Intratumor heterogeneity and precision of microarray-based predictors of breast cancer biology and clinical outcome

Radiomic nomogram based on bi-parametric magnetic resonance imaging to predict the International Society of Urological Pathology grading ≥ 3 prostate cancer: a multicenter study

PURPOSE

To create a reliable radiomic nomogram for the prediction of the International Society of Urological Pathology (ISUP) grading ≥ 3 prostate cancer (PCa) patients.

METHODS

patients with verified PCa were obtained from three different hospitals. The patients were divided into training, internal validation, and two external validation groups. A radiomic signature (rad-score) extracted from T2WI, diffusion-weighted imaging, and apparent diffusion coefficient (ADC) maps were constructed in the training cohort. Eight clinical features were performed to develop a clinical model using univariate and multivariate logistic regression. The combined model incorporated the radiomic signature and clinical model. The model's performance was assessed by the receiver operating characteristic (ROC) curve.

RESULTS

Rad-score, magnetic resonance imaging T-stage, and ADC value were significant predictors of ISUP ≥ 3 PCa. A nomogram of these three factors was shown to have greater diagnostic accuracy than using only the radiomic signature or clinical model alone. The area under the ROC curve was 0.85, 0.88, 0.81, 0.81 for the training, internal, and two external validation cohorts, respectively. In the stratified analysis based on the MR scanner model, the area under the ROC curve of predicting ISUP ≥ 3 PCa for GE, Siemens, and combined groups were 0.84, 0.83, and 0.84, respectively, in the combined training group and an internal validation group.

CONCLUSIONS

The proposed nomogram has the potential to predict the differentiation degree of ISUP PCa patients.

Fusion Radiomics-Based Prediction of Response to Neoadjuvant Chemotherapy for Osteosarcoma

RATIONALE AND OBJECTIVES

Neoadjuvant chemotherapy (NAC) is the most crucial prognostic factor for osteosarcoma (OS), it significantly prolongs progression-free survival and improves the quality of life. This study aims to develop a deep learning radiomics (DLR) model to accurately predict the response to NAC in patients diagnosed with OS using preoperative MR images.

METHODS

We reviewed axial T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted (T1CE) of 106 patients pathologically confirmed as OS. First, the Auto3DSeg framework was utilized for automated OS segmentation. Second, using three feature extraction methods, nine risk classification models were constructed based on three classifiers. The area under the receiver operating curve (AUC), sensitivity, specificity, accuracy, negative predictive value and positive predictive value were calculated for performance evaluation. Additionally, we developed a deep learning radiomics nomogram with clinical indicators.

RESULTS

The model for OS automatic segmentation achieved a Dice coefficient of 0.868 across datasets. To predict the response to NAC, the DLR model achieved the highest prediction performance with an accuracy of 93.8% and an AUC of 0.961 in the test sets. We used calibration curves to assess the predictive ability of the models and performed decision curve analysis to evaluate the clinical net benefit of the DLR model.

CONCLUSION

The DLR model can serve as a pragmatic prediction tool, capable of identifying patients with poor response to NAC, providing information for risk counseling, and assisting in making clinical treatment decisions. Poor responders are better advised to undergo immunotherapy and receive the best supportive care.

Radiogenomic-based multiomic analysis reveals imaging intratumor heterogeneity phenotypes and therapeutic targets

Intratumor heterogeneity (ITH) profoundly affects therapeutic responses and clinical outcomes. However, the widespread methods for assessing ITH based on genomic sequencing or pathological slides, which rely on limited tissue samples, may lead to inaccuracies due to potential sampling biases. Using a newly established multicenter breast cancer radio-multiomic dataset (n = 1474) encompassing radiomic features extracted from dynamic contrast-enhanced magnetic resonance images, we formulated a noninvasive radiomics methodology to effectively investigate ITH. Imaging ITH (IITH) was associated with genomic and pathological ITH, predicting poor prognosis independently in breast cancer. Through multiomic analysis, we identified activated oncogenic pathways and metabolic dysregulation in high-IITH tumors. Integrated metabolomic and transcriptomic analyses highlighted ferroptosis as a vulnerability and potential therapeutic target of high-IITH tumors. Collectively, this work emphasizes the superiority of radiomics in capturing ITH. Furthermore, we provide insights into the biological basis of IITH and propose therapeutic targets for breast cancers with elevated IITH.

Reproducibility and intratumoral heterogeneity of the PAM50 breast cancer assay

BACKGROUND

The PAM50 assay is used routinely in clinical practice to determine breast cancer prognosis and management; however, research assessing how technical variation and intratumoral heterogeneity contribute to misclassification and reproducibility of these tests is limited.

METHODS

We evaluated the impact of intratumoral heterogeneity on the reproducibility of results for the PAM50 assay by testing RNA extracted from formalin-fixed paraffin embedded breast cancer blocks sampled at distinct spatial locations. Samples were classified according to intrinsic subtype (Luminal A, Luminal B, HER2-enriched, Basal-like, or Normal-like) and risk of recurrence with proliferation score (ROR-P, high, medium, or low). Intratumoral heterogeneity and technical reproducibility (replicate assays on the same RNA) were assessed as percent categorical agreement between paired intratumoral and replicate samples. Euclidean distances between samples, calculated across the PAM50 genes and the ROR-P score, were compared for concordant vs. discordant samples.

RESULTS

Technical replicates (N = 144) achieved 93% agreement for ROR-P group and 90% agreement on PAM50 subtype. For spatially distinct biological replicates (N = 40 intratumoral replicates), agreement was lower (81% for ROR-P and 76% for PAM50 subtype). The Euclidean distances between discordant technical replicates were bimodal, with discordant samples showing higher Euclidian distance and biologic heterogeneity.

CONCLUSION

The PAM50 assay achieved very high technical reproducibility for breast cancer subtyping and ROR-P, but intratumoral heterogeneity is revealed by the assay in a small proportion of cases.

A Comprehensive Landscape of Imaging Feature-Associated RNA Expression Profiles in Human Breast Tissue

The expression abundance of transcripts in nondiseased breast tissue varies among individuals. The association study of genotypes and imaging phenotypes may help us to understand this individual variation. Since existing reports mainly focus on tumors or lesion areas, the heterogeneity of pathological image features and their correlations with RNA expression profiles for nondiseased tissue are not clear. The aim of this study is to discover the association between the nucleus features and the transcriptome-wide RNAs. We analyzed both microscopic histology images and RNA-sequencing data of 456 breast tissues from the Genotype-Tissue Expression (GTEx) project and constructed an automatic computational framework. We classified all samples into four clusters based on their nucleus morphological features and discovered feature-specific gene sets. The biological pathway analysis was performed on each gene set. The proposed framework evaluates the morphological characteristics of the cell nucleus quantitatively and identifies the associated genes. We found image features that capture population variation in breast tissue associated with RNA expressions, suggesting that the variation in expression pattern affects population variation in the morphological traits of breast tissue. This study provides a comprehensive transcriptome-wide view of imaging-feature-specific RNA expression for healthy breast tissue. Such a framework could also be used for understanding the connection between RNA expression and morphology in other tissues and organs. Pathway analysis indicated that the gene sets we identified were involved in specific biological processes, such as immune processes.

Multiclassifier Radiomics Analysis of Ultrasound for Prediction of Extrathyroidal Extension in Papillary Thyroid Carcinoma in Children

Objective: To explore extrathyroidal extension (ETE) in children and adolescents with papillary thyroid carcinoma using a multiclassifier ultrasound radiomic model. Methods: In this study, data from 164 pediatric patients with papillary thyroid cancer (PTC) were retrospectively analyzed and patients were randomly divided into a training cohort (115) and a validation cohort (49) in a 7:3 ratio. To extract radiomics features from ultrasound images of the thyroid, areas of interest (ROIs) were delineated layer by layer along the edge of the tumor contour. The feature dimension was then reduced using the correlation coefficient screening method, and 16 features with a nonzero coefficient were chosen using Lasso. Then, in the training cohort, four supervised machine learning radiomics models (k-nearest neighbor, random forest, support vector machine [SVM], and LightGBM) were developed. ROC and decision-making curves were utilized to compare model performance, which was validated using validation cohorts. In addition, the SHapley Additive exPlanations (SHAP) framework was applied to explain the optimal model. Results: In the training cohort, the average area under the curve (AUC) was 0.880 (0.835-0.927), 0.873 (0.829-0.916), 0.999 (0.999-1.000), and 0.926 (0.892-0.926) for the SVM, KNN, random forest, and LightGBM, respectively. In the validation cohort, the AUC for the SVM was 0.784 (0.680-0.889), for the KNN, it was 0.720 (0.615-0.825), for the random forest, it was 0.728 (0.622-0.834), and for the LightGBM, it was 0.832 (0.742-0.921). Generally, the LightGBM model performed well in both the training and validation cohorts. From the SHAP results, original_shape_MinorAxisLength,original_shape_Maximum2DDiameterColumn, and wavelet-HHH_glszm_SmallAreaLowGrayLevelEmphasis have the most significant effect on the model. Conclusions: Our combined model based on machine learning and ultrasonic radiomics demonstrate the excellent predictive ability for extrathyroidal extension (ETE) in pediatric PTC.

