Refine
Year of publication
- 2018 (2) (remove)
Document Type
- Article (2) (remove)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- ALK gene (1)
- Gene fusion (1)
- Lung adenocarcinoma (1)
- Machine learning (1)
- Medical imaging (1)
- Next generation sequencing (1)
- Pathology (1)
- Prostate cancer (1)
Institute
- Medizin (2) (remove)
The Gleason grading system remains the most powerful prognostic predictor for patients with prostate cancer since the 1960s. Its application requires highly-trained pathologists, is tedious and yet suffers from limited inter-pathologist reproducibility, especially for the intermediate Gleason score 7. Automated annotation procedures constitute a viable solution to remedy these limitations. In this study, we present a deep learning approach for automated Gleason grading of prostate cancer tissue microarrays with Hematoxylin and Eosin (H&E) staining. Our system was trained using detailed Gleason annotations on a discovery cohort of 641 patients and was then evaluated on an independent test cohort of 245 patients annotated by two pathologists. On the test cohort, the inter-annotator agreements between the model and each pathologist, quantified via Cohen’s quadratic kappa statistic, were 0.75 and 0.71 respectively, comparable with the inter-pathologist agreement (kappa = 0.71). Furthermore, the model’s Gleason score assignments achieved pathology expert-level stratification of patients into prognostically distinct groups, on the basis of disease-specific survival data available for the test cohort. Overall, our study shows promising results regarding the applicability of deep learning-based solutions towards more objective and reproducible prostate cancer grading, especially for cases with heterogeneous Gleason patterns.
Oncogenic rearrangements leading to targetable gene fusions are well-established cancer driver events in lung adenocarcinoma. Accurate and reliable detection of these gene fusions is crucial to select the appropriate targeted therapy for each patient. We compared the targeted next-generation-sequencing Oncomine Focus Assay (OFA; Thermo Fisher Scientific) with conventional ALK FISH and anti-Alk immunohistochemistry in a cohort of 52 lung adenocarcinomas (10 ALK rearranged, 18 non-ALK rearranged, and 24 untested cases). We found a sensitivity and specificity of 100% for detection of ALK rearrangements using the OFA panel. In addition, targeted next generation sequencing allowed us to analyze a set of 23 driver genes in a single assay. Besides EML4-ALK (11/52 cases), we detected EZR-ROS1 (1/52 cases), KIF5B-RET (1/52 cases) and MET-MET (4/52 cases) fusions. All EML4-ALK, EZR-ROS1 and KIF5B-RET fusions were confirmed by multiplexed targeted next generation sequencing assay (Oncomine Solid Tumor Fusion Transcript Kit, Thermo Fisher Scientific). All cases with EML4-ALK rearrangement were confirmed by Alk immunohistochemistry and all but one by ALK FISH. In our experience, targeted next-generation sequencing is a reliable and timesaving tool for multiplexed detection of targetable rearrangements. Therefore, targeted next-generation sequencing represents an efficient alternative to time-consuming single target assays currently used in molecular pathology.