection and in situ techniques, we demonstrated that the somatic genetic alterations detected by MPS were restricted to the mesenchymal component. Furthermore, mitochondrial DNA sequencing of microdissected samples revealed that epithelial and mesenchymal components of UAs were clonally unrelated. In conclusion, here we provide evidence that UAs are genetically heterogeneous lesions and mesenchymal neoplasms.

Microglandular adenosis associated with triple-negative breast cancer is a neoplastic lesion of triple-negative phenotype harbouring TP53 somatic mutations.

Microglandular adenosis (MGA) is a rare proliferative lesion of the breast composed of small glands lacking myoepithelial cells and lined by S100-positive, oestrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative epithelial cells. There is evidence to suggest that MGA may constitute a non-obligate precursor of triple-negative breast cancer (TNBC). We sought to define the genomic landscape of pure MGA and of MGA, atypical MGA (AMGA) and associated TNBCs, and to determine whether synchronous MGA, AMGA, and TNBCs would be clonally related. Two pure MGAs and eight cases of MGA and/or AMGA associated with in situ or invasive TNBC were collected, microdissected, and subjected to massively parallel sequencing targeting all coding regions of 236 genes recurrently mutated in breast cancer or related to DNA repair. Pure MGAs lacked clonal non-synonymous somatic mutations and displayed limited copy number alterations (CNAs); conversely, all MGAs (n = 7) and AMGAs (n = 3) associated with TNBC harboured at least one somatic non-synonymous mutation (range 3-14 and 1-10, respectively). In all cases where TNBCs were analyzed, identical TP53 mutations and similar patterns of gene CNAs were found in the MGA and/or AMGA and in the associated TNBC. In the MGA/AMGA associated with TNBC lacking TP53 mutations, somatic mutations affecting PI3K pathway-related genes (eg PTEN, PIK3CA, and INPP4B) and tyrosine kinase receptor signalling-related genes (eg ERBB3 and FGFR2) were identified. At diagnosis, MGAs associated with TNBC were found to display subclonal populations, and clonal shifts in the progression from MGA to AMGA and/or to TNBC were observed. Our results demonstrate the heterogeneity of MGAs, and that MGAs associated with TNBC, but not necessarily pure MGAs, are genetically advanced, clonal, and neoplastic lesions harbouring recurrent mutations in TP53 and/or other cancer genes, supporting the notion that a subset of MGAs and AMGAs may constitute non-obligate precursors of TNBCs.

Genetic alterations of triple negative breast cancer by targeted next-generation sequencing and correlation with tumor morphology.

Triple negative breast cancer represents a heterogeneous group of breast carcinomas, both at the histologic and genetic level. Although recent molecular studies have comprehensively characterized the genetic landscape of these tumors, few have integrated a detailed histologic examination into the analysis. In this study, we defined the genetic alterations in 39 triple negative breast cancers using a high-depth targeted massively parallel sequencing assay and correlated the findings with a detailed morphologic analysis. We obtained representative frozen tissue of primary triple negative breast cancers from patients treated at our institution between 2002 and 2010. We characterized tumors according to their histologic subtype and morphologic features. DNA was extracted from paired frozen primary tumor and normal tissue samples and was subjected to a targeted massively parallel sequencing platform comprising 229 cancer-associated genes common across all experiments. The average number of non-synonymous mutations was 3 (range 0-10) per case. The most frequent somatic alterations were mutations in TP53 (74%) and PIK3CA (10%) and MYC amplifications (26%). Triple negative breast cancers with apocrine differentiation less frequently harbored TP53 mutations (25%) and MYC gains (0%), and displayed a high mutation frequency in PIK3CA and other PI3K signaling pathway-related genes (75%). Using a targeted massively parallel sequencing platform, we identified the key somatic genetic alterations previously reported in triple negative breast cancers. Furthermore, our findings show that triple negative breast cancers with apocrine differentiation constitute a distinct subset, characterized by a high frequency of PI3K pathway alterations similar to luminal subtypes of breast cancer.