Comparing Genetic Risk and Clinical Risk Classification in Luminal-like Breast Cancer Patients Using a 23-Gene Classifier

Background: A 23-gene classifier has been developed based on gene expression profiles of Taiwanese luminal-like breast cancer. We aim to stratify risk of relapse and identify patients who may benefit from adjuvant chemotherapy based on genetic model among distinct clinical risk groups. Methods: There were 248 luminal (hormone receptor-positive and human epidermal growth factor receptor II-negative) breast cancer patients with 23-gene classifier results. Using the modified Adjuvant! Online definition, clinical high/low-risk groups were tabulated with the genetic model. The primary endpoint was a recurrence-free interval (RFI) at 5 years. Results: There was a significant difference between the high/low-risk groups defined by the 23-gene classifier for the 5-year prognosis of recurrence (16 recurrences in high-risk and 3 recurrences in low-risk; log-rank test: p < 0.0001). Among the clinically high-risk group, the 5-year RFI of high risk defined by the 23-gene classifier was significantly higher than that of the low-risk group (15 recurrences in high-risk and 2 recurrences in low-risk; log-rank test: p < 0.0001). Conclusion: This study showed that 23-gene classifier can be used to stratify clinically high-risk patients into distinct survival patterns based on genomic risks and displays the potentiality to guide adjuvant chemotherapy. The 23-gene classifier can provide a better estimation of breast cancer prognosis which can help physicians make a better treatment decision.

Multi-region sequencing with spatial information enables accurate heterogeneity estimation and risk stratification in liver cancer

BACKGROUND

Numerous studies have used multi-region sampling approaches to characterize intra-tumor heterogeneity (ITH) in hepatocellular carcinoma (HCC). However, conventional multi-region sampling strategies do not preserve the spatial details of samples, and thus, the potential influences of spatial distribution on patient-wise ITH (represents the overall heterogeneity level of the tumor in a given patient) have long been overlooked. Furthermore, gene-wise transcriptional ITH (represents the expression pattern of genes across different intra-tumor regions) in HCC is also under-explored, highlighting the need for a comprehensive investigation.

METHODS

To address the problem of spatial information loss, we propose a simple and easy-to-implement strategy called spatial localization sampling (SLS). We performed multi-region sampling and sequencing on 14 patients with HCC, collecting a total of 75 tumor samples with spatial information and molecular data. Normalized diversity score and integrated heterogeneity score (IHS) were then developed to measure patient-wise and gene-wise ITH, respectively.

RESULTS

A significant correlation between spatial and molecular heterogeneity was uncovered, implying that spatial distribution of sampling sites did influence ITH estimation in HCC. We demonstrated that the normalized diversity score had the ability to overcome sampling location bias and provide a more accurate estimation of patient-wise ITH. According to this metric, HCC tumors could be divided into two classes (low-ITH and high-ITH tumors) with significant differences in multiple biological properties. Through IHS analysis, we revealed a highly heterogenous immune microenvironment in HCC and identified some low-ITH checkpoint genes with immunotherapeutic potential. We also constructed a low-heterogeneity risk stratification (LHRS) signature based on the IHS results which could accurately predict the survival outcome of patients with HCC on a single tumor biopsy sample.

CONCLUSIONS

This study provides new insights into the complex phenotypes of HCC and may serve as a guide for future studies in this field.

Identification of CD161 expression as a novel prognostic biomarker in breast cancer correlated with immune infiltration

Background:CD161 has been identified as a prognostic biomarker in many neoplasms, but its role in breast cancer (BC) has not been fully explained. We aimed to investigate the molecular mechanism and prognostic value of CD161 in BC.

Methods:CD161 expression profile was extracted from TIMER, Oncomine, UALCAN databases, and verified by the Gene Expression Omnibus (GEO) database and quantitative real-time polymerase chain reaction (qRT-PCR). The prognostic value of CD161 was assessed via GEPIA, Kaplan–Meier plotter and PrognoScan databases. The Cox regression and nomogram analyses were conducted to further validate the association between CD161 expression and survival. Gene set enrichment analysis (GSEA), Gene Ontology (GO) analysis, and KEGG pathway enrichment analysis were performed to probe the tumor-associated annotations of CD161. CIBERSORT and ssGSEA were employed to investigate the correlation between CD161 expression and immune cell infiltration in BC, and the result was verified by TIMER and TISIDB.

Results: Multiple BC cohorts showed that CD161 expression was decreased in BC, and a high CD161 expression was associated with a preferable prognosis. Therefore, we identified the combined model including CD161, age and PR status to predict the survival (C index = 0.78) of BC patients. Functional enrichment analysis indicated that CD161 and its co-expressed genes were closely related to several cancerous and immune signaling pathways, suggesting its involvement in immune response during cancer development. Moreover, immune infiltration analysis revealed that CD161 expression was correlated with immune infiltration.

Conclusion: Collectively, our findings revealed that CD161 may serve as a potential biomarker for favorable prognosis and a promising immune therapeutic target in BC.

Serial Tumor Molecular Profiling of Newly Diagnosed HER2-Negative Breast Cancers During Chemotherapy in Combination with Angiogenesis Inhibitors

Background

Breast cancers are heterogeneous with variable clinical courses and treatment responses.

Objective

We sought to evaluate dynamic changes in the molecular landscape of HER2-negative tumors treated with chemotherapy and anti-angiogenic agents.

Patients and Methods

Newly diagnosed HER2-negative breast cancer patients received low-dose sunitinib or bevacizumab prior to four 2-weekly cycles of dose-dense doxorubicin and cyclophosphamide. Tumor biopsies were obtained at baseline, after 2 weeks and after 8 weeks of chemotherapy. Next-generation sequencing was performed to assess for single nucleotide variants (SNVs) and copy number alterations (CNAs) of 440 cancer-related genes (ACTOnco^®). Observed genomic changes were correlated with the Miller-Payne histological response to treatment.

Results

Thirty-four patients received sunitinib and 18 received bevacizumab. In total, 77% were hormone receptor positive (HER2−/HR+) and 23% were triple negative breast cancers (TNBC). New therapy-induced mutations were infrequent, occurring only in 13%, and appeared early after a single cycle of treatment. Seventy-two percent developed changes in the variant allele frequency (VAF) of pathogenic SNVs; the majority (51%) of these changes occurred early at 2 weeks and were sustained for 8 weeks. Changes in VAF of SNVs were most commonly seen in the PI3K/mTOR/AKT pathway; 13% developed changes in pathogenic mutations, which potentially confer sensitivity to PIK3CA inhibitors. Tumors with poor Miller-Payne response to treatment were less likely to experience changes in VAF of SNVs compared with those with good response (50% [7/14] vs 15% [4/24] had no changes observed at any timepoint, p = 0.029).

Conclusions

Serial molecular profiling identifies early therapy-induced genomic alterations, which may guide future selection of targeted therapies in breast cancer patients who progress after standard chemotherapy.

Clinical trial registration

ClinicalTrials.gov: NCT02790580 (first posted June 6, 2016).

Supplementary Information

The online version contains supplementary material available at 10.1007/s11523-022-00886-x.

Magnetic Resonance Imaging Radiomics-Based Nomogram From Primary Tumor for Pretreatment Prediction of Peripancreatic Lymph Node Metastasis in Pancreatic Ductal Adenocarcinoma: A Multicenter Study

BACKGROUND

Determining the absence or presence of peripancreatic lymph nodal metastasis (PLNM) is important to the pathologic staging, prognostication, and guidance of treatment in pancreatic ductal adenocarcinoma (PDAC) patients. Computed tomography and MRI had a poor sensitivity and diagnostic accuracy in the assessment of PLNM.

PURPOSES

To develop and validate a 3 T MRI primary tumor radiomics-based nomogram from multicenter datasets for pretreatment prediction of the PLNM in PDAC patients.

STUDY TYPE

Retrospective.

SUBJECTS

A total of 251 patients (156 men and 95 women; mean age, 60.85 ± 8.23 years) with histologically confirmed pancreatic ductal adenocarcinoma from three hospitals.

FIELD STRENGTH AND SEQUENCES

A 3.0 T and fat-suppressed T1-weighted imaging.

ASSESSMENT

Quantitative imaging features were extracted from fat-suppressed T1-weighted (FS T1WI) images at the arterial phase.

STATISTICAL TESTS

Normally distributed data were compared by using t-tests, while the Mann-Whitney U test was used to evaluate non-normally distributed data. The diagnostic performances of the preoperative and postoperative nomograms were assessed in the external validation cohort with the area under receiver operating characteristics curve (AUC), calibration curve, and decision curve analysis (DCA). AUCs were compared with the De Long test. A p value below 0.05 was considered to be statistically significant.

RESULTS

The AUCs of magnetic resonance imaging (MRI) Rad-score were 0.868 (95% confidence level [CI]: 0.613-0.852) and 0.772 (95% CI: 0.659-0.879) in the training and internal validation cohort, respectively. The preoperative and postoperative nomograms could accurately predict PLNM in the training cohort (AUC = 0.909 and 0.851) and were validated in both the internal and external cohorts (AUC = 0.835 and 0.805, 0.808 and 0.733, respectively). DCA indicated that the two novel nomograms are of similar clinical usefulness.

DATA CONCLUSION

Pre-/postoperative nomograms and the constructed radiomics signature from primary tumor based on FS T1WI of arterial phase could serve as a potential tool to predict PLNM in patients with PDAC. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Radiomics Analysis of Computed Tomography for Prediction of Thyroid Capsule Invasion in Papillary Thyroid Carcinoma: A Multi-Classifier and Two-Center Study

OBJECTIVE

To investigate the application of computed tomography (CT)-based radiomics model for prediction of thyroid capsule invasion (TCI) in papillary thyroid carcinoma (PTC).

