Development and Validation of a Natural Language Processing Model to Identify Low-Risk Pulmonary Embolism in Real Time to Facilitate Safe Outpatient Management.

STUDY OBJECTIVE: This study aimed to (1) develop and validate a natural language processing model to identify the presence of pulmonary embolism (PE) based on real-time radiology reports and (2) identify low-risk PE patients based on previously validated risk stratification scores using variables extracted from the electronic health record at the time of diagnosis. The combination of these approaches yielded an natural language processing-based clinical decision support tool that can identify patients presenting to the emergency department (ED) with low-risk PE as candidates for outpatient management. METHODS: Data were curated from all patients who received a PE-protocol computed tomography pulmonary angiogram (PE-CTPA) imaging study in the ED of a 3-hospital academic health system between June 1, 2018 and December 31, 2020 (n=12,183). The "preliminary" radiology reports from these imaging studies made available to ED clinicians at the time of diagnosis were adjudicated as positive or negative for PE by the clinical team. The reports were then divided into development, internal validation, and temporal validation cohorts in order to train, test, and validate an natural language processing model that could identify the presence of PE based on unstructured text. For risk stratification, patient- and encounter-level data elements were curated from the electronic health record and used to compute a real-time simplified pulmonary embolism severity (sPESI) score at the time of diagnosis. Chart abstraction was performed on all low-risk PE patients admitted for inpatient management. RESULTS: When applied to the internal validation and temporal validation cohorts, the natural language processing model identified the presence of PE from radiology reports with an area under the receiver operating characteristic curve of 0.99, sensitivity of 0.86 to 0.87, and specificity of 0.99. Across cohorts, 10.5% of PE-CTPA studies were positive for PE, of which 22.2% were classified as low-risk by the sPESI score. Of all low-risk PE patients, 74.3% were admitted for inpatient management. CONCLUSION: This study demonstrates that a natural language processing-based model utilizing real-time radiology reports can accurately identify patients with PE. Further, this model, used in combination with a validated risk stratification score (sPESI), provides a clinical decision support tool that accurately identifies patients in the ED with low-risk PE as candidates for outpatient management.

Risk of in-hospital Deterioration for Children with Single Ventricle Physiology.

Children with single ventricle physiology (SV) are at high risk of in-hospital morbidity and mortality. Identifying children at risk for deterioration may allow for earlier escalation of care and subsequently decreased mortality.We conducted a retrospective chart review of all admissions to the pediatric cardiology non-ICU service from 2014 to 2018 for children < 18 years old. We defined clinical deterioration as unplanned transfer to the ICU or inpatient mortality. We selected children with SV by diagnosis codes and defined infants as children < 1 year old. We compared demographic, vital sign, and lab values between infants with and without a deterioration event. We evaluated vital sign and medical therapy changes before deterioration events.Among infants with SV (129 deterioration events over 225 admissions, overall 25% with hypoplastic left heart syndrome), those who deteriorated were younger (p = 0.001), had lower baseline oxygen saturation (p = 0.022), and higher baseline respiratory rate (p = 0.022), heart rate (p = 0.023), and hematocrit (p = 0.008). Median Duke Pediatric Early Warning Score increased prior to deterioration (p < 0.001). Deterioration was associated with administration of additional oxygen support (p = 0.012), a fluid bolus (p < 0.001), antibiotics (p < 0.001), vasopressor support (p = 0.009), and red blood cell transfusion (p < 0.001).Infants with SV are at high risk for deterioration. Integrating baseline and dynamic patient data from the electronic health record to identify the highest risk patients may allow for earlier detection and intervention to prevent clinical deterioration.

Respiratory Syncytial Virus During the COVID-19 Pandemic Compared to Historic Levels: A Retrospective Cohort Study of a Health System.

