Deep-Learning Model for Tumor-Type Prediction Using Targeted Clinical Genomic Sequencing Data.

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. SIGNIFICANCE: 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.

Enhanced clinical assessment of hematologic malignancies through routine paired tumor and normal sequencing.

Genomic profiling of hematologic malignancies has augmented our understanding of variants that contribute to disease pathogenesis and supported development of prognostic models that inform disease management in the clinic. Tumor only sequencing assays are limited in their ability to identify definitive somatic variants, which can lead to ambiguity in clinical reporting and patient management. Here, we describe the MSK-IMPACT Heme cohort, a comprehensive data set of somatic alterations from paired tumor and normal DNA using a hybridization capture-based next generation sequencing platform. We highlight patterns of mutations, copy number alterations, and mutation signatures in a broad set of myeloid and lymphoid neoplasms. We also demonstrate the power of appropriate matching to make definitive somatic calls, including in patients who have undergone allogeneic stem cell transplant. We expect that this resource will further spur research into the pathobiology and clinical utility of clinical sequencing for patients with hematologic neoplasms.