The Australian Breast Cancer Tissue Bank (ABCTB)

Carpenter, Jane; Marsh, Deborah J.; Mariasegaram, Mythily; Clarke, Christine L.

doi:10.5334/ojb.aa

Cited by 9 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, they trained and evaluated their classifier to predict PR and HER2 status and obtained an AUC of 0.810 (95% CI: 0.769-0.846) on PR and an AUC of 0.778 (95% CI: 0.730-0.825) on HER2. Rawat et al (34), trained a CNN classifier on 939 TCGA H&E images to predict HER2 status and were able to achieve slide-level AUCs of 0.71 through internal cross-validation, though unexpectedly they observed a higher AUC (0.79) on an independent set (35). Bychkov et al (36), investigated whether predicting HER2 status using a CNN model can guide the choice of therapy.…”

Section: Introductionmentioning

confidence: 99%

“…Bychkov et al (36), investigated whether predicting HER2 status using a CNN model can guide the choice of therapy. The study utilized cancer tissue samples from FinProg patient series (35), the FinProg validation series (37), and the FinHer clinical trial (38), all of which had HER2 amplification determined by CISH. Their CNN model, trained on 693 H&E-stained patient samples from the FinProg series, was able to predict tile-level HER2 status with AUC 0.70 in a 5-fold cross validation and AUC 0.67 on 712 test images from the FinHer dataset.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

et al. 2022

View full text Add to dashboard Cite

The current standard of care for many patients with HER2-positive breast cancer is neoadjuvant chemotherapy in combination with anti-HER2 agents, based on HER2 amplification as detected by in situ hybridization (ISH) or protein immunohistochemistry (IHC). However, hematoxylin & eosin (H&E) tumor stains are more commonly available, and accurate prediction of HER2 status and anti-HER2 treatment response from H&E would reduce costs and increase the speed of treatment selection. Computational algorithms for H&E have been effective in predicting a variety of cancer features and clinical outcomes, including moderate success in predicting HER2 status. In this work, we present a novel convolutional neural network (CNN) approach able to predict HER2 status with increased accuracy over prior methods. We trained a CNN classifier on 188 H&E whole slide images (WSIs) manually annotated for tumor regions of interest (ROIs) by our pathology team. Our classifier achieved an area under the curve (AUC) of 0.90 in crossvalidation of slide-level HER2 status and 0.81 on an independent TCGA test set. Within slides, we observed strong agreement between pathologist annotated ROIs and blinded computational predictions of tumor regions / HER2 status. Moreover, we trained our classifier on pre-treatment samples from 187 HER2+ patients that subsequently received trastuzumab therapy. Our classifier achieved an AUC of 0.80 in a five-fold cross validation. Our work provides an H&E-based algorithm that can predict HER2 status and trastuzumab response in breast cancer at an accuracy that is better than IHC and may benefit clinical evaluations..

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Their model achieved a slide-level HER2 AUC of 0.71 (TCGA, n = 124) in a 5-fold cross validation. They also tested the generalizability of their model using an independent cohort from The Australian Breast Cancer Tissue Bank (ABCTB) (35). Their model achieved a slide-level Her2 AUC = 0.79 (ABCTB, n = 2487).…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning trained on H&E tumor ROIs predicts HER2 status and Trastuzumab treatment response in HER2+ breast cancer

Farahmand

Fernandez

Ahmed

et al. 2021

Preprint

View full text Add to dashboard Cite

The current standard of care for many patients with HER2-positive breast cancer is neoadjuvant chemotherapy in combination with anti-HER2 agents, based on HER2 amplification as detected by in situ hybridization (ISH) or protein immunohistochemistry (IHC). However, hematoxylin & eosin (H&E) tumor stains are more commonly available, and accurate prediction of HER2 status and anti-HER2 treatment response from H&E would reduce costs and increase the speed of treatment selection. Computational algorithms for H&E have been effective in predicting a variety of cancer features and clinical outcomes, including moderate success in predicting HER2 status. In this work, we present a novel convolutional neural network (CNN) approach able to predict HER2 status with increased accuracy over prior methods. We trained a CNN classifier on 188 H&E whole slide images (WSIs) manually annotated for tumor regions of interest (ROIs) by our pathology team. Our classifier achieved an area under the curve (AUC) of 0.90 in cross-validation of slide-level HER2 status and 0.81 on an independent TCGA test set. Within slides, we observed strong agreement between pathologist annotated ROIs and blinded computational predictions of tumor regions / HER2 status. Moreover, we trained our classifier on pre-treatment samples from 187 HER2+ patients that subsequently received trastuzumab therapy. Our classifier achieved an AUC of 0.80 in a five-fold cross validation. Our work provides an H&E-based algorithm that can predict HER2 status and trastuzumab response in breast cancer at an accuracy that is better than IHC and may benefit clinical evaluations.

show abstract

A Global View of Breast Tissue Banking

Wilson

Botfield

Speirs

2015

Advances in Experimental Medicine and Biology

View full text Add to dashboard Cite

The importance of accessing high quality clinical samples for translational research is now firmly recognised. Traditionally these samples were collected and curated by individuals with an interest in a particular disease type. In recent years the idea of centralising and storing tissue collections in the form of tissue banks or biobanks has developed. As a result a number of biobanks have been established in many different countries. These can be either single centres or multi centre collaborations, often in the form of a federated network. This chapter outlines the development of breast tissue banking in a global context and discusses some of the challenges that lie ahead for the field, in particular how to meet the growing needs of researchers, how to make the best use of donated samples and how to increase the visibility of samples residing in biobanks to researchers.

show abstract

The Australian Breast Cancer Tissue Bank (ABCTB)

Cited by 9 publications

References 7 publications

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

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

Deep learning trained on H&E tumor ROIs predicts HER2 status and Trastuzumab treatment response in HER2+ breast cancer

A Global View of Breast Tissue Banking

Contact Info

Product

Resources

About