BACKGROUND: Surveillance in 2020-2021 showed that seasonal respiratory illnesses were below levels seen during prior seasons, with the exception of interseasonal respiratory syncytial virus (RSV). METHODS: Electronic health record data of infants aged <1 year visiting the Duke University Health System from 4 October 2015 to 28 March 2020 (pre-COVID-19) and 29 March 2020 to 30 October 2021 (COVID-19) were assessed. International Classification of Diseases-Tenth Revision (ICD-10) codes for RSV (B97.4, J12.1, J20.5, J21.0) and bronchiolitis (RSV codes plus J21.8, J21.9) were used to detail encounters in the inpatient (IP), emergency department (ED), outpatient (OP), urgent care (UC), and telemedicine (TM) settings. RESULTS: Pre-COVID-19, 88% of RSV and 92% of bronchiolitis encounters were seen in ambulatory settings. During COVID-19, 94% and 93%, respectively, occurred in ambulatory settings. Pre-COVID-19, the highest RSV proportion was observed in December-January (up to 38% in ED), while the peaks during COVID-19 were seen in July-September (up to 41% in ED) across all settings. RSV laboratory testing among RSV encounters was low during pre-COVID-19 (IP, 51%; ED, 51%; OP, 41%; UC, 84%) and COVID-19 outside of UC (IP, 33%; ED, 47%; OP, 47%; UC, 87%). Full-term, otherwise healthy infants comprised most RSV encounters (pre-COVID-19, up to 57% in OP; COVID-19, up to 82% in TM). CONCLUSIONS: With the interruption of historical RSV epidemiologic trends and the emergence of interseasonal disease during COVID-19, continued monitoring of RSV is warranted across all settings as the changing RSV epidemiology could affect the distribution of health care resources and public health policy.

Chemoradiation treatment patterns among United States Veteran Health Administration patients with unresectable stage III non-small cell lung cancer.

BACKGROUND: The Veterans Health Administration (VHA) is the largest integrated health care system in the United States (US). Among VHA patients, the rate of use of concurrent chemoradiation therapy (CCRT) among those with unresectable, stage III non-small cell lung cancer (NSCLC) is unknown. The objective was to report recent CCRT treatment patterns in VHA patients and identify characteristics associated with receipt of CCRT. METHODS: Using Department of Veteran Affairs (VA) Cancer Registry System data linked to VA electronic medical records, we determined rates of CCRT, sequential CRT (SCRT), radiation therapy (RT) only, chemotherapy (CT) only, and neither treatment. RESULTS: Among 4054 VHA patients who met study criteria, CCRT rates slightly increased from 44 to 50% between 2013 and 2017. Factors associated with decreased odds of CCRT receipt compared to any other treatment included increasing age (adjusted odds ratio [aOR] per 10 years = 0.67; 95% CI: 0.60-0.76) and Charlson-Deyo comorbidity score (aOR = 0.94; 95% CI: 0.91-0.97). White race was associated with increased odds of CCRT receipt (aOR = 1.24; 95% CI: 1.004-1.53). In a chart review sample of 200 patients, less than half (n = 85) had a documented reason for not receiving CCRT. Among these, 29% declined treatment, and 71% did not receive CCRT due to "not being a candidate" for reasons related to frailty or lung nodules being too far apart for radiation therapy. CONCLUSIONS: CCRT rates among VHA patients with unresectable, stage III NSCLC slightly increased from 2013 to 2017; however in 2017, only half were receiving CCRT. Older patients and those with multiple comorbidities were less likely to receive CCRT and even when controlling for these factors, non-white patients were less likely to receive CCRT.

New insights into Hoogsteen base pairs in DNA duplexes from a structure-based survey.

Hoogsteen (HG) base pairs (bps) provide an alternative pairing geometry to Watson-Crick (WC) bps and can play unique functional roles in duplex DNA. Here, we use structural features unique to HG bps (syn purine base, HG hydrogen bonds and constricted C1'-C1' distance across the bp) to search for HG bps in X-ray structures of DNA duplexes in the Protein Data Bank. The survey identifies 106 A•T and 34 G•C HG bps in DNA duplexes, many of which are undocumented in the literature. It also uncovers HG-like bps with syn purines lacking HG hydrogen bonds or constricted C1'-C1' distances that are analogous to conformations that have been proposed to populate the WC-to-HG transition pathway. The survey reveals HG preferences similar to those observed for transient HG bps in solution by nuclear magnetic resonance, including stronger preferences for A•T versus G•C bps, TA versus GG steps, and also suggests enrichment at terminal ends with a preference for 5'-purine. HG bps induce small local perturbations in neighboring bps and, surprisingly, a small but significant degree of DNA bending (∼14°) directed toward the major groove. The survey provides insights into the preferences and structural consequences of HG bps in duplex DNA.

