American Association for Cancer Research
Browse
cd-23-0996_figure_s12_results_flow_for_met_site_histology_prior_suppsf12.pdf (1.32 MB)

Figure S12: Results flow for Met Site, Histology Prior from Deep-Learning Model for Tumor-Type Prediction Using Targeted Clinical Genomic Sequencing Data

Download (1.32 MB)
journal contribution
posted on 2024-06-03, 07:40 authored by Madison Darmofal, Shalabh Suman, Gurnit Atwal, Michael Toomey, Jie-Fu Chen, Jason C. Chang, Efsevia Vakiani, Anna M. Varghese, Anoop Balakrishnan Rema, Aijazuddin Syed, Nikolaus Schultz, Michael F. Berger, Quaid Morris

Flow of results for combination prior using either metastatic site (left) or histology (right). Metastatic Site prior is only applied to all metastatic samples (n = 2166), histological prior is applied to all test set examples with histological annotations (n = 4571). Arrow base represents pre-adjustment category, arrow head represents post-adjustment. Circle arrow indicates the number of samples that did not change categories after adjustment, i.e. 1406 samples that were correct and high confidence before and after skewing for metastatic biopsy site annotations.

Funding

National Cancer Institute (NCI)

United States Department of Health and Human Services

Find out more...

History

ARTICLE ABSTRACT

Tumor type guides clinical treatment decisions in cancer, but histology-based diagnosis remains challenging. Genomic alterations are highly diagnostic of tumor type, and tumor-type classifiers trained on genomic features have been explored, but the most accurate methods are not clinically feasible, relying on features derived from whole-genome sequencing (WGS), or predicting across limited cancer types. We use genomic features from a data set of 39,787 solid tumors sequenced using a clinically targeted cancer gene panel to develop Genome-Derived-Diagnosis Ensemble (GDD-ENS): a hyperparameter ensemble for classifying tumor type using deep neural networks. GDD-ENS achieves 93% accuracy for high-confidence predictions across 38 cancer types, rivaling the performance of WGS-based methods. GDD-ENS can also guide diagnoses of rare type and cancers of unknown primary and incorporate patient-specific clinical information for improved predictions. Overall, integrating GDD-ENS into prospective clinical sequencing workflows could provide clinically relevant tumor-type predictions to guide treatment decisions in real time. We describe a highly accurate tumor-type prediction model, designed specifically for clinical implementation. Our model relies only on widely used cancer gene panel sequencing data, predicts across 38 distinct cancer types, and supports integration of patient-specific nongenomic information for enhanced decision support in challenging diagnostic situations.See related commentary by Garg, p. 906.This article is featured in Selected Articles from This Issue, p. 897