METHODS

This retrospective study recruited 412 consecutive PTC patients from two independent institutions and randomly assigned to training (n=265), internal test (n=114) and external test (n=33) cohorts. Radiomics features were extracted from non-contrast (NC) and artery phase (AP) CT scans. We also calculated delta radiomics features, which are defined as the absolute differences between the extracted radiomics features. One-way analysis of variance and least absolute shrinkage and selection operator were used to select optimal radiomics features. Then, six supervised machine learning radiomics models (k-nearest neighbor, logistic regression, decision tree, linear support vector machine [L-SVM], Gaussian-SVM, and polynomial-SVM) were constructed. Univariate was used to select clinicoradiological risk factors. Combined models including optimal radiomics features and clinicoradiological risk factors were constructed by these six classifiers. The prediction performance was evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

RESULTS

In the internal test cohort, the best combined model (L-SVM, AUC=0.820 [95% CI 0.758-0.888]) performed better than the best radiomics model (L-SVM, AUC = 0.733 [95% CI 0.654-0.812]) and the clinical model (AUC = 0.709 [95% CI 0.649-0.783]). Combined-L-SVM model combines 23 radiomics features and 1 clinicoradiological risk factor (CT-reported TCI). In the external test cohort, the AUC was 0.776 (0.625-0.904) in the combined-L-SVM model, showing that the model is stable. DCA demonstrated that the combined model was clinically useful.

CONCLUSIONS

Our combined model based on machine learning incorporated with CT radiomics features and the clinicoradiological risk factor shows good predictive ability for TCI in PTC.

Multi-Phase CT-Based Radiomics Nomogram for Discrimination Between Pancreatic Serous Cystic Neoplasm From Mucinous Cystic Neoplasm

Purpose

This study aimed to develop and verify a multi-phase (MP) computed tomography (CT)-based radiomics nomogram to differentiate pancreatic serous cystic neoplasms (SCNs) from mucinous cystic neoplasms (MCNs), and to compare the diagnostic efficacy of radiomics models for different phases of CT scans.

Materials and Methods

A total of 170 patients who underwent surgical resection between January 2011 and December 2018, with pathologically confirmed pancreatic cystic neoplasms (SCN=115, MCN=55) were included in this single-center retrospective study. Radiomics features were extracted from plain scan (PS), arterial phase (AP), and venous phase (VP) CT scans. Algorithms were performed to identify the optimal features to build a radiomics signature (Radscore) for each phase. All features from these three phases were analyzed to develop the MP-Radscore. A combined model comprised the MP-Radscore and imaging features from which a nomogram was developed. The accuracy of the nomogram was evaluated using receiver operating characteristic (ROC) curves, calibration tests, and decision curve analysis.

Results

For each scan phase, 1218 features were extracted, and the optimal ones were selected to construct the PS-Radscore (11 features), AP-Radscore (11 features), and VP-Radscore (12 features). The MP-Radscore (14 features) achieved better performance based on ROC curve analysis than any single phase did [area under the curve (AUC), training cohort: MP-Radscore 0.89, PS-Radscore 0.78, AP-Radscore 0.83, VP-Radscore 0.85; validation cohort: MP-Radscore 0.88, PS-Radscore 0.77, AP-Radscore 0.83, VP-Radscore 0.84]. The combination nomogram performance was excellent, surpassing those of all other nomograms in both the training cohort (AUC, 0.91) and validation cohort (AUC, 0.90). The nomogram also performed well in the calibration and decision curve analyses.

Conclusions

Radiomics for arterial and venous single-phase models outperformed the plain scan model. The combination nomogram that incorporated the MP-Radscore, tumor location, and cystic number had the best discriminatory performance and showed excellent accuracy for differentiating SCN from MCN.

MRI-based radiomics signature for pretreatment prediction of pathological response to neoadjuvant chemotherapy in osteosarcoma: a multicenter study

OBJECTIVE

To develop and validate a radiomics signature based on magnetic resonance imaging (MRI) from multicenter datasets for preoperative prediction of pathologic response to neoadjuvant chemotherapy (NAC) in patients with osteosarcoma.

METHODS

We retrospectively enrolled 102 patients with histologically confirmed osteosarcoma who received chemotherapy before treatment from 4 hospitals (68 in the primary cohort and 34 in the external validation cohort). Quantitative imaging features were extracted from contrast-enhanced fat-suppressed T1-weighted images (CE FS T1WI). Four classification methods, i.e., the least absolute shrinkage and selection operator logistic regression (LASSO-LR), support vector machine (SVM), Gaussian process (GP), and Naive Bayes (NB) algorithm, were compared for feature selection and radiomics signature construction. The predictive performance of the radiomics signatures was assessed with the area under receiver operating characteristics curve (AUC), calibration curve, and decision curve analysis (DCA).

RESULTS

Thirteen radiomics features selected based on the LASSO-LR classifier were adopted to construct the radiomics signature, which was significantly associated with the pathologic response. The prediction model achieved the best performance between good and poor responders with an AUC of 0.882 (95% CI, 0.837-0.918) in the primary cohort. Calibration curves showed good agreement. Similarly, findings were validated in the external validation cohort with good performance (AUC, 0.842 [95% CI, 0.793-0.883]) and good calibration. DCA analysis confirmed the clinical utility of the selected radiomics signature.

CONCLUSION

The constructed CE FS T1WI-radiomics signature with excellent performance could provide a potential tool to predict pathologic response to NAC in patients with osteosarcoma.

KEY POINTS

• The radiomics signature based on multicenter contrast-enhanced MRI was useful to predict response to NAC. • The prediction model obtained with the LASSO-LR classifier achieved the best performance. • The baseline clinical characteristics were not associated with response to NAC.

Screening and Identification of Key Common and Specific Genes and Their Prognostic Roles in Different Molecular Subtypes of Breast Cancer

Breast cancer is one of the most common cancers. Although the present molecular classification improves the treatment effect and prognosis of breast cancer, the heterogeneity of the molecular subtype remains very complex, and the applicability and effectiveness of treatment methods are still limited leading to poorer patient prognosis than expected. Further identification of more refined molecular typing based on gene expression profile will yield better understanding of the heterogeneity, improving treatment effects and prolonging prognosis of patients. Here, we downloaded the mRNA expression profiles and corresponding clinical data of patients with breast cancer from public databases and performed typical molecular typing using PAM50 (Prediction Analysis of Microarray 50) method. Comparative analyses were performed to screen the common and specific differentially expressed genes (DEGs) between cancer and corresponding para-cancerous tissues in each breast cancer subtype. The GO and KEGG analyses of the DEGs were performed to enrich the common and specific functional progress and signaling pathway involved in breast cancer subtypes. A total of 38 key common and specific DEGs were identified and selected based on the validated results, GO/KEGG enrichments, and the priority of expression, including four common DEGs and 34 specific DEGs in different subtypes. The prognostic value of these key common and specific DEGs was further analyzed to obtain useful potential markers in clinic. Finally, the potential roles and the specific prognostic values of the common and specific DEGs were speculated and summarized in total breast cancer and different subtype breast cancer based on the results of these analyses. The findings of our study provide the basis of more refined molecular typing of breast cancer, potential new therapeutic targets and prognostic markers for different breast cancer subtypes

Revisiting the Concept of Stress in the Prognosis of Solid Tumors: A Role for Stress Granules Proteins?

Simple Summary

Stress Granules (SGs) were discovered in 1999 and while the first decade of research has focused on some fundamental questions, the field is now investigating their role in human pathogenesis. Since then, evidences of a link between SGs and cancerology are accumulating in vitro and in vivo. In this work we summarized the role of SGs proteins in cancer development and their prognostic values. We find that level of expression of protein involved in SGs formation (and not mRNA level) could serve a prognostic marker in cancer. With this review we strongly suggest that SGs (proteins) could be targets of choice to block cancer development and counteract resistance to improve patients care.

Abstract

Cancer treatments are constantly evolving with new approaches to improve patient outcomes. Despite progresses, too many patients remain refractory to treatment due to either the development of resistance to therapeutic drugs and/or metastasis occurrence. Growing evidence suggests that these two barriers are due to transient survival mechanisms that are similar to those observed during stress response. We review the literature and current available open databases to study the potential role of stress response and, most particularly, the involvement of Stress Granules (proteins) in cancer. We propose that Stress Granule proteins may have prognostic value for patients.

A Radiomics Nomogram for the Preoperative Prediction of Lymph Node Metastasis in Pancreatic Ductal Adenocarcinoma

PURPOSE

To construct and verify a CT-based multidimensional nomogram for the evaluation of lymph node (LN) status in pancreatic ductal adenocarcinoma (PDAC).

MATERIALS AND METHODS

We retrospectively assessed data from 172 patients with clinicopathologically confirmed PDAC surgically resected between February 2014 and November 2016. Patients were assigned to either a training cohort (n = 121) or a validation cohort (n = 51). We acquired radiomics features from the preoperative venous phase (VP) CT images. The maximum relevance-minimum redundancy (mRMR) algorithm and the least absolute shrinkage and selection operator (LASSO) methods were used to select the optimal features. We used multivariable logistic regression to construct a combined radiomics model for visualization in the form of a nomogram. Performance of the nomogram was evaluated by the receiver operating characteristic (ROC) curve approach, calibration testing, and analysis of clinical usefulness.

RESULTS

A Rad score consisting of 10 LN status-related radiomics features was found to be significantly associated with the actual LN status (P < 0.01). A nomogram that consisted of Rad scores, CT-reported parenchymal atrophy, and CT-reported LN status performed well in terms of predictive power in the training cohort (area under the curve, 0.92), and this was confirmed in the validation cohort (area under the curve, 0.95). The nomogram also performed well in the calibration test and decision curve analysis, demonstrating its potential clinical value.

