MOCOnet: Robust Motion Correction of Cardiovascular Magnetic Resonance T1 Mapping Using Convolutional Neural Networks

Background: Quantitative cardiovascular magnetic resonance (CMR) T1 mapping has shown promise for advanced tissue characterisation in routine clinical practise. However, T1 mapping is prone to motion artefacts, which affects its robustness and clinical interpretation. Current methods for motion correction on T1 mapping are model-driven with no guarantee on generalisability, limiting its widespread use. In contrast, emerging data-driven deep learning approaches have shown good performance in general image registration tasks. We propose MOCOnet, a convolutional neural network solution, for generalisable motion artefact correction in T1 maps.Methods: The network architecture employs U-Net for producing distance vector fields and utilises warping layers to apply deformation to the feature maps in a coarse-to-fine manner. Using the UK Biobank imaging dataset scanned at 1.5T, MOCOnet was trained on 1,536 mid-ventricular T1 maps (acquired using the ShMOLLI method) with motion artefacts, generated by a customised deformation procedure, and tested on a different set of 200 samples with a diverse range of motion. MOCOnet was compared to a well-validated baseline multi-modal image registration method. Motion reduction was visually assessed by 3 human experts, with motion scores ranging from 0% (strictly no motion) to 100% (very severe motion).Results: MOCOnet achieved fast image registration (<1 second per T1 map) and successfully suppressed a wide range of motion artefacts. MOCOnet significantly reduced motion scores from 37.1±21.5 to 13.3±10.5 (p < 0.001), whereas the baseline method reduced it to 15.8±15.6 (p < 0.001). MOCOnet was significantly better than the baseline method in suppressing motion artefacts and more consistently (p = 0.007).Conclusion: MOCOnet demonstrated significantly better motion correction performance compared to a traditional image registration approach. Salvaging data affected by motion with robustness and in a time-efficient manner may enable better image quality and reliable images for immediate clinical interpretation.

Expert Curation of Somatic FLT3 Variants By the Clingen Somatic Hematologic Cancer Taskforce (ClinGen HCT)

Clinical significance of somatic gene variants needs to be comprehensively characterized for their diagnostic, prognostic and/or therapeutic actionability in patient management. However, challenges remain due to discrepancies in interpretation and reporting of these somatic variants among different testing labs. Therefore, standardized curation, clinical interpretation and reporting of somatic variants in hematologic cancers is critical. To address this issue, the Hematologic Cancer Taskforce (HCT), composed of 52 multi-disciplinary experts including oncologists, molecular pathologists, lab directors, genomic scientists and biocurators, was formed in January 2020 within the ClinGen Somatic Cancer Clinical Domain Working Group (CDWG) with a goal to undertake systematic curation and evidence-based clinical interpretation of genes/somatic variants associated with hematologic malignancies. In collaboration with the Clinical Interpretation of Variants in Cancer (CIViC) (civicdb.org) knowledgebase, HCT members curate, edit, and verify Evidence Items for each variant extracted from peer-reviewed publications. Monthly discussions based on these Evidence Items lead to the preparation of variant Assertions, which summarize the state of the field consensus variant interpretation and include tiering based on the AMP/ASCO/CAP guidelines (PMID: 27993330). FMS-like tyrosine kinase 3 (FLT3) encodes a class III receptor tyrosine kinase expressed in hematopoietic cells. FLT3 mutations, including both internal tandem duplication (ITD) and mutations in the tyrosine kinase domain (TKD), are the most common mutations in acute myeloid leukemia (AML), occurring in approximately 30% of all AML cases. Implementing FLT3 tyrosine kinase inhibitors (TKIs) in different treatment regimens for FLT3 mutated AML patients has led to significantly improved overall survival. Type I FLT3 inhibitors, including midostaurin, gilteritinib, sunitinib, lestaurtinib, and crenolanib, bind to the ATP-binding site when the receptor is in active conformation. Type II FLT3 inhibitors, including sorafenib, ponatinib, and quizartinib, interact with a hydrophobic region directly adjacent to the ATP-binding domain that is only accessible when the receptor is inactive, which prevents receptor activation. Generally in AML cells, type I FLT3 inhibitors prevent activity for both ITD and TKD mutations, while Type II inhibitors target ITD but lack efficiency against TKD mutations. The development of TKD mutations in AML cells with ITD have proved to be a mechanism of acquired, or secondary resistance to Type II FLT3 inhibitors. The HCT is piloting curation assessments of FLT3 alterations, including ITD, TKD and non-TKD variants, in AML. So far, the HCT has curated 75 evidence items for FLT3 somatic variants. FLT3-ITD, as well as D835 and I836 were asserted as tier 1 level A variants based on the prediction of response to gilteritinib in relapsed/refractory AML (PMIDs: 27993330, 31665578, 28645776, 28516360, 27908881). Recent curation activities are focused on FLT3 D839G and N676K, as clinical trials using large AML patient cohorts are lacking in their ability to validate drug response/resistance associations of these two TKD variants due to their low frequency. Functional studies showed both variants result in increased proliferation and protection from apoptosis, supporting the oncogenic potential of these two variants (PMIDs: 26891877, 2468088). FLT3 D839G combined with ITD confers resistance to pexidartinib and ponatinib, both Type II FLT3 inhibitors (PMIDs: 25847190, 23430109). FLT3 N676K predicts response to the Type I FLT3 inhibitor, gilteritinib, when N676K is present alone or in combination with ITD. Interestingly, FLT3 N676K in the absence of ITD predicts response to sorafenib, a Type II FLT3 inhibitor (PMIDs: 32040554, 32984009). However, these results are mostly derived from in vitro studies. Two separate Tier II, Level D Assertions have been submitted for FLT3-ITD&D839G for its response to pexidartinib and ponatinib, and more evidence is being collected to form an Assertion for FLT3 N676K. The complexity of the prediction of response/resistance associated with FLT3 D839G and N676K supports the importance of evidence-based curation and collection for these variants in the context of the overall mutation profile, disease context and specific FLT3 TKIs to clearly define their clinical impact. Disclosures Pullarkat: Stemline Therapeutics: Honoraria.