Molprobity's ultimate rotamer-library distributions for model validation.

Here we describe the updated MolProbity rotamer-library distributions derived from an order-of-magnitude larger and more stringently quality-filtered dataset of about 8000 (vs. 500) protein chains, and we explain the resulting changes and improvements to model validation as seen by users. To include only side-chains with satisfactory justification for their given conformation, we added residue-specific filters for electron-density value and model-to-density fit. The combined new protocol retains a million residues of data, while cleaning up false-positive noise in the multi- χ datapoint distributions. It enables unambiguous characterization of conformational clusters nearly 1000-fold less frequent than the most common ones. We describe examples of local interactions that favor these rare conformations, including the role of authentic covalent bond-angle deviations in enabling presumably strained side-chain conformations. Further, along with favored and outlier, an allowed category (0.3-2.0% occurrence in reference data) has been added, analogous to Ramachandran validation categories. The new rotamer distributions are used for current rotamer validation in MolProbity and PHENIX, and for rotamer choice in PHENIX model-building and refinement. The multi-dimensional χ distributions and Top8000 reference dataset are freely available on GitHub. These rotamers are termed "ultimate" because data sampling and quality are now fully adequate for this task, and also because we believe the future of conformational validation should integrate side-chain with backbone criteria. Proteins 2016; 84:1177-1189. © 2016 Wiley Periodicals, Inc.

The crystal structure of Mtr4 reveals a novel arch domain required for rRNA processing.

The essential RNA helicase, Mtr4, performs a critical role in RNA processing and degradation as an activator of the nuclear exosome. The molecular basis for this vital function is not understood and detailed analysis is significantly limited by the lack of structural data. In this study, we present the crystal structure of Mtr4. The structure reveals a new arch-like domain that is specific to Mtr4 and Ski2 (the cytosolic homologue of Mtr4). In vivo and in vitro analyses demonstrate that the Mtr4 arch domain is required for proper 5.8S rRNA processing, and suggest that the arch functions independently of canonical helicase activity. In addition, extensive conservation along the face of the putative RNA exit site highlights a potential interface with the exosome. These studies provide a molecular framework for understanding fundamental aspects of helicase function in exosome activation, and more broadly define the molecular architecture of Ski2-like helicases.

EGFR mutation testing and TKI treatment patterns among veterans with stage III and IV non-small cell lung cancer.

BACKGROUND: Epidermal growth factor receptor (EGFR) mutation testing is recommended in metastatic non-small cell lung cancer (NSCLC). The objective of this study was to assess changes in EGFR mutation testing patterns and tyrosine kinase inhibitor (TKI) use in US veterans with stage III-IV NSCLC between 2013 and 2017. PATIENTS AND METHODS: Retrospective study using linked data from Department of Veterans Affairs (VA) Cancer Registry System, Corporate Data Warehouse, commercial laboratories, and clinical notes. Generalized linear mixed models accounting for clustering by VA facility were used to determine factors associated with EGFR mutation testing. RESULTS: From 2013 to 2017, EGFR mutation testing increased from 29.5% to 38.4% among veterans with stage III-IV NSCLC and from 47.0% to 57.4% among veterans with stage IV non-squamous disease. Factors associated with increased odds of testing included being married, Medicare enrollment, and adenocarcinoma histology. Factors associated with decreased odds of testing included Medicaid eligibility, stage III disease, increasing age, being a current or former smoker, increasing Charlson-Deyo comorbidity score, and receiving cancer care in the South. Appropriate use of a TKI rose from 2013 to 2017 (17.2% to 74.1%). CONCLUSION: EGFR mutation testing rates increased to almost 60% in the stage IV non-squamous NSCLC population in 2017, with residual opportunity for further increase. Several sociodemographic characteristics, comorbidities, and geographic regions were associated with EGFR mutation testing suggestive of inequitable testing decisions. Appropriate use of TKI improved drastically from 2013 to 2017 demonstrating rapidly changing practice patterns through the adoption phase of new treatment options.