CONCLUSION

A multidimensional radiomics nomogram consisting of Rad scores, CT-reported parenchymal atrophy, and CT-reported LN status may contribute to the non-invasive evaluation of LN status in PDAC patients.

Concordance of Genomic Alterations between Circulating Tumor DNA and Matched Tumor Tissue in Chinese Patients with Breast Cancer

Purpose

Circulating tumor DNA (ctDNA) served as a noninvasive method with less side effects using peripheral blood. Given the studies on concordance rate between liquid and solid biopsies in Chinese breast cancer (BC) patients were limited, we sought to examine the concordance rate of different kinds of genomic alterations between paired tissue biopsies and ctDNA samples in Chinese BC cohorts.

Materials and Methods

In this study, we analyzed the genomic alteration profiles of 81 solid BC samples and 41 liquid BC samples. The concordance across 136 genes was evaluated.

Results

The median mutation counts per sample in 41 ctDNA samples was higher than the median in 81 tissue samples (p=0.0254; Wilcoxon rank sum test). For mutation at the protein-coding level, 39.0% (16/41) samples had at least one concordant mutation in two biopsies. 20.0% tissue-derived mutations could be detected via ctDNA-based sequencing, whereas 11.7% ctDNA-derived mutations could be found in paired tissues. At gene amplification level, the overall concordant rate was 68.3% (28/41). The concordant rate at gene level for each patient ranged from 83.8% (114/136) to 99.3% (135/136). And, the mean level of variant allele frequency (VAF) for concordant mutations in ctDNA was statistically higher than that for the discordant ones (p < 0.001; Wilcoxon rank sum test). Across five representative genes, the overall sensitivity and specificity were 49.0% and 85.9%, respectively.

Conclusion

Our results indicated that ctDNA could provide complementary information on genetic characterizations in detecting single nucleotide variants (SNVs) and insertions and deletions (InDels).

Inda M, Smid M, van Ooijen H, van de Stolpe A, Martens JWM, Verhaegh WFJ. ER and PI3K Pathway Activity in Primary ER Positive Breast Cancer Is Associated with Progression-Free Survival of Metastatic Patients under First-Line Tamoxifen

: Estrogen receptor positive (ER+) breast cancer patients are eligible for hormonal treatment, but only around half respond. A test with higher specificity for prediction of endocrine therapy response is needed to avoid hormonal overtreatment and to enable selection of alternative treatments. A novel testing method was reported before that enables measurement of functional signal transduction pathway activity in individual cancer tissue samples, using mRNA levels of target genes of the respective pathway-specific transcription factor. Using this method, 130 primary breast cancer samples were analyzed from non-metastatic ER+ patients, treated with surgery without adjuvant hormonal therapy, who subsequently developed metastatic disease that was treated with first-line tamoxifen. Quantitative activity levels were measured of androgen and estrogen receptor (AR and ER), PI3K-FOXO, Hedgehog (HH), NFκB, TGFβ, and Wnt pathways. Based on samples with known pathway activity, thresholds were set to distinguish low from high activity. Subsequently, pathway activity levels were correlated with the tamoxifen treatment response and progression-free survival. High ER pathway activity was measured in 41% of the primary tumors and was associated with longer time to progression (PFS) of metastases during first-line tamoxifen treatment. In contrast, high PI3K, HH, and androgen receptor pathway activity was associated with shorter PFS, and high PI3K and TGFβ pathway activity with worse treatment response. Potential clinical utility of assessment of ER pathway activity lies in predicting response to hormonal therapy, while activity of PI3K, HH, TGFβ, and AR pathways may indicate failure to respond, but also opens new avenues for alternative or complementary targeted treatments.

The therapeutic response of ER+/HER2- breast cancers differs according to the molecular Basal or Luminal subtype

The genomics-based molecular classifications aim at identifying more homogeneous classes than immunohistochemistry, associated with a more uniform clinical outcome. We conducted an in silico analysis on a meta-dataset including gene expression data from 5342 clinically defined ER+/HER2- breast cancers (BC) and DNA copy number/mutational and proteomic data. We show that the Basal (16%) versus Luminal (74%) subtypes as defined using the 80-gene signature differ in terms of response/vulnerability to systemic therapies of BC. The Basal subtype is associated with better chemosensitivity, lesser benefit from adjuvant hormone therapy, and likely better sensitivity to PARP inhibitors, platinum salts and immune therapy, and other targeted therapies under development such as FGFR inhibitors. The Luminal subtype displays potential better sensitivity to CDK4/6 inhibitors and vulnerability to targeted therapies such as PIK3CA, AR and Bcl-2 inhibitors. Expression profiles are very different, showing an intermediate position of the ER+/HER2- Basal subtype between the ER+/HER2- Luminal and ER- Basal subtypes, and let suggest a different cell-of-origin. Our data suggest that the ER+/HER2- Basal and Luminal subtypes should not be assimilated and treated as a homogeneous group.

Addressing cellular heterogeneity in tumor and circulation for refined prognostication

Delineation of intratumor heterogeneity (ITH) has been a subject of growing interest for defining and tracking the evolution of cancer. Yet, the clinical consequences of such ITH on risk prediction remain unclear. Here we show ITH-driven variance on patient stratification and argue that the level of ITH of individual genes should be considered when developing single sector-based prognostic multigene tests (MGTs) in non–small-cell lung cancer (NSCLC). Single-cell molecular analysis of enriched, patient-derived circulating tumor cells (CTCs) further revealed predictive biomarkers for metastatic risk. Through systematic analysis of genes implicated in multiple steps of the metastatic spectrum, we demonstrate that the refined signatures achieve superior accuracy in identifying patients with early-stage disease at high risk of recurrence of NSCLC.

Despite pronounced genomic and transcriptomic heterogeneity in non–small-cell lung cancer (NSCLC) not only between tumors, but also within a tumor, validation of clinically relevant gene signatures for prognostication has relied upon single-tissue samples, including 2 commercially available multigene tests (MGTs). Here we report an unanticipated impact of intratumor heterogeneity (ITH) on risk prediction of recurrence in NSCLC, underscoring the need for a better genomic strategy to refine prognostication. By leveraging label-free, inertial-focusing microfluidic approaches in retrieving circulating tumor cells (CTCs) at single-cell resolution, we further identified specific gene signatures with distinct expression profiles in CTCs from patients with differing metastatic potential. Notably, a refined prognostic risk model that reconciles the level of ITH and CTC-derived gene expression data outperformed the initial classifier in predicting recurrence-free survival (RFS). We propose tailored approaches to providing reliable risk estimates while accounting for ITH-driven variance in NSCLC.

Radiomic analysis for preoperative prediction of cervical lymph node metastasis in patients with papillary thyroid carcinoma

PURPOSE

Cervical lymph node (LN) metastasis of papillary thyroid carcinoma (PTC) is critical for treatment and prognosis. We explored the feasibility of using radiomics to preoperatively predict cervical LN metastasis in PTC patients.

METHOD

Total 221 PTC patients (training cohort: n = 154; validation cohort: n = 67; divided randomly at the ratio of 7:3) were enrolled and divided into 2 groups based on LN pathologic diagnosis (N0: n = 118; N1a and N1b: n = 88 and 15, respectively). We extracted 546 radiomic features from non-contrast and venous contrast-enhanced computed tomography (CT) images. We selected 8 groups of candidate feature sets by minimum redundancy maximum relevance (mRMR), and obtained 8 radiomic sub-signatures by support vector machine (SVM) to construct the radiomic signature. Incorporating the radiomic signature, CT-reported cervical LN status and clinical risk factors, a nomogram was constructed using multivariable logistic regression. The nomogram's calibration, discrimination, and clinical utility were assessed.

RESULTS

The radiomic signature was associated significantly with cervical LN status (p < 0.01 for both training and validation cohorts). The radiomic signature showed better predictive performance than any radiomic sub-signatures devised by SVM. Addition of radiomic signature to the nomogram improved the predictive value (area under the curve (AUC), 0.807 to 0.867) in the training cohort; this was confirmed in an independent validation cohort (AUC, 0.795 to 0.822). Good agreement was observed using calibration curves in both cohorts. Decision curve analysis demonstrated the radiomic nomogram was worthy of clinical application.

CONCLUSIONS

Our radiomic nomogram improved the preoperative prediction of cervical LN metastasis in PTC patients.

Predictive value of single-nucleotide polymorphism signature for recurrence in localised renal cell carcinoma: a retrospective analysis and multicentre validation study

BACKGROUND

Identification of high-risk localised renal cell carcinoma is key for the selection of patients for adjuvant treatment who are at truly higher risk of reccurrence. We developed a classifier based on single-nucleotide polymorphisms (SNPs) to improve the predictive accuracy for renal cell carcinoma recurrence and investigated whether intratumour heterogeneity affected the precision of the classifier.

