Deep learning trained on hematoxylin and eosin tumor region of Interest predicts HER2 status and trastuzumab treatment response in HER2+ breast cancer

Farahmand, Saman; Fernandez, Aileen I.; Ahmed, Fahad; Rimm, David L.; Chuang, Jeffrey H.; Reisenbichler, Emily; Zarringhalam, Kourosh

doi:10.1038/s41379-021-00911-w

Cited by 96 publications

(71 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Alternative methodologies (e.g., RT-PCR, digital pathology) have already been proposed for the detection of this subset of patients ( Jiang et al, 2016 ). Among these, machine learning-based predictors showed significant results in terms of speed, accuracy, and cost-effectiveness of predicting both HER2 status and anti-HER2 treatment response ( Fusco et al, 2013 ; La Barbera et al, 2020 ; Yousif et al, 2021 ; Farahmand et al, 2022 ). However, the IHC-ISH combined test remains the gold standard.…”

Section: Rebooting Her2 Testing In Breast Cancermentioning

confidence: 99%

HER2 Low, Ultra-low, and Novel Complementary Biomarkers: Expanding the Spectrum of HER2 Positivity in Breast Cancer

Venetis

Crimini

Sajjadi

et al. 2022

Front. Mol. Biosci.

View full text Add to dashboard Cite

HER2 status in breast cancer is assessed to select patients eligible for targeted therapy with anti-HER2 therapies. According to the American Society of Clinical Oncology (ASCO) and College of American Pathologists (CAP), the HER2 test positivity is defined by protein overexpression (score 3+) at immunohistochemistry (IHC) and/or gene amplification at in situ hybridization (ISH). The introduction of novel anti-HER2 compounds, however, is changing this paradigm because some breast cancers with lower levels of protein expression (i.e. score 1+/2+ with no gene amplification) benefited from HER2 antibody-drug conjugates (ADC). Recently, a potential for HER2 targeting in HER2 “ultra-low” (i.e. score 0 with incomplete and faint staining in ≤10% of tumor cells) and MutL-deficient estrogen receptor (estrogen receptor)-positive/HER2-negative breast cancers has been highlighted. All these novel findings are transforming the traditional dichotomy of HER2 status and have dramatically raised the expectations in this field. Still, a more aware HER2 status assessment coupled with the comprehensive characterization of the clinical and molecular features of these tumors is required. Here, we seek to provide an overview of the current state of HER2 targeting in breast cancers beyond the canonical HER2 positivity and to discuss the practical implications for pathologists and oncologists.

show abstract

Section: Rebooting Her2 Testing In Breast Cancermentioning

confidence: 99%

HER2 Low, Ultra-low, and Novel Complementary Biomarkers: Expanding the Spectrum of HER2 Positivity in Breast Cancer

Venetis

Crimini

Sajjadi

et al. 2022

Front. Mol. Biosci.

View full text Add to dashboard Cite

show abstract

“…Radiomics has successfully adapted to diverse immune phenotypes. Radiomic traits extracted from CT imaging of patients with solid tumors ( 117 ), H&E images of melanoma and lung cancer ( 149 ), and WSIs of patients with breast cancer ( 152 ) can determine the response to anti-PD-1/PD-L1 and anti-PD-1 and trastuzumab, respectively.…”

Section: Discussionmentioning

confidence: 99%

“…Farahmand et al. ( 152 ) developed an H&E-based deep learning algorithm to determine human EGFR 2 statuses and trastuzumab treatment response in patients with breast cancer with an AUC of 0.81 and 0.80, respectively, independent of the TCGA dataset. They demonstrated power classification within the level of interobserver variability.…”

Section: Some Case Studies and Applicationsmentioning

confidence: 99%

Deep Learning With Radiomics for Disease Diagnosis and Treatment: Challenges and Potential

Zhang

et al. 2022

Front. Oncol.

