Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study

Niehues, Jan Moritz; Quirke, Philip; West, Nicholas P.; Grabsch, Heike; Treeck, Marko van; Schirris, Yoni; Veldhuizen, Gregory Patrick; Hutchins, Gordon; Richman, Susan D.; Foersch, Sebastian; Brinker, Titus J.; Fukuoka, Junya; Bychkov, Andrey; Uegami, Wataru; Truhn, Daniel; Brenner, Hermann; Brobeil, Alexander; Hoffmeister, Michael; Kather, Jakob Nikolas

doi:10.1016/j.xcrm.2023.100980

Cited by 23 publications

(13 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We speculated that chemotherapeutic agents are more easily able to exert antitumor functions in patients with a low pathomics signature because of low tumor intensity, but further study is needed to investigate this hypothesis. Studies have reported that pathological features of H-E–stained slides were associated with microsatellite instability or stability, KRAS alteration, or other molecular subtypes . Hence, an integrated analysis of pathomics features and other omics technologies (eg, genomics, transcriptomics, proteomics) may be a feasible approach to unraveling the underlying mechanisms of the pathomics signature with adjuvant chemotherapy benefits …”

Section: Discussionmentioning

confidence: 99%

“…Studies have reported that pathological features of H-E-stained slides were associated with microsatellite instability or stability, KRAS alteration, or other molecular subtypes. 31,35,36 Hence, an integrated analysis of pathomics features and other omics technologies (eg, genomics, transcriptomics, proteomics) may be a feasible approach to unraveling the underlying mechanisms of the pathomics signature with adjuvant chemotherapy benefits. 36,37 For patients who do not benefit from chemotherapy, it is necessary to consider other potential treatment strategies, such as triple-drug combination chemotherapy, targeted therapy,…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Pathomics Signature for Prognosis and Chemotherapy Benefits in Stage III Colon Cancer

Jiang,

Wang,

Dong

et al. 2024

JAMA Surg

View full text Add to dashboard Cite

ImportanceThe current TNM staging system may not provide adequate information for prognostic purposes and to assess the potential benefits of chemotherapy for patients with stage III colon cancer.ObjectiveTo develop and validate a pathomics signature to estimate prognosis and benefit from chemotherapy using hematoxylin-eosin (H-E)–stained slides.Design, Setting, and ParticipantsThis retrospective prognostic study used data from consecutive patients with histologically confirmed stage III colon cancer at 2 medical centers between January 2012 and December 2015. A total of 114 pathomics features were extracted from digital H-E–stained images from Nanfang Hospital of Southern Medical University, Guangzhou, China, and a pathomics signature was constructed using a least absolute shrinkage and selection operator Cox regression model in the training cohort. The associations of the pathomics signature with disease-free survival (DFS) and overall survival (OS) were evaluated. Patients at the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, China, formed the validation cohort. Data analysis was conducted from September 2022 to March 2023.Main Outcomes and MeasuresThe prognostic accuracy of the pathomics signature as well as its association with chemotherapy response were evaluated.ResultsThis study included 785 patients (mean [SD] age, 62.7 [11.1] years; 437 [55.7%] male). A pathomics signature was constructed based on 4 features. Multivariable analysis revealed that the pathomics signature was an independent factor associated with DFS (hazard ratio [HR], 2.46 [95% CI, 2.89-4.13]; P &lt; .001) and OS (HR, 2.78 [95% CI, 2.34-3.31]; P &lt; .001) in the training cohort. Incorporating the pathomics signature into pathomics nomograms resulted in better performance for the estimation of prognosis than the traditional model in a concordance index comparison in the training cohort (DFS: HR, 0.88 [95% CI, 0.86-0.89] vs HR, 0.73 [95% CI, 0.71-0.75]; P &lt; .001; OS: HR, 0.85 [95% CI, 0.84-0.86] vs HR, 0.74 [95% CI, 0.72-0.76]; P &lt; .001) and validation cohort (DFS: HR, 0.83 [95% CI, 0.82-0.85] vs HR, 0.70 [95% CI, 0.67-0.72]; P &lt; .001; OS: HR, 0.80 [95% CI, 0.78-0.82] vs HR, 0.69 [0.67-0.72]; P &lt; .001). Further analysis revealed that patients with a low pathomics signature were more likely to benefit from chemotherapy (eg, combined cohort: DFS: HR, 0.44 [95% CI, 0.28-0.69]; P = .001; OS: HR, 0.43 [95% CI, 0.29-0.64]; P &lt; .001).Conclusions and RelevanceThese findings suggest that a pathomics signature could help identify patients most likely to benefit from chemotherapy in stage III colon cancer.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Pathomics Signature for Prognosis and Chemotherapy Benefits in Stage III Colon Cancer