METHODS

In this retrospective analysis and multicentre validation study, we used paraffin-embedded specimens from the training set of 227 patients from Sun Yat-sen University (Guangzhou, Guangdong, China) with localised clear cell renal cell carcinoma to examine 44 potential recurrence-associated SNPs, which were identified by exploratory bioinformatics analyses of a genome-wide association study from The Cancer Genome Atlas (TCGA) Kidney Renal Clear Cell Carcinoma (KIRC) dataset (n=114, 906 600 SNPs). We developed a six-SNP-based classifier by use of LASSO Cox regression, based on the association between SNP status and patients' recurrence-free survival. Intratumour heterogeneity was investigated from two other regions within the same tumours in the training set. The six-SNP-based classifier was validated in the internal testing set (n=226), the independent validation set (Chinese multicentre study; 428 patients treated between Jan 1, 2004 and Dec 31, 2012, at three hospitals in China), and TCGA set (441 retrospectively identified patients who underwent resection between 1998 and 2010 for localised clear cell renal cell carcinoma in the USA). The main outcome was recurrence-free survival; the secondary outcome was overall survival.

FINDINGS

Although intratumour heterogeneity was found in 48 (23%) of 206 cases in the internal testing set with complete SNP information, the predictive accuracy of the six-SNP-based classifier was similar in the three different regions of the training set (areas under the curve [AUC] at 5 years: 0·749 [95% CI 0·660-0·826] in region 1, 0·734 [0·651-0·814] in region 2, and 0·736 [0·649-0·824] in region 3). The six-SNP-based classifier precisely predicted recurrence-free survival of patients in three validation sets (hazard ratio [HR] 5·32 [95% CI 2·81-10·07] in the internal testing set, 5·39 [3·38-8·59] in the independent validation set, and 4·62 [2·48-8·61] in the TCGA set; all p<0·0001), independently of patient age or sex and tumour stage, grade, or necrosis. The classifier and the clinicopathological risk factors (tumour stage, grade, and necrosis) were combined to construct a nomogram, which had a predictive accuracy significantly higher than that of each variable alone (AUC at 5 years 0·811 [95% CI 0·756-0·861]).

INTERPRETATION

Our six-SNP-based classifier could be a practical and reliable predictor that can complement the existing staging system for prediction of localised renal cell carcinoma recurrence after surgery, which might enable physicians to make more informed treatment decisions about adjuvant therapy. Intratumour heterogeneity does not seem to hamper the accuracy of the six-SNP-based classifier as a reliable predictor of recurrence. The classifier has the potential to guide treatment decisions for patients at differing risks of recurrence.

FUNDING

National Key Research and Development Program of China, National Natural Science Foundation of China, Guangdong Provincial Science and Technology Foundation of China, and Guangzhou Science and Technology Foundation of China.

SALMON: Survival Analysis Learning With Multi-Omics Neural Networks on Breast Cancer

Improved cancer prognosis is a central goal for precision health medicine. Though many models can predict differential survival from data, there is a strong need for sophisticated algorithms that can aggregate and filter relevant predictors from increasingly complex data inputs. In turn, these models should provide deeper insight into which types of data are most relevant to improve prognosis. Deep Learning-based neural networks offer a potential solution for both problems because they are highly flexible and account for data complexity in a non-linear fashion. In this study, we implement Deep Learning-based networks to determine how gene expression data predicts Cox regression survival in breast cancer. We accomplish this through an algorithm called SALMON (Survival Analysis Learning with Multi-Omics Neural Networks), which aggregates and simplifies gene expression data and cancer biomarkers to enable prognosis prediction. The results revealed improved performance when more omics data were used in model construction. Rather than use raw gene expression values as model inputs, we innovatively use eigengene modules from the result of gene co-expression network analysis. The corresponding high impact co-expression modules and other omics data are identified by feature selection technique, then examined by conducting enrichment analysis and exploiting biological functions, escalated the interpretation of input feature from gene level to co-expression modules level. Our study shows the feasibility of discovering breast cancer related co-expression modules, sketch a blueprint of future endeavors on Deep Learning-based survival analysis. SALMON source code is available at https://github.com/huangzhii/SALMON/.

Androgen Receptor mRNA Expression in Urothelial Carcinoma of the Bladder: A Retrospective Analysis of Two Independent Cohorts

INTRODUCTION: Gender-specific differences have led to the androgen receptor (AR) being considered a possible factor in the pathophysiology of urothelial carcinoma of the bladder (UCB), but the exact role remains unclear. MATERIALS AND METHODS: The association of AR mRNA expression with clinicopathological features was retrospectively analyzed in two previously described cohorts. The first cohort consisted of 41 patients with all stages of UCB treated at Aarhus University Hospital, Denmark. The second cohort consisted of 323 patients with muscle-invasive bladder cancer (MIBC) accumulated by the Cancer Genome Atlas (TCGA) Research Network. RESULTS: AR mRNA expression is significantly higher in non-muscle-invasive bladder cancer (NMIBC) when compared to MIBC (P = .0004), with no relevant changes within the different stages of MIBC. AR mRNA expression was significantly associated with TCGA molecular subtypes (P < .0001). In the total cohort, there was no association between AR expression and gender (P = .23). When analyzed separately, females showed a significantly worse disease-free (P = .03) and overall survival (P = .02) when expressing AR mRNA above median level, while the same was not observed for men. Multivariable Cox's regression analyses revealed AR mRNA expression to be an independent prognostic marker for disease-free survival in women (P = .007). CONCLUSIONS: AR mRNA expression is significantly higher in NMIBC than in MIBC, while high AR mRNA expression is associated with worse survival in females with MIBC. Further studies need to investigate the gender-specific role of AR in UCB.

Analysis of the prognostic relevance of sex-steroid hormonal receptor mRNA expression in muscle-invasive urothelial carcinoma of the urinary bladder

Muscle-invasive urothelial carcinoma of the urinary bladder (UCB) often recurs following radical cystectomy (RC). An altered expression of sex-steroid hormone receptors has been associated with oncological outcomes of UCB and may represent therapeutic targets. Here the expression of different hormone receptors was measured on mRNA levels in patients treated by RC and associated with outcomes. Androgen receptor (AR), estrogen receptor 1 (ESR1), and progesterone receptor (PGR) mRNA expression was assessed by quantitative reverse transcription polymerase chain reaction (RT-qPCR) in RC samples of 87 patients with a median age of 66 (39-88) years. Univariate and multivariate analyses were performed to test associations with pathological and clinical characteristics as well as recurrence-free (RFS) and disease-specific survival (DSS). AR mRNA expression was lower in comparison with ESR1 and PGR expression (p < 0.0001). In univariate analysis, high expression levels of AR were associated with reduced RFS (HR 2.8, p = 0.015) and DSS (HR 2.8, p = 0.010). High AR mRNA expression and a positive lymph node status were independent predictors for reduced RFS (HR 2.5, p = 0.0049) and DSS (HR 3.4, p = 0.009). In patients with low AR mRNA expression, an increased ESR1 and PGR mRNA expression were associated with reduced RFS and DSS. High expression levels of AR are significantly associated with adverse outcome in patients with muscle-invasive UCB following RC. ESR1 and PGR expression status can further stratify patients with low AR expression into subgroups with significantly reduced RFS and DSS. Therapeutic targeting of AR may influence outcomes in patients with UCB.

Intra-individual Gene Expression Variability of Histologically Normal Breast Tissue

Several studies have sought to identify novel transcriptional biomarkers in normal breast or breast microenvironment to predict tumor risk and prognosis. However, systematic efforts to evaluate intra-individual variability of gene expression within normal breast have not been reported. This study analyzed the microarray gene expression data of 288 samples from 170 women in the Normal Breast Study (NBS), wherein multiple histologically normal breast samples were collected from different block regions and different sections at a given region. Intra-individual differences in global gene expression and selected gene expression signatures were quantified and evaluated in association with other patient-level factors. We found that intra-individual reliability was relatively high in global gene expression, but differed by signatures, with composition-related signatures (i.e., stroma) having higher intra-individual variability and tumorigenesis-related signatures (i.e., proliferation) having lower intra-individual variability. Histological stroma composition was the only factor significantly associated with heterogeneous breast tissue (defined as > median intra-individual variation; high nuclear density, odds ratio [OR] = 3.42, 95% confidence interval [CI] = 1.15–10.15; low area, OR = 0.29, 95% CI = 0.10–0.86). Other factors suggestively influencing the variability included age, BMI, and adipose nuclear density. Our results underscore the importance of considering intra-individual variability in tissue-based biomarker development, and have important implications for normal breast research.

Limited utility of tissue micro-arrays in detecting intra-tumoral heterogeneity in stem cell characteristics and tumor progression markers in breast cancer

Background

Intra-tumoral heterogeneity has been recently addressed in different types of cancer, including breast cancer. A concept describing the origin of intra-tumoral heterogeneity is the cancer stem-cell hypothesis, proposing the existence of cancer stem cells that can self-renew limitlessly and therefore lead to tumor progression. Clonal evolution in accumulated single cell genomic alterations is a further possible explanation in carcinogenesis. In this study, we addressed the question whether intra-tumoral heterogeneity can be reliably detected in tissue-micro-arrays in breast cancer by comparing expression levels of conventional predictive/prognostic tumor markers, tumor progression markers and stem cell markers between central and peripheral tumor areas.

Methods

We analyzed immunohistochemical expression and/or gene amplification status of conventional prognostic tumor markers (ER, PR, HER2, CK5/6), tumor progression markers (PTEN, PIK3CA, p53, Ki-67) and stem cell markers (mTOR, SOX2, SOX9, SOX10, SLUG, CD44, CD24, TWIST) in 372 tissue-micro-array samples from 72 breast cancer patients. Expression levels were compared between central and peripheral tumor tissue areas and were correlated to histopathological grading. 15 selected cases additionally underwent RNA sequencing for transcriptome analysis.

