We followed a systematic approach based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses to identify existing clinical natural language processing (NLP) systems that generate structured information from unstructured free text. Seven literature databases were searched with a query combining the concepts of natural language processing and structured data capture. Two reviewers screened all records for relevance during two screening phases, and information about clinical NLP systems was collected from the final set of papers. A total of 7149 records (after removing duplicates) were retrieved and screened, and 86 were determined to fit the review criteria. These papers contained information about 71 different clinical NLP systems, which were then analyzed. The NLP systems address a wide variety of important clinical and research tasks. Certain tasks are well addressed by the existing systems, while others remain as open challenges that only a small number of systems attempt, such as extraction of temporal information or normalization of concepts to standard terminologies. This review has identified many NLP systems capable of processing clinical free text and generating structured output, and the information collected and evaluated here will be important for prioritizing development of new approaches for clinical NLP.
The use and benefit of adjuvant chemotherapy to treat stage II colorectal cancer (CRC) patients is not well understood since the majority of these patients are cured by surgery alone. Identification of biological markers of relapse is a critical challenge to effectively target treatments to the ~20% of patients destined to relapse. We have integrated molecular profiling results of several “omics” data types to determine the most reliable prognostic biomarkers for relapse in CRC using data from 40 stage I and II CRC patients. We identified 31 multi-omics features that highly correlate with relapse. The data types were integrated using multi-step analytical approach with consecutive elimination of redundant molecular features. For each data type a systems biology analysis was performed to identify pathways biological processes and disease categories most affected in relapse. The biomarkers detected in tumors urine and blood of patients indicated a strong association with immune processes including aberrant regulation of T-cell and B-cell activation that could lead to overall differences in lymphocyte recruitment for tumor infiltration and markers indicating likelihood of future relapse. The immune response was the biologically most coherent signature that emerged from our analyses among several other biological processes and corroborates other studies showing a strong immune response in patients less likely to relapse.
Background Immune checkpoint blockade (ICB) shows lasting benefits in advanced melanoma; however, not all patients respond to this treatment and many develop potentially life-threatening immune-related adverse events (irAEs). Identifying individuals who will develop irAEs is critical in order to improve the quality of care. Here, we prospectively demonstrate that the gut microbiome predicts irAEs in melanoma patients undergoing ICB. Methods Pre-, during, and post-treatment stool samples were collected from 27 patients with advanced stage melanoma treated with IPI (anti-CTLA-4) and NIVO (anti-PD1) ICB inhibitors at NYU Langone Health. We completed 16S rRNA gene amplicon sequencing, DNA deep shotgun metagenomic, and RNA-seq metatranscriptomic sequencing. The divisive amplicon denoising algorithm (DADA2) was used to process 16S data. Taxonomy for shotgun sequencing data was assigned using MetaPhlAn2, and gene pathways were assigned using HUMAnN 2.0. Compositionally aware differential expression analysis was performed using ANCOM. The Cox-proportional hazard model was used to assess the prospective role of the gut microbiome (GMB) in irAES, with adjustment for age, sex, BMI, immune ICB treatment type, and sequencing batch. Results Two natural GMB clusters with distinct community compositions were identified from the analysis of 16S rRNA data (R2 = 0.16, p < 0.001). In Cox-proportional hazard modeling, these two clusters showed a near 7-fold differential risk for developing irAEs within 1 year of initiating treatment (HR = 6.89 [95% CI: 1.33–35.58]). Using shotgun metagenomics, we further identified 37 bacterial strains differentially expressed between the risk groups, with specific dominance of Bacteroides dorei within the high-risk GMB cluster and Bacteroides vulgatus in the low-risk cluster. The high-risk cluster also appeared to have elevated expression of several functional pathways, including those associated with adenosine metabolism (all FDR < 0.05). A sub-analysis of samples (n = 10 participants) at baseline and 6 and 12 weeks after the start of treatment revealed that the microbiome remained stable over the course of treatment (R2 = 0.88, p < 0.001). Conclusions We identified two distinct fecal bacterial community clusters which are associated differentially with irAEs in ICB-treated advanced melanoma patients.