Barriers to Prescribing Targeted Therapies for Patients With NSCLC With Highly Actionable Gene Variants in the Veterans Affairs National Precision Oncology Program.

PURPOSE: Next-generation sequencing (NGS) gene panels are frequently completed for patients with advanced non-small-cell lung cancer (NSCLC). Patients with highly actionable gene variants have improved outcomes and reduced toxicities with the use of corresponding targeted agents. We sought to identify barriers to targeted agent use within the Veterans Health Affairs' National Precision Oncology Program (NPOP). METHODS: A retrospective evaluation of patients with NSCLC who underwent NGS multigene panels through NPOP between July 2015 and February 2019 was conducted. Patients who were assigned level 1 or 2A evidence for oncogenic gene variants by an artificial intelligence offering (IBM Watson for Genomics [WfG]) and NPOP staff were selected. Antineoplastic drug prescriptions and provider notes were reviewed. Reasons for withholding targeted treatments were categorized. RESULTS: Of 1,749 patients with NSCLC who successfully underwent NGS gene panel testing, 112 (6.4%) patients were assigned level 1 and/or 2A evidence for available targeted treatments by WfG and NPOP staff. All highly actionable gene variants were within ALK, BRAF, EGFR, ERBB2, MET, RET, and ROS1. Of these, 36 (32.1%) patients were not prescribed targeted agents. The three most common reasons were (1) patient did not carry a diagnosis of metastatic disease (33.3%), (2) treating provider did not comment on the NGS results (25.0%), and (3) provider felt that patient could not tolerate therapy (19.4%). No patients were denied access to level 1 or 2A targeted drugs because of rejection of a nonformulary drug request. CONCLUSION: A substantial minority of patients with NSCLC bearing highly actionable gene variants are not prescribed targeted agents. Further provider- and pathologist-directed educational efforts and implementation of health informatics systems to provide real-time decision support for test ordering and interpretation are needed.

Comparison of Annotation Services for Next-Generation Sequencing in a Large-Scale Precision Oncology Program.

PURPOSE: Next-generation sequencing (NGS) multigene panel testing has become widespread, including the Veterans Affairs (VA), through the VA National Precision Oncology Program (NPOP). The interpretation of genomic alterations remains a bottleneck for realizing precision medicine. We sought to examine the concordance for pathogenicity determination and clinical actionability of annotation services in NPOP. METHODS: Unique gene variants were generated from NGS gene panel results using two sequencing services. For each unique gene variant, annotations were provided through N-of-One (NoO), IBM Watson for Genomics (WfG), and OncoKB. Annotations for pathogenicity (all three sources) and actionability (WfG and OncoKB) were examined for concordance. Cohen's kappa statistic was calculated to measure agreement between annotation services. RESULTS: Among 1,227 NGS results obtained between 2015 and 2017, 1,388 unique variants were identified in 117 genes. The genes with the largest number of variants included TP53 (270), STK11 (92), and CDKN2A (81). The most common cancer type was lung adenocarcinoma (440), followed by colon adenocarcinoma (113). For pathogenic and likely pathogenic variants, there was 30% agreement between WfG and NoO (kappa, -0.26), 76% agreement between WfG and OncoKB (kappa, 0.22), and 42% agreement between NoO and OncoKB (kappa, -0.07). For level 1 drug actionability of gene variant-diagnosis combinations, there was moderate agreement between WfG and OncoKB (96.9%; kappa, 0.44), with 27 combinations identified as level 1 by both services, 58 by WfG alone, and 6 variants by OncoKB alone. CONCLUSION: There is substantial variability in pathogenicity assessment of NGS variants in solid tumors by annotation services. In addition, there was only moderate agreement in level 1 therapeutic actionability recommendations between WfG and OncoKB. Improvement in the precision of NGS multigene panel annotation is needed.

Medical oncologists' perspectives of the Veterans Affairs National Precision Oncology Program.