Results

No significant difference in any of the analyzed between central and peripheral tumor areas was seen with any of the analyzed methods/or results that showed difference. Except mTOR, PIK3CA and SOX9 (nuclear) protein expression, all markers correlated significantly (p < 0.05) with histopathological grading both in central and peripheral areas.

Conclusion

Our results suggest that intra-tumoral heterogeneity of stem-cell and tumor-progression markers cannot be reliably addressed in tissue-micro-array samples in breast cancer. However, most markers correlated strongly with histopathological grading confirming prognostic information as expression profiles were independent on the site of the biopsy was taken.

Electronic supplementary material

The online version of this article (10.1186/s12967-018-1495-6) contains supplementary material, which is available to authorized users.

A New Approach to Predict Progression-free Survival in Stage IV EGFR-mutant NSCLC Patients with EGFR-TKI Therapy

Purpose: We established a CT-derived approach to achieve accurate progression-free survival (PFS) prediction to EGFR tyrosine kinase inhibitors (TKI) therapy in multicenter, stage IV EGFR-mutated non-small cell lung cancer (NSCLC) patients.Experimental Design: A total of 1,032 CT-based phenotypic characteristics were extracted according to the intensity, shape, and texture of NSCLC pretherapy images. On the basis of these CT features extracted from 117 stage IV EGFR-mutant NSCLC patients, a CT-based phenotypic signature was proposed using a Cox regression model with LASSO penalty for the survival risk stratification of EGFR-TKI therapy. The signature was validated using two independent cohorts (101 and 96 patients, respectively). The benefit of EGFR-TKIs in stratified patients was then compared with another stage-IV EGFR-mutant NSCLC cohort only treated with standard chemotherapy (56 patients). Furthermore, an individualized prediction model incorporating the phenotypic signature and clinicopathologic risk characteristics was proposed for PFS prediction, and also validated by multicenter cohorts.Results: The signature consisted of 12 CT features demonstrated good accuracy for discriminating patients with rapid and slow progression to EGFR-TKI therapy in three cohorts (HR: 3.61, 3.77, and 3.67, respectively). Rapid progression patients received EGFR TKIs did not show significant difference with patients underwent chemotherapy for progression-free survival benefit (P = 0.682). Decision curve analysis revealed that the proposed model significantly improved the clinical benefit compared with the clinicopathologic-based characteristics model (P < 0.0001).Conclusions: The proposed CT-based predictive strategy can achieve individualized prediction of PFS probability to EGFR-TKI therapy in NSCLCs, which holds promise of improving the pretherapy personalized management of TKIs. Clin Cancer Res; 24(15); 3583-92. ©2018 AACR.

A Novel Statistical Method to Diagnose, Quantify and Correct Batch Effects in Genomic Studies

Genome projects now generate large-scale data often produced at various time points by different laboratories using multiple platforms. This increases the potential for batch effects. Currently there are several batch evaluation methods like principal component analysis (PCA; mostly based on visual inspection), and sometimes they fail to reveal all of the underlying batch effects. These methods can also lead to the risk of unintentionally correcting biologically interesting factors attributed to batch effects. Here we propose a novel statistical method, finding batch effect (findBATCH), to evaluate batch effect based on probabilistic principal component and covariates analysis (PPCCA). The same framework also provides a new approach to batch correction, correcting batch effect (correctBATCH), which we have shown to be a better approach to traditional PCA-based correction. We demonstrate the utility of these methods using two different examples (breast and colorectal cancers) by merging gene expression data from different studies after diagnosing and correcting for batch effects and retaining the biological effects. These methods, along with conventional visual inspection-based PCA, are available as a part of an R package exploring batch effect (exploBATCH; https://github.com/syspremed/exploBATCH).

A preliminary study of 18F-FES PET/CT in predicting metastatic breast cancer in patients receiving docetaxel or fulvestrant with docetaxel

The present explorative study was initiated to evaluate the clinical value of ¹⁸F-FES PET/CT in monitoring the change of estrogen receptor (ER) expression and potential predictive value in metastatic breast cancer patients. Twenty-two pathology-confirmed breast cancer patients were prospectively enrolled and randomly divided into two groups (T: docetaxel, n = 14 and TF: docetaxel + fulvestrant, n = 8). The percentage of patients without disease progression after 12 months (PFS > 12 months) was 62.5% in group TF compared with 21.4% in group T (P = 0.08). According to ¹⁸F-FES PET/CT scans, the SUVmax (maximum standard uptake value) of all the metastatic lesions decreased in group TF after 2 cycles of treatment (6 weeks ± 3 days). However, 6 of 9 patients in group T had at least one lesion with higher post-treatment SUVmax. There was a significant difference in the reduction of ER expression between these two groups (P = 0.028). In group TF, the patients with PFS > 12 months had significantly greater SUVmax changes of ¹⁸F-FES than those with PFS < 12 months (PFS > 12 months: 91.0 ± 12.0% versus PFS < 12 months: 20.7 ± 16.2%; t = −4.64, P = 0.01). Our preliminary study showed that ¹⁸F-FES PET/CT, as a noninvasive method to monitor ER expression, could be utilized to predict prognosis based on changes in SUVmax.

Heterogeneous Enhancement Patterns of Tumor-adjacent Parenchyma at MR Imaging Are Associated with Dysregulated Signaling Pathways and Poor Survival in Breast Cancer

Purpose To identify the molecular basis of quantitative imaging characteristics of tumor-adjacent parenchyma at dynamic contrast material-enhanced magnetic resonance (MR) imaging and to evaluate their prognostic value in breast cancer. Materials and Methods In this institutional review board-approved, HIPAA-compliant study, 10 quantitative imaging features depicting tumor-adjacent parenchymal enhancement patterns were extracted and screened for prognostic features in a discovery cohort of 60 patients. By using data from The Cancer Genome Atlas (TCGA), a radiogenomic map for the tumor-adjacent parenchymal tissue was created and molecular pathways associated with prognostic parenchymal imaging features were identified. Furthermore, a multigene signature of the parenchymal imaging feature was built in a training cohort (n = 126), and its prognostic relevance was evaluated in two independent cohorts (n = 879 and 159). Results One image feature measuring heterogeneity (ie, information measure of correlation) was significantly associated with prognosis (false-discovery rate < 0.1), and at a cutoff of 0.57 stratified patients into two groups with different recurrence-free survival rates (log-rank P = .024). The tumor necrosis factor signaling pathway was identified as the top enriched pathway (hypergeometric P < .0001) among genes associated with the image feature. A 73-gene signature based on the tumor profiles in TCGA achieved good association with the tumor-adjacent parenchymal image feature (R² = 0.873), which stratified patients into groups regarding recurrence-free survival (log-rank P = .029) and overall survival (log-rank P = .042) in an independent TCGA cohort. The prognostic value was confirmed in another independent cohort (Gene Expression Omnibus GSE 1456), with log-rank P = .00058 for recurrence-free survival and log-rank P = .0026 for overall survival. Conclusion Heterogeneous enhancement patterns of tumor-adjacent parenchyma at MR imaging are associated with the tumor necrosis signaling pathway and poor survival in breast cancer. ^© RSNA, 2017 Online supplemental material is available for this article.

Unsupervised Clustering of Quantitative Image Phenotypes Reveals Breast Cancer Subtypes with Distinct Prognoses and Molecular Pathways

Purpose: To identify novel breast cancer subtypes by extracting quantitative imaging phenotypes of the tumor and surrounding parenchyma and to elucidate the underlying biologic underpinnings and evaluate the prognostic capacity for predicting recurrence-free survival (RFS).Experimental Design: We retrospectively analyzed dynamic contrast-enhanced MRI data of patients from a single-center discovery cohort (n = 60) and an independent multicenter validation cohort (n = 96). Quantitative image features were extracted to characterize tumor morphology, intratumor heterogeneity of contrast agent wash-in/wash-out patterns, and tumor-surrounding parenchyma enhancement. On the basis of these image features, we used unsupervised consensus clustering to identify robust imaging subtypes and evaluated their clinical and biologic relevance. We built a gene expression-based classifier of imaging subtypes and tested their prognostic significance in five additional cohorts with publically available gene expression data but without imaging data (n = 1,160).Results: Three distinct imaging subtypes, that is, homogeneous intratumoral enhancing, minimal parenchymal enhancing, and prominent parenchymal enhancing, were identified and validated. In the discovery cohort, imaging subtypes stratified patients with significantly different 5-year RFS rates of 79.6%, 65.2%, 52.5% (log-rank P = 0.025) and remained as an independent predictor after adjusting for clinicopathologic factors (HR, 2.79; P = 0.016). The prognostic value of imaging subtypes was further validated in five independent gene expression cohorts, with average 5-year RFS rates of 88.1%, 74.0%, 59.5% (log-rank P from <0.0001 to 0.008). Each imaging subtype was associated with specific dysregulated molecular pathways that can be therapeutically targeted.Conclusions: Imaging subtypes provide complimentary value to established histopathologic or molecular subtypes and may help stratify patients with breast cancer. Clin Cancer Res; 23(13); 3334-42. ©2017 AACR.

Application of single-cell technology in cancer research

In this review, we have outlined the application of single-cell technology in cancer research. Single-cell technology has made encouraging progress in recent years and now provides the means to detect rare cancer cells such as circulating tumor cells and cancer stem cells. We reveal how this technology has advanced the analysis of intratumor heterogeneity and tumor epigenetics, and guided individualized treatment strategies. The future prospects now are to bring single-cell technology into the clinical arena. We believe that the clinical application of single-cell technology will be beneficial in cancer diagnostics and treatment, and ultimately improve survival in cancer patients.

