Predicting tumour mutational burden from histopathological images using multiscale deep learning

@article{Jain2020PredictingTM,
  title={Predicting tumour mutational burden from histopathological images using multiscale deep learning},
  author={Mika Sarkin Jain and Tarik F. Massoud},
  journal={Nature Machine Intelligence},
  year={2020},
  volume={2},
  pages={356-362}
}
Tumour mutational burden (TMB) is an important biomarker for predicting the response to immunotherapy in patients with cancer. Gold-standard measurement of TMB is performed using whole exome sequencing (WES), which is not available at most hospitals because of its high cost, operational complexity and long turnover times. We have developed a machine learning algorithm, Image2TMB, which can predict TMB from readily available lung adenocarcinoma histopathological images. Image2TMB integrates the… 
Deep Learning of Histopathological Features for the Prediction of Tumour Molecular Genetics
TLDR
Recent developments in the field of deep learning for predicting molecular genetics from histopathological slides are outlined, as well as to highlight limitations and pitfalls of working with histopathology slides in deep learning.
Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer with Deep Learning
TLDR
The histomorphologic features determining the LNM prediction were revealed indicating the model ability to learn informative latent features, which indicates its prediction accuracy and potential adaptability among multiple centres.
Deep learning in cancer diagnosis, prognosis and treatment selection
TLDR
This review provides an overview of emerging deep learning techniques and how they are being applied to oncology, and focuses on the deep learning applications for omics data types, including genomic, methylation and transcriptomic data, as well as histopathology-based genomic inference.
Clinicopathological Features, Tumor Mutational Burden, and Tumour-Infiltrating Lymphocyte Interplay in ERBB2-Mutated Breast Cancer: In Silico Analysis
TLDR
Evidence is provided for a link between ERBB2 mutations and high TMB in BC, particularly in invasive lobular carcinoma.
Bayesian hierarchical finite mixture of regression for histopathological imaging‐based cancer data analysis
TLDR
This study provides a new venue for FMR analysis for cancer and other complex diseases by simultaneously use two types of histopathological imaging features, which are extracted based on domain‐specific biomedical knowledge and using automated signal processing software, respectively.
All You Need is Color: Image based Spatial Gene Expression Prediction using Neural Stain Learning
TLDR
The proposed method with only 11 trainable weight parameters outperforms both classical regression models with cellular composition and morphological features as well as deep learning methods and shows higher correlations with true expression values obtained through sequencing for a larger set of genes in comparison to other approaches.
Multi_Scale_Tools: A Python Library to Exploit Multi-Scale Whole Slide Images
TLDR
Multi_Scale_Tools, a library aiming to facilitate exploiting the multi-scale structure of WSIs, and outperform the single-magnification CNN for both classification and segmentation tasks.
Review of deep learning: concepts, CNN architectures, challenges, applications, future directions
TLDR
This review attempts to provide a more comprehensive survey of the most important aspects of DL and including those enhancements recently added to the field, and outlines the importance of DL, and presents the types of DL techniques and networks.
...
...

References

SHOWING 1-10 OF 26 REFERENCES
Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning
TLDR
A deep convolutional neural network model is trained on whole-slide images obtained from The Cancer Genome Atlas to accurately and automatically classify them into LUAD, LUSC or normal lung tissue and predicts the ten most commonly mutated genes in LUAD.
Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features
TLDR
It is suggested that automatically derived image features can predict the prognosis of lung cancer patients and thereby contribute to precision oncology.
Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden
TLDR
Measurements of TMB from comprehensive genomic profiling are strongly reflective of measurements from whole exome sequencing and model that below 0.5 Mb the variance in measurement increases significantly, demonstrating that many disease types have a substantial portion of patients with high TMB who might benefit from immunotherapy.
Tumor Mutational Burden as an Independent Predictor of Response to Immunotherapy in Diverse Cancers
TLDR
Higher TMB predicts favorable outcome to PD-1/PD-L1 blockade across diverse cancers treated with various immunotherapies, and Benefit from dual checkpoint blockade did not show a similarly strong dependence on TMB.
Tumor mutational load predicts survival after immunotherapy across multiple cancer types
TLDR
Analysis of advanced cancer patients treated with immune-checkpoint inhibitors shows that tumor mutational burden, as assessed by targeted next-generation sequencing, predicts survival after immunotherapy across multiple cancer types.
Size matters: Dissecting key parameters for panel‐based tumor mutational burden analysis
TLDR
It is demonstrated that panel size is a critical parameter that influences confidence intervals (CIs) and cutoff values as well as important test parameters including sensitivity, specificity, and positive predictive value, and the Illumina TSO500 panel is evaluated.
Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer
TLDR
In the setting of a challenge competition, some deep learning algorithms achieved better diagnostic performance than a panel of 11 pathologists participating in a simulation exercise designed to mimic routine pathology workflow; algorithm performance was comparable with an expert pathologist interpreting whole-slide images without time constraints.
Blood-based tumor mutational burden as a predictor of clinical benefit in non-small-cell lung cancer patients treated with atezolizumab
TLDR
It is shown that high bTMB is a clinically actionable biomarker for atezolizumab in NSCLC, and a blood-based DNA sequencing assay to infer tumor mutational burden in the absence of tumor biopsy predicts response to PD-L1 blockade in patients with non-small-cell lung cancer.
...
...