We have developed a Decision Support Environment (DSE) for medical experts at the US Food and Drug Administration (FDA). The DSE contains two integrated systems: The Event-based Text-mining of Health Electronic Records (ETHER) and the Pattern-based and Advanced Network Analyzer for Clinical Evaluation and Assessment (PANACEA). These systems assist medical experts in reviewing reports submitted to the Vaccine Adverse Event Reporting System (VAERS) and the FDA Adverse Event Reporting System (FAERS). In this manuscript, we describe the DSE architecture and key functionalities, and examine its potential contributions to the signal management process by focusing on four use cases: the identification of missing cases from a case series, the identification of duplicate case reports, retrieving cases for a case series analysis, and community detection for signal identification and characterization.
The sheer volume of textual information that needs to be reviewed and analyzed in many clinical settings requires the automated retrieval of key clinical and temporal information. The existing natural language processing systems are often challenged by the low quality of clinical texts and do not demonstrate the required performance. In this study, we focus on medical product safety report narratives and investigate the association of the clinical events with appropriate time information. We developed a novel algorithm for tagging and extracting temporal information from the narratives, and associating it with related events. The proposed algorithm minimizes the performance dependency on text quality by relying only on shallow syntactic information and primitive properties of the extracted event and time entities. We demonstrated the effectiveness of the proposed algorithm by evaluating its tagging and time assignment capabilities on 140 randomly selected reports from the US Vaccine Adverse Event Reporting System (VAERS) and the FDA (Food and Drug Administration) Adverse Event Reporting System (FAERS). We compared the performance of our tagger with the SUTime and HeidelTime taggers, and our algorithm's event-time associations with the Temporal Awareness and Reasoning Systems for Question Interpretation (TARSQI). We further evaluated the ability of our algorithm to correctly identify the time information for the events in the 2012 Informatics for Integrating Biology and the Bedside (i2b2) Challenge corpus. For the time tagging task, our algorithm performed better than the SUTime and the HeidelTime taggers (F-measure in VAERS and FAERS: Our algorithm: 0.86 and 0.88, SUTime: 0.77 and 0.74, and HeidelTime 0.75 and 0.42, respectively). In the event-time association task, our algorithm assigned an inappropriate timestamp for 25% of the events, while the TARSQI toolkit demonstrated a considerably lower performance, assigning inappropriate timestamps in 61.5% of the same events. Our algorithm also supported the correct calculation of 69% of the event relations to the section time in the i2b2 testing set.
Purpose: While immune checkpoint inhibitors (ICI) have revolutionized the treatment of cancer by producing durable antitumor responses, only 10%–30% of treated patients respond and the ability to predict clinical benefit remains elusive. Several studies, small in size and using variable analytic methods, suggest the gut microbiome may be a novel, modifiable biomarker for tumor response rates, but the specific bacteria or bacterial communities putatively impacting ICI responses have been inconsistent across the studied populations. Experimental Design: We have reanalyzed the available raw 16S rRNA amplicon and metagenomic sequencing data across five recently published ICI studies (n = 303 unique patients) using a uniform computational approach. Results: Herein, we identify novel bacterial signals associated with clinical responders (R) or nonresponders (NR) and develop an integrated microbiome prediction index. Unexpectedly, the NR-associated integrated index shows the strongest and most consistent signal using a random effects model and in a sensitivity and specificity analysis (P < 0.01). We subsequently tested the integrated index using validation cohorts across three distinct and diverse cancers (n = 105). Conclusions: Our analysis highlights the development of biomarkers for nonresponse, rather than response, in predicting ICI outcomes and suggests a new approach to identify patients who would benefit from microbiome-based interventions to improve response rates.
Organ transplant recipients are not routinely included in clinical trials, and as a result there is a paucity of data to guide clinicians in the treatment of malignancies in this unique patient population. This is a case report and focused review of the treatment of hepatocellular carcinoma in patients with orthotopic liver transplants. We describe a single patient’s treatment over a period of 4 years from the time of diagnosis to submission of this case report. We submit evidence that the anti-CTLA-4 antibody ipilimumab can produce a durable response, with a tolerable adverse event profile and without associated allograft rejection.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.