Gene expression profiles among murine strains segregate with distinct differences in the progression of radiation-induced lung disease

Molecular mechanisms underlying development of acute pneumonitis and/or late fibrosis following thoracic irradiation remain poorly understood. Here, we hypothesize that heterogeneity in disease progression and phenotypic expression of radiation-induced lung disease (RILD) across murine strains presents an opportunity to better elucidate mechanisms driving tissue response toward pneumonitis and/or fibrosis. Distinct differences in disease progression were observed in age- and sex-matched CBA/J, C57L/J and C57BL/6J mice over 1 year after graded doses of whole-thorax lung irradiation (WTLI). Separately, comparison of gene expression profiles in lung tissue 24 h post-exposure demonstrated >5000 genes to be differentially expressed (P<0.01; >twofold change) between strains with early versus late onset of disease. An immediate divergence in early tissue response between radiation-sensitive and -resistant strains was observed. In pneumonitis-prone C57L/J mice, differentially expressed genes were enriched in proinflammatory pathways, whereas in fibrosis-prone C57BL/6J mice, genes were enriched in pathways involved in purine and pyrimidine synthesis, DNA replication and cell division. At 24 h post-WTLI, different patterns of cellular damage were observed at the ultrastructural level among strains but microscopic damage was not yet evident under light microscopy. These data point toward a fundamental difference in patterns of early pulmonary tissue response to WTLI, consistent with the macroscopic expression of injury manifesting weeks to months after exposure. Understanding the mechanisms underlying development of RILD might lead to more rational selection of therapeutic interventions to mitigate healthy tissue damage.

Summary: Rational mouse model selection is crucial for identifying new therapeutic targets and screening medical interventions in acute pneumonitis and/or late fibrosis following thoracic irradiation.

Pathophysiological mechanisms underlying phenotypic differences in pulmonary radioresponse

Differences in the pathogenesis of radiation-induced lung injury among murine strains offer a unique opportunity to elucidate the molecular mechanisms driving the divergence in tissue response from repair and recovery to organ failure. Here, we utilized two well-characterized murine models of radiation pneumonitis/fibrosis to compare and contrast differential gene expression in lungs 24 hours after exposure to a single dose of whole thorax lung irradiation sufficient to cause minor to major morbidity/mortality. Expression of 805 genes was altered as a general response to radiation; 42 genes were identified whose expression corresponded to the threshold for lethality. Three genes were discovered whose expression was altered within the lethal, but not the sublethal, dose range. Time-course analysis of the protein product of the most promising gene, resistin-like molecule alpha, demonstrated a significant difference in expression between radiosensitive versus radiotolerant strains, suggesting a unique role for this protein in acute lung injury.

Genomic Adequacy from Solid Tumor Core Needle Biopsies of ex Vivo Tissue and in Vivo Lung Masses: Prospective Study

Purpose To identify the variables and factors that affect the quantity and quality of nucleic acid yields from imaging-guided core needle biopsy. Materials and Methods This study was approved by the institutional review board and compliant with HIPAA. The authors prospectively obtained 232 biopsy specimens from 74 patients (177 ex vivo biopsy samples from surgically resected masses were obtained from 49 patients and 55 in vivo lung biopsy samples from computed tomographic [CT]-guided lung biopsies were obtained from 25 patients) and quantitatively measured DNA and RNA yields with respect to needle gauge, number of needle passes, and percentage of the needle core. RNA quality was also assessed. Significance of correlations among variables was assessed with analysis of variance followed by linear regression. Conditional probabilities were calculated for projected sample yields. Results The total nucleic acid yield increased with an increase in the number of needle passes or a decrease in needle gauge (two-way analysis of variance, P < .0001 for both). However, contrary to calculated differences in volume yields, the effect of needle gauge was markedly greater than the number of passes. For example, the use of an 18-gauge versus a 20-gauge biopsy needle resulted in a 4.8-5.7 times greater yield, whereas a double versus a single pass resulted in a 2.4-2.8 times greater yield for 18- versus 20-gauge needles, respectively. Ninety-eight of 184 samples (53%) had an RNA integrity number of at least 7 (out of a possible score of 10). Conclusion With regard to optimizing nucleic acid yields in CT-guided lung core needle biopsies used for genomic analysis, there should be a preference for using lower gauge needles over higher gauge needles with more passes. ^©RSNA, 2016 Online supplemental material is available for this article. An earlier incorrect version of this article appeared online. This article was corrected on October 21, 2016.

Multimodal Assessment of Estrogen Receptor mRNA Profiles to Quantify Estrogen Pathway Activity in Breast Tumors

BACKGROUND

Molecular markers have transformed our understanding of the heterogeneity of breast cancer and have allowed the identification of genomic profiles of estrogen receptor (ER)-α signaling. However, our understanding of the transcriptional profiles of ER signaling remains inadequate. Therefore, we sought to identify the genomic indicators of ER pathway activity that could supplement traditional immunohistochemical (IHC) assessments of ER status to better understand ER signaling in the breast tumors of individual patients.

MATERIALS AND METHODS

We reduced ESR1 (gene encoding the ER-α protein) mRNA levels using small interfering RNA in ER⁺ MCF7 breast cancer cells and assayed for transcriptional changes using Affymetrix HG U133 Plus 2.0 arrays. We also compared 1034 ER⁺ and ER^- breast tumors from publicly available microarray data. The principal components of ER activity generated from these analyses and from other published estrogen signatures were compared with ESR1 expression, ER-α IHC, and patient survival.

RESULTS

Genes differentially expressed in both analyses were associated with ER-α IHC and ESR1 mRNA expression. They were also significantly enriched for estrogen-driven molecular pathways associated with ESR1, cyclin D1 (CCND1), MYC (v-myc avian myelocytomatosis viral oncogene homolog), and NFKB (nuclear factor kappa B). Despite their differing constituent genes, the principal components generated from these new analyses and from previously published ER-associated gene lists were all associated with each other and with the survival of patients with breast cancer treated with endocrine therapies.

CONCLUSION

A biomarker of ER-α pathway activity, generated using ESR1-responsive mRNAs in MCF7 cells, when used alongside ER-α IHC and ESR1 mRNA expression, could provide a method for further stratification of patients and add insight into ER pathway activity in these patients.

Attempt towards a novel classification of triple-negative breast cancer using immunohistochemical markers

Significant efforts have been made to gain a better understanding of the heterogeneity of triple-negative breast cancers from the histological to the molecular and genomic levels. In this study, we attempted to bring forward gene expression subtypes of triple-negative breast cancer (TBNC) to the clinic, by translating gene stratification to clinically accessible immunohistochemical (IHC) classification. Using IHC analysis, we categorized 154 TBNC cases into three main subclasses. Differences in the frequencies of basic characteristics and clinicopathological parameters between the subtypes were examined using Chi-square tests. We defined three main groups among the 154 triple-negative cases. The basal-like (BL) group expressed cytokeratin (CK) 5/6 and/or CK14 (83 cases), the AR⁺ group demonstrated positivity for androgen receptor (18 cases), and the final group exhibited a CD44+CD24-^/low phenotype (39 cases). There were three overlapping cases between the BL subgroup and the CD44+CD24-/low phenotype subgroup, and 11 unclassified cases. In this new IHC classification, three subcategories exhibited a statistical difference with regard to age, tumor size, histological grade, tumor necrosis, Ki67 labeling index, relapse-free survival, breast cancer-specific survival and response to chemotherapy. According to our definition, the BL group and CD44+CD24-^/low phenotype could be observed in tumors that were not triple-negative, and BL tumors that were triple-negative demonstrated almost undistinguishable clinicopathological characteristics compared with BL tumors that were not triple-negative. The same observation was made with CD44+CD24-/low tumors that were triple-negative vs. CD44+CD24-^/low tumors that were not. The AR+ group demonstrated undistinguishable clinicopathological characteristics compared with the luminal subtype. We successfully distinguished three subtypes exhibiting diverse clinicopathological and prognostic characteristics with the minimum use of IHC markers.

Intratumor Heterogeneity Affects Gene Expression Profile Test Prognostic Risk Stratification in Early Breast Cancer

PURPOSE

To examine the effect of intratumor heterogeneity (ITH) on detection of genes within gene expression panels (GEPs) and the subsequent ability to predict prognostic risk.

EXPERIMENTAL DESIGN

Multiplexed barcoded RNA analysis was used to measure the expression of 141 genes from five GEPs (Oncotype Dx, MammaPrint, PAM50, EndoPredict, and Breast Cancer Index) in breast cancer tissue sections and tumor-rich cores from 71 estrogen receptor (ER)-positive node-negative tumors, on which clinical Oncotype Dx testing was previously performed. If the tumor had foci of high Ki67 (n = 26), low/negative progesterone receptor (PR; n = 13), or both (n = 5), additional cores were obtained. In total, 181 samples were processed. Oncotype Dx recurrence scores were calculated from NanoString nCounter gene expression data.

RESULTS

Hierarchical clustering using all GEP genes showed that majority (61 of 71) of tumor samples clustered by patient, indicating greater interpatient heterogeneity (IPH) than ITH. We found a strikingly high correlation between Oncotype Dx recurrence scores obtained from whole sections versus tumor-rich cores (r = 0.94). However, high Ki67 and low PR cores had slightly higher but not statistically significant recurrence scores. For 18 of 71 (25%) patients, scores were divergent between sections and cores and crossed the boundaries for low, intermediate, and high risk.