Jiang,

Wang,

Dong

et al. 2024

JAMA Surg

View full text Add to dashboard Cite

show abstract

“…Most AI studies in histopathology employ supervised DL. Of particular relevance are “weakly” supervised approaches, in which the objective of the system is to predict a “label” for the WSI in its entirety [ 13 , 21 , 22 ]. A “label” can refer to any of the basic and advanced categories, including properties of slides (presence of tumor), properties of tumors (subtype or genetic alterations), and of patients (survival or response) [ 13 ].…”

Section: In Histopathologymentioning

confidence: 99%

“…Histopathology images are widely available and inexpensive, but only show tissue phenotype, not necessarily underlying molecular changes. Therefore, it was shown that already the addition of clinical parameters from the patient could improve the generalizability of DL models improving the predictions [ 21 ]. Genomic methods, on the other hand, can offer a glimpse into the underlying machinery within the cells, but there is still the disadvantage that a certain amount of material is required to obtain such information, which is not always feasible.…”

Section: Multimodalitymentioning

confidence: 99%

Deep learning in cancer genomics and histopathology

Unger,

Kather

2024

Genome Med

Self Cite

View full text Add to dashboard Cite

Histopathology and genomic profiling are cornerstones of precision oncology and are routinely obtained for patients with cancer. Traditionally, histopathology slides are manually reviewed by highly trained pathologists. Genomic data, on the other hand, is evaluated by engineered computational pipelines. In both applications, the advent of modern artificial intelligence methods, specifically machine learning (ML) and deep learning (DL), have opened up a fundamentally new way of extracting actionable insights from raw data, which could augment and potentially replace some aspects of traditional evaluation workflows. In this review, we summarize current and emerging applications of DL in histopathology and genomics, including basic diagnostic as well as advanced prognostic tasks. Based on a growing body of evidence, we suggest that DL could be the groundwork for a new kind of workflow in oncology and cancer research. However, we also point out that DL models can have biases and other flaws that users in healthcare and research need to know about, and we propose ways to address them.

show abstract

“…MSI and HRD are both abnormalities impacting the deoxyribonucleic acid damage repair (DDR) process in tumors. Early recognition of those biomarkers may benefit the patients through specific therapies targeting DDR-related genomic alterations (19, 20). This is of particular interest in breast cancer (21, 56) and colorectal cancer (11, 21).…”

Section: Experimental and Evaluation Setupmentioning

confidence: 99%

Scaling Self-Supervised Learning for Histopathology with Masked Image Modeling

Filiot¹,

Ghermi²,

Olivier³

et al. 2023

Preprint

View full text Add to dashboard Cite

Computational pathology is revolutionizing the field of pathology by integrating advanced computer vision and machine learning technologies into diagnostic workflows. Recently, self-supervised learning (SSL) has emerged as a promising solution to learn representations from histology patches, leveraging large volumes of unannotated whole slide images (WSI). In particular, Masked Image Modeling (MIM) showed remarkable results and robustness over purely contrastive learning methods. In this work, we explore the application of MIM to histology using iBOT, a self-supervised transformer-based framework. Through a wide range of downstream tasks over seven cancer indications, we provide recommendations on the pre-training of large models for histology data using MIM. First, we demonstrate that in-domain pre-training with iBOT outperforms both ImageNet pre-training and a model pre-trained with a purely contrastive learning objective, MoCo V2. Second, we show that Vision Transformers models (ViT), when scaled appropriately, have the capability to learn pan-cancer representations that benefit a large variety of downstream tasks. Finally, our iBOT ViT-Base model, pre-trained on more than 40 million histology images from 16 different cancer types, achieves state-of-the-art performance in most weakly-supervised WSI classification tasks compared to other SSL frameworks.

show abstract

Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: A retrospective multi-centric study

Cited by 23 publications

References 53 publications

Pathomics Signature for Prognosis and Chemotherapy Benefits in Stage III Colon Cancer

Pathomics Signature for Prognosis and Chemotherapy Benefits in Stage III Colon Cancer

Deep learning in cancer genomics and histopathology

Scaling Self-Supervised Learning for Histopathology with Masked Image Modeling

Contact Info

Product

Resources

About