BACKGROUND: To support the rising need for testing and to standardize tumor DNA sequencing practices within the U.S. Department of Veterans Affairs (VA)'s Veterans Health Administration (VHA), the National Precision Oncology Program (NPOP) was launched in 2016. We sought to assess oncologists' practices, concerns, and perceptions regarding Next-Generation Sequencing (NGS) and the NPOP. MATERIALS AND METHODS: Using a purposive total sampling approach, oncologists who had previously ordered NGS for at least one tumor sample through the NPOP were invited to participate in semi-structured interviews. Questions assessed the following: expectations for the NPOP, procedural requirements, applicability of testing results, and the summative utility of the NPOP. Interviews were assessed using an open coding approach. Thematic analysis was conducted to evaluate the completed codebook. Themes were defined deductively by reviewing the direct responses to interview questions as well as inductively by identifying emerging patterns of data. RESULTS: Of the 105 medical oncologists who were invited to participate, 20 (19%) were interviewed from 19 different VA medical centers in 14 states. Five recurrent themes were observed: (1) Educational Efforts Regarding Tumor DNA Sequencing Should be Undertaken, (2) Pathology Departments Share a Critical Role in Facilitating Test Completion, (3) Tumor DNA Sequencing via NGS Serves as the Most Comprehensive Testing Modality within Precision Oncology, (4) The Availability of the NPOP Has Expanded Options for Select Patients, and (5) The Completion of Tumor DNA Sequencing through the NPOP Could Help Improve Research Efforts within VHA Oncology Practices. CONCLUSION: Medical oncologists believe that the availability of tumor DNA sequencing through the NPOP could potentially lead to an improvement in outcomes for veterans with metastatic solid tumors. Efforts should be directed toward improving oncologists' understanding of sequencing, strengthening collaborative relationships between oncologists and pathologists, and assessing the role of comprehensive NGS panels within the battery of precision tests.

Genomic Analysis of Metastatic Solid Tumors in Veterans: Findings From the VHA National Precision Oncology Program.

PURPOSE: The Veterans Health Administration (VHA) is the largest cancer care provider in the United States, with the added challenge of serving more than twice the percentage of patients with cancer in rural areas than the national average. The VHA established the National Precision Oncology Program in 2016 to implement and standardize the practice of precision oncology across the diverse VHA system. METHODS: Tumor or peripheral blood specimens from veterans with advanced solid tumors who were eligible for treatment were submitted for next-generation sequencing (NGS) at two commercial laboratories. Annotated results were generated by the laboratories and independently using IBM Watson for Genomics. Levels-of-evidence treatment recommendations were based on OncoKB criteria. RESULTS: From July 2016 to June 2018, 3,698 samples from 72 VHA facilities were submitted for NGS testing, of which 3,182 samples (86%) were successfully sequenced. Most samples came from men with lung, prostate, and colorectal cancers. Thirty-four percent of samples were from patients who lived in a rural area. TP53, ATM, and KRAS were among the most commonly mutated genes. Approximately 70% of samples had at least one actionable mutation, with clinical trials identified as the recommended option in more than 50%. Mutations in genes associated with a neuroendocrine prostate cancer phenotype were expressed at increased frequency among veterans than in the general population. The most frequent therapies prescribed in response to NGS testing were immune checkpoint inhibitors, EGFR kinase inhibitors, and PARP inhibitors. CONCLUSION: Clinical implementation of precision oncology is feasible across the VHA health care system, including rural sites. Veterans have unique occupational exposures that might inform the nature of the mutational signatures identified here. Importantly, these results underscore the importance of increasing clinical trial availability to veterans.

Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix.

Diffraction (X-ray, neutron and electron) and electron cryo-microscopy are powerful methods to determine three-dimensional macromolecular structures, which are required to understand biological processes and to develop new therapeutics against diseases. The overall structure-solution workflow is similar for these techniques, but nuances exist because the properties of the reduced experimental data are different. Software tools for structure determination should therefore be tailored for each method. Phenix is a comprehensive software package for macromolecular structure determination that handles data from any of these techniques. Tasks performed with Phenix include data-quality assessment, map improvement, model building, the validation/rebuilding/refinement cycle and deposition. Each tool caters to the type of experimental data. The design of Phenix emphasizes the automation of procedures, where possible, to minimize repetitive and time-consuming manual tasks, while default parameters are chosen to encourage best practice. A graphical user interface provides access to many command-line features of Phenix and streamlines the transition between programs, project tracking and re-running of previous tasks.