CONCLUSIONS

Our study indicates that in patients with highly heterogeneous tumors, GEP recurrence scores from a single core could under- or overestimate prognostic risk. Hence, it may be a useful strategy to assess multiple samples (both representative and atypical cores) to fully account for the ITH-driven variation in risk prediction. Clin Cancer Res; 22(21); 5362-9. ©2016 AACR.

"RADIOTRANSCRIPTOMICS": A synergy of imaging and transcriptomics in clinical assessment

Recent advances in quantitative imaging and "omics" technology have generated a wealth of mineable biological "big data". With the push towards a P4 "predictive, preventive, personalized, and participatory" approach to medicine, researchers began integrating complementary tools to further tune existing diagnostic and therapeutic models. The field of radiogenomics has long pioneered such multidisciplinary investigations in neuroscience and oncology, correlating genotypic and phenotypic signatures to study structural and functional changes in relation to altered molecular behavior. Given the innate dynamic nature of complex disorders and the role of environmental and epigenetic factors in pathogenesis, the transcriptome can further elucidate serial modifications undetected at the genome level. We therefore propose "radiotranscriptomics" as a new member of the P4 medicine initiative, combining transcriptome information, including gene expression and isoform variation, and quantitative image annotations.

Low Concordance between Gene Expression Signatures in ER Positive HER2 Negative Breast Carcinoma Could Impair Their Clinical Application

BACKGROUND

Numerous prognostic gene expression signatures have been recently described. Among the signatures there is variation in the constituent genes that are utilized. We aim to evaluate prognostic concordance among eight gene expression signatures, on a large dataset of ER positive HER2 negative breast cancers.

METHODS

We analysed the performance of eight gene expression signatures on six different datasets of ER+ HER2- breast cancers. Survival analyses were performed using the Kaplan-Meier estimate of survival function. We assessed discrimination and concordance between the 8 signatures on survival and recurrence rates The Nottingham Prognostic Index (NPI) was used to to stratify the risk of recurrence/death.

RESULTS

The discrimination ability of the whole signatures, showed fair discrimination performances, with AUC ranging from 0.64 (95%CI 0.55-0.73 for the 76-genes signatures, to 0.72 (95%CI 0.64-0.8) for the Molecular Prognosis Index T17. Low concordance was found in predicting events in the intermediate and high-risk group, as defined by the NPI. Low risk group was the only subgroup with a good signatures concordance.

CONCLUSION

Genomic signatures may be a good option to predict prognosis as most of them perform well at the population level. They exhibit, however, a high degree of discordance in the intermediate and high-risk groups. The major benefit that we could expect from gene expression signatures is the standardization of proliferation assessment.

Aberrant Splicing of Estrogen Receptor, HER2, and CD44 Genes in Breast Cancer

Breast cancer (BC) is the most common cause of cancer-related death among women under the age of 50 years. Established biomarkers, such as hormone receptors (estrogen receptor [ER]/progesterone receptor) and human epidermal growth factor receptor 2 (HER2), play significant roles in the selection of patients for endocrine and trastuzumab therapies. However, the initial treatment response is often followed by tumor relapse with intrinsic resistance to the first-line therapy, so it has been expected to identify novel molecular markers to improve the survival and quality of life of patients. Alternative splicing of pre-messenger RNAs is a ubiquitous and flexible mechanism for the control of gene expression in mammalian cells. It provides cells with the opportunity to create protein isoforms with different, even opposing, functions from a single genomic locus. Aberrant alternative splicing is very common in cancer where emerging tumor cells take advantage of this flexibility to produce proteins that promote cell growth and survival. While a number of splicing alterations have been reported in human cancers, we focus on aberrant splicing of ER, HER2, and CD44 genes from the viewpoint of BC development. ERα36, a splice variant from the ER1 locus, governs nongenomic membrane signaling pathways triggered by estrogen and confers 4-hydroxytamoxifen resistance in BC therapy. The alternative spliced isoform of HER2 lacking exon 20 (Δ16HER2) has been reported in human BC; this isoform is associated with transforming ability than the wild-type HER2 and recapitulates the phenotypes of endocrine therapy-resistant BC. Although both CD44 splice isoforms (CD44s, CD44v) play essential roles in BC development, CD44v is more associated with those with favorable prognosis, such as luminal A subtype, while CD44s is linked to those with poor prognosis, such as HER2 or basal cell subtypes that are often metastatic. Hence, the detection of splice variants from these loci will provide keys to understand the pathogenesis, predict the prognosis, and choose specific therapies for BC.

A CpG-methylation-based assay to predict survival in clear cell renal cell carcinoma

Clear cell renal cell carcinomas (ccRCCs) display divergent clinical behaviours. Molecular markers might improve risk stratification of ccRCC. Here we use, based on genome-wide CpG methylation profiling, a LASSO model to develop a five-CpG-based assay for ccRCC prognosis that can be used with formalin-fixed paraffin-embedded specimens. The five-CpG-based classifier was validated in three independent sets from China, United States and the Cancer Genome Atlas data set. The classifier predicts the overall survival of ccRCC patients (hazard ratio=2.96−4.82; P=3.9 × 10⁻⁶−2.2 × 10⁻⁹), independent of standard clinical prognostic factors. The five-CpG-based classifier successfully categorizes patients into high-risk and low-risk groups, with significant differences of clinical outcome in respective clinical stages and individual ‘stage, size, grade and necrosis' scores. Moreover, methylation at the five CpGs correlates with expression of five genes: PITX1, FOXE3, TWF2, EHBP1L1 and RIN1. Our five-CpG-based classifier is a practical and reliable prognostic tool for ccRCC that can add prognostic value to the staging system.

Using molecular markers is a useful way to predict the prognosis of cancer patients. Here, Wei et al. describe a five gene methylation signature that can predict the prognosis of renal clear cell cancer and validate its use in multiple patient cohorts.

Receptor status change from primary to residual breast cancer after neoadjuvant chemotherapy and analysis of survival outcomes

BACKGROUND

To evaluate the frequency of receptor change from pretreatment to residual breast cancer after NCT and their correlation with outcomes.

PATIENTS AND METHODS

Three hundred ninety-eight women were identified retrospectively. Estrogen receptor, progesterone receptor, and HER2 were reviewed. Patients were classified as not having receptor change versus any receptor change. Kaplan-Meier was used to estimate survival outcomes according to changes. Cox proportional hazards models were used to determine the association of receptor status changes with outcomes after adjustment for patient and tumor characteristics.

RESULTS

One hundred sixty-two (40.7%) patients had a change in at least 1 of the receptors from pretreatment to residual disease. Patients who had no change in receptor status had a significantly greater triple-negative breast cancer (TNBC) rate at baseline (P = .0001). Of the 193 hormone receptor (HR)-positive tumors, 9 (4.7%) and 29 (15.1%) became HER2-positive and TNBC, respectively. Of the 72 HER2-positive tumors, 20 (27.8%) and 9 (12.5%) became HR-positive and TNBC, respectively. Of the 128 TNBC tumors, only 2 (1.6%) and 33 (25.8%) became HER2-positive and HR-positive, respectively. At a median follow up of 40 months, 5-year overall survival (OS) was 73% and 63%; and 5-year relapse-free survival (RFS) was 63% and 48% for patients with or without any receptor change (P = .07 and P = .003), respectively. Any receptor change was associated with better RFS (hazard ratio, 0.63; 95% confidence interval [CI], 0.44-0.9) but not OS. (hazard ratio, 0.79; 95% CI, 0.53-1.18).

CONCLUSION

Changes in receptor status between the pretreatment and residual disease after NCT are frequent and appear to be associated with improved RFS because of the receptor stability of TNBC.

Claudin-low breast cancers: clinical, pathological, molecular and prognostic characterization

BACKGROUND

The lastly identified claudin-low (CL) subtype of breast cancer (BC) remains poorly described as compared to the other molecular subtypes. We provide a comprehensive characterization of the largest series of CL samples reported so far.

METHODS

From a data set of 5447 invasive BC profiled using DNA microarrays, we identified 673 CL samples (12,4%) that we describe comparatively to the other molecular subtypes at several levels: clinicopathological, genomic, transcriptional, survival, and response to chemotherapy.

RESULTS

CL samples display profiles different from other subtypes. For example, they differ from basal tumors regarding the hormone receptor status, with a lower frequency of triple negative (TN) tumors (52% vs 76% for basal cases). Like basal tumors, they show high genomic instability with many gains and losses. At the transcriptional level, CL tumors are the most undifferentiated tumors along the mammary epithelial hierarchy. Compared to basal tumors, they show enrichment for epithelial-to-mesenchymal transition markers, immune response genes, and cancer stem cell-like features, and higher activity of estrogen receptor (ER), progesterone receptor (PR), EGFR, SRC and TGFβ pathways, but lower activity of MYC and PI3K pathways. The 5-year disease-free survival of CL cases (67%) and the rate of pathological complete response (pCR) to primary chemotherapy (32%) are close to those of poor-prognosis and good responder subtypes (basal and ERBB2-enriched). However, the prognostic features of CL tumors are closer to those observed in the whole BC series and in the luminal A subtype, including proliferation-related gene expression signatures (GES). Immunity-related GES valuable in basal breast cancers are not significant in CL tumors. By contrast, the GES predictive for pCR in CL tumors resemble more to those of basal and HER2-enriched tumors than to those of luminal A tumors.

CONCLUSIONS

Many differences exist between CL and the other subtypes, notably basal. An unexpected finding concerns the relatively high numbers of ER-positive and non-TN tumors within CL subtype, suggesting a larger heterogeneity than in basal and luminal A subtypes.