Germline Sequencing Improves Tumor-Only Sequencing Interpretation in a Precision Genomic Study of Patients With Pediatric Solid Tumor

PURPOSE Molecular tumor profiling is becoming a routine part of clinical cancer care, typically involving tumor-only panel testing without matched germline. We hypothesized that integrated germline sequencing could improve clinical interpretation and enhance the identification of germline variants with significant hereditary risks. MATERIALS AND METHODS Tumors from pediatric patients with high-risk, extracranial solid malignancies were sequenced with a targeted panel of cancer-associated genes. Later, germline DNA was analyzed for a subset of these genes. We performed a post hoc analysis to identify how an integrated analysis of tumor and germline data would improve clinical interpretation. RESULTS One hundred sixty participants with both tumor-only and germline sequencing reports were eligible for this analysis. Germline sequencing identified 38 pathogenic or likely pathogenic variants among 35 (22%) patients. Twenty-five (66%) of these were included in the tumor sequencing report. The remaining germline pathogenic or likely pathogenic variants were single-nucleotide variants filtered out of tumor-only analysis because of population frequency or copy-number variation masked by additional copy-number changes in the tumor. In tumor-only sequencing, 308 of 434 (71%) single-nucleotide variants reported were present in the germline, including 31% with suggested clinical utility. Finally, we provide further evidence that the variant allele fraction from tumor-only sequencing is insufficient to differentiate somatic from germline events. CONCLUSION A paired approach to analyzing tumor and germline sequencing data would be expected to improve the efficiency and accuracy of distinguishing somatic mutations and germline variants, thereby facilitating the process of variant curation and therapeutic interpretation for somatic reports, as well as the identification of variants associated with germline cancer predisposition.

Practical Fundamentals of Clinical MEG Interpretation in Epilepsy

Magnetoencephalography (MEG) is a neurophysiologic test that offers a functional localization of epileptic sources in patients considered for epilepsy surgery. The understanding of clinical MEG concepts, and the interpretation of these clinical studies, are very involving processes that demand both clinical and procedural expertise. One of the major obstacles in acquiring necessary proficiency is the scarcity of fundamental clinical literature. To fill this knowledge gap, this review aims to explain the basic practical concepts of clinical MEG relevant to epilepsy with an emphasis on single equivalent dipole (sECD), which is one the most clinically validated and ubiquitously used source localization method, and illustrate and explain the regional topology and source dynamics relevant for clinical interpretation of MEG-EEG.

Integrating molecular profiles into clinical frameworks through the Molecular Oncology Almanac to prospectively guide precision oncology

Tumor molecular profiling of single gene-variant (‘first-order’) genomic alterations informs potential therapeutic approaches. Interactions between such first-order events and global molecular features (for example, mutational signatures) are increasingly associated with clinical outcomes, but these ‘second-order’ alterations are not yet accounted for in clinical interpretation algorithms and knowledge bases. We introduce the Molecular Oncology Almanac (MOAlmanac), a paired clinical interpretation algorithm and knowledge base to enable integrative interpretation of multimodal genomic data for point-of-care decision making and translational-hypothesis generation. We benchmarked MOAlmanac to a first-order interpretation method across multiple retrospective cohorts and observed an increased number of clinical hypotheses from evaluation of molecular features and profile-to-cell line matchmaking. When applied to a prospective precision oncology trial cohort, MOAlmanac nominated a median of two therapies per patient and identified therapeutic strategies administered in 47% of patients. Overall, we present an open-source computational method for integrative clinical interpretation of individualized molecular profiles.