View full text Add to dashboard Cite

The high-throughput extraction of quantitative imaging features from medical images for the purpose of radiomic analysis, i.e., radiomics in a broad sense, is a rapidly developing and emerging research field that has been attracting increasing interest, particularly in multimodality and multi-omics studies. In this context, the quantitative analysis of multidimensional data plays an essential role in assessing the spatio-temporal characteristics of different tissues and organs and their microenvironment. Herein, recent developments in this method, including manually defined features, data acquisition and preprocessing, lesion segmentation, feature extraction, feature selection and dimension reduction, statistical analysis, and model construction, are reviewed. In addition, deep learning-based techniques for automatic segmentation and radiomic analysis are being analyzed to address limitations such as rigorous workflow, manual/semi-automatic lesion annotation, and inadequate feature criteria, and multicenter validation. Furthermore, a summary of the current state-of-the-art applications of this technology in disease diagnosis, treatment response, and prognosis prediction from the perspective of radiology images, multimodality images, histopathology images, and three-dimensional dose distribution data, particularly in oncology, is presented. The potential and value of radiomics in diagnostic and therapeutic strategies are also further analyzed, and for the first time, the advances and challenges associated with dosiomics in radiotherapy are summarized, highlighting the latest progress in radiomics. Finally, a robust framework for radiomic analysis is presented and challenges and recommendations for future development are discussed, including but not limited to the factors that affect model stability (medical big data and multitype data and expert knowledge in medical), limitations of data-driven processes (reproducibility and interpretability of studies, different treatment alternatives for various institutions, and prospective researches and clinical trials), and thoughts on future directions (the capability to achieve clinical applications and open platform for radiomics analysis).

show abstract

“…Deep learning models have achieved high accuracy for detecting tumor regions 14 and identifying cancer subtypes 15 from hematoxylin and eosin (H&E)-stained whole slide images. Such models have also been able to predict several clinically relevant genetic features, such as microsatellite instability (MSI) 16 and mutation status of key genes [17][18][19] with moderate accuracy. Deep learning models using WSIs have been studied to stratify patients based on survival risk 20 .…”

Section: Introductionmentioning

confidence: 99%

Integrative deep learning analysis improves colon adenocarcinoma patient stratification at risk for mortality

Zhou

pour

Deirawan

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Colorectal cancers are the fourth most commonly diagnosed cancer and the second leading cancer in number of deaths. Many clinical variables, pathological features, and genomic signatures are associated with patient risk, but reliable patient stratification in the clinic remains a challenging task. Here we assess how image, clinical, and genomic features can be combined to predict risk. We first observe that deep learning models based only on whole slide images (WSIs) from The Cancer Genome Atlas accurately separate high risk (OS<3years, N=38) from low risk (OS>5years, N=25) patients (AUC=0.81±0.08, 5year survival p-value=2.13e-25, 5year relative risk=5.09±0.05) though such models are less effective at predicting OS for moderate risk (3years<OS<5years, N=45) patients (5year survival p-value=0.5, 5year relative risk=1.32±0.09). However, we find that novel integrative models combining whole slide images, clinical variables, and mutation signatures can improve patient stratification for moderate risk patients (5year survival p-value=6.69e-30, 5year relative risk=5.32±0.07). Our integrative model combining image and clinical variables is also effective on an independent pathology dataset generated by our team (3year survival p-value=1.14e-09, 5year survival p-value=2.15e-05, 3year relative risk=3.25±0.06, 5year relative-risk=3.07±0.08). The integrative model substantially outperforms models using only images or only clinical variables, indicating beneficial cross-talk between the data types. Pathologist review of image-based heatmaps suggests that nuclear shape, nuclear size pleomorphism, intense cellularity, and abnormal structures are associated with high risk, while low risk regions tend to have more regular and small cells. The improved stratification of colorectal cancer patients from our computational methods can be beneficial for preemptive development of management and treatment plans for individual patients, as well as for informed enrollment of patients in clinical trials.

show abstract

Deep learning trained on hematoxylin and eosin tumor region of Interest predicts HER2 status and trastuzumab treatment response in HER2+ breast cancer

Cited by 96 publications

References 44 publications

HER2 Low, Ultra-low, and Novel Complementary Biomarkers: Expanding the Spectrum of HER2 Positivity in Breast Cancer

HER2 Low, Ultra-low, and Novel Complementary Biomarkers: Expanding the Spectrum of HER2 Positivity in Breast Cancer

Deep Learning With Radiomics for Disease Diagnosis and Treatment: Challenges and Potential

Integrative deep learning analysis improves colon adenocarcinoma patient stratification at risk for mortality

Contact Info

Product

Resources

About