Model validation: local diagnosis, correction and when to quit.

Traditionally, validation was considered to be a final gatekeeping function, but refinement is smoother and results are better if model validation actively guides corrections throughout structure solution. This shifts emphasis from global to local measures: primarily geometry, conformations and sterics. A fit into the wrong local minimum conformation usually produces outliers in multiple measures. Moving to the right local minimum should be prioritized, rather than small shifts across arbitrary borderlines. Steric criteria work best with all explicit H atoms. `Backrub' motions should be used for side chains and `P-perp' diagnostics to correct ribose puckers. A `water' may actually be an ion, a relic of misfitting or an unmodeled alternate. Beware of wishful thinking in modeling ligands. At high resolution, internally consistent alternate conformations should be modeled and geometry in poor density should not be downweighted. At low resolution, CaBLAM should be used to diagnose protein secondary structure and ERRASER to correct RNA backbone. All atoms should not be forced inside density, beware of sequence misalignment, and very rare conformations such as cis-non-Pro peptides should be avoided. Automation continues to improve, but the crystallographer still must look at each outlier, in the context of density, and correct most of them. For the valid few with unambiguous density and something that is holding them in place, a functional reason should be sought. The expectation is a few outliers, not zero.

MolProbity: More and better reference data for improved all-atom structure validation.

This paper describes the current update on macromolecular model validation services that are provided at the MolProbity website, emphasizing changes and additions since the previous review in 2010. There have been many infrastructure improvements, including rewrite of previous Java utilities to now use existing or newly written Python utilities in the open-source CCTBX portion of the Phenix software system. This improves long-term maintainability and enhances the thorough integration of MolProbity-style validation within Phenix. There is now a complete MolProbity mirror site at http://molprobity.manchester.ac.uk. GitHub serves our open-source code, reference datasets, and the resulting multi-dimensional distributions that define most validation criteria. Coordinate output after Asn/Gln/His "flip" correction is now more idealized, since the post-refinement step has apparently often been skipped in the past. Two distinct sets of heavy-atom-to-hydrogen distances and accompanying van der Waals radii have been researched and improved in accuracy, one for the electron-cloud-center positions suitable for X-ray crystallography and one for nuclear positions. New validations include messages at input about problem-causing format irregularities, updates of Ramachandran and rotamer criteria from the million quality-filtered residues in a new reference dataset, the CaBLAM Cα-CO virtual-angle analysis of backbone and secondary structure for cryoEM or low-resolution X-ray, and flagging of the very rare cis-nonProline and twisted peptides which have recently been greatly overused. Due to wide application of MolProbity validation and corrections by the research community, in Phenix, and at the worldwide Protein Data Bank, newly deposited structures have continued to improve greatly as measured by MolProbity's unique all-atom clashscore.

Mismodeled purines: implicit alternates and hidden Hoogsteens.

Hoogsteen base pairs are seen in DNA crystal structures, but only rarely. This study tests whether Hoogsteens or other syn purines are either under-modeled or over-modeled, which are known problems for rare conformations. Candidate purines needing a syn/anti 180° flip were identified by diagnostic patterns of difference electron-density peaks. Manual inspection narrowed 105 flip candidates to 20 convincing cases, all at ≤2.7 Šresolution. Rebuilding and refinement confirmed that 14 of these were authentic purine flips. Seven examples are modeled as Watson-Crick base pairs but should be Hoogsteens (commonest at duplex termini), and three had the opposite issue. Syn/anti flips were also needed for some single-stranded purines. Five of the 20 convincing cases arose from an unmodeled alternate duplex running in the opposite direction. These are in semi-palindromic DNA sequences bound by a homodimeric protein and show flipped-purine-like difference peaks at residues where the palindrome is imperfect. This study documents types of incorrect modeling which are worth avoiding. However, the primary conclusions are that such mistakes are infrequent, the bias towards fitting anti purines is very slight, and the occurrence rate of Hoogsteen base pairs in DNA crystal structures remains unchanged from earlier estimates at ∼0.3%.