Using large language models for safety-related table summarization in clinical study reports.

Objectives: The generation of structured documents for clinical trials is a promising application of large language models (LLMs). We share opportunities, insights, and challenges from a competitive challenge that used LLMs for automating clinical trial documentation. Materials and Methods: As part of a challenge initiated by Pfizer (organizer), several teams (participant) created a pilot for generating summaries of safety tables for clinical study reports (CSRs). Our evaluation framework used automated metrics and expert reviews to assess the quality of AI-generated documents. Results: The comparative analysis revealed differences in performance across solutions, particularly in factual accuracy and lean writing. Most participants employed prompt engineering with generative pre-trained transformer (GPT) models. Discussion: We discuss areas for improvement, including better ingestion of tables, addition of context and fine-tuning. Conclusion: The challenge results demonstrate the potential of LLMs in automating table summarization in CSRs while also revealing the importance of human involvement and continued research to optimize this technology.

The impact of immunosuppressive agents on immune checkpoint inhibitor efficacy in patients with advanced melanoma: A real-world, multicenter, retrospective study.

BACKGROUND: Immune-related adverse events (irAEs) associated with immune checkpoint inhibitors (ICIs) are often managed via immunosuppressive agents (ISAs); however, their impact on ICI efficacy is not well studied. The impact of the use of ISAs on ICI efficacy in patients with advanced melanoma was therefore investigated. METHODS: This is a real-world, multicenter, retrospective cohort study of patients with advanced melanoma who received ICIs (n = 370). Overall survival (OS) and time to treatment failure (TTF) from the time of ICI initiation were compared among patients in subgroups of interest by unadjusted and 12-week landmark sensitivity-adjusted analyses. The association of irAEs and their management with OS and TTF were evaluated using univariate and multivariable Cox proportional hazards regression models. RESULTS: Overall, irAEs of any grade and of grade ≥3 occurred in 57% and 23% of patients, respectively. Thirty-seven percent of patients received steroids, and 3% received other ISAs. Median OS was longest among patients receiving both (not reached [NR]), shorter among those receiving only systemic steroids (SSs) (84.2 months; 95% CI, 40.2 months to NR), and shortest among those who did not experience irAEs (10.3 months; 95% CI, 6-20.1 months) (p < .001). Longer OS was significantly associated with the occurrence of irAEs and the use of SSs with or without ISAs upon multivariable-adjusted analysis (p < .001). Similar results were noted with anti-programmed death 1 (PD-1) monotherapy and combination anti-PD-1 plus anti-cytotoxic T-lymphocyte antigen 4 (CTLA-4) therapy, and with 12-week landmark sensitivity analysis (p = .01). CONCLUSIONS: These findings in patients with melanoma who were treated with ICIs suggest that the use of SSs or ISAs for the management of irAEs is not associated with inferior disease outcomes, which supports the use of these agents when necessary.

Patient-Reported Data Augment Prediction Models of Persistent Opioid Use after Elective Upper Extremity Surgery.

BACKGROUND: Opioids play a role in pain management after surgery, but prolonged use contributes to developing opioid use disorder. Identifying patients at risk of prolonged use is critical for deploying interventions that reduce or avoid opioids; however, available predictive models do not incorporate patient-reported data (PRD), and it remains unclear whether PRD can predict postoperative use behavior. The authors used a machine learning approach leveraging preoperative PRD and electronic health record data to predict persistent opioid use after upper extremity surgery. METHODS: Included patients underwent upper extremity surgery, completed preoperative PRD questionnaires, and were prescribed opioids after surgery. The authors trained models using a 2018 cohort and tested in a 2019 cohort. Opioid use was determined by patient report and filled prescriptions up to 6 months after surgery. The authors assessed model performance using area under the receiver operating characteristic, sensitivity, specificity, and Brier score. RESULTS: Among 1656 patients, 19% still used opioids at 6 weeks, 11% at 3 months, and 9% at 6 months. The XGBoost model trained on PRD plus electronic health record data achieved area under the receiver operating characteristic 0.73 at 6 months. Factors predictive of prolonged opioid use included income; education; tobacco, drug, or alcohol abuse; cancer; depression; and race. Protective factors included preoperative Patient-Reported Outcomes Measurement Information System Global Physical Health and Upper Extremity scores. CONCLUSIONS: This opioid use prediction model using preintervention data had good discriminative performance. PRD variables augmented electronic health record-based machine learning algorithms in predicting postsurgical use behaviors and were some of the strongest predictors. PRD should be used in future efforts to guide proper opioid stewardship. CLINICAL QUESTION/LEVEL OF EVIDENCE: Risk, III.

Exploring approaches for predictive cancer patient digital twins: Opportunities for collaboration and innovation.

We are rapidly approaching a future in which cancer patient digital twins will reach their potential to predict cancer prevention, diagnosis, and treatment in individual patients. This will be realized based on advances in high performance computing, computational modeling, and an expanding repertoire of observational data across multiple scales and modalities. In 2020, the US National Cancer Institute, and the US Department of Energy, through a trans-disciplinary research community at the intersection of advanced computing and cancer research, initiated team science collaborative projects to explore the development and implementation of predictive Cancer Patient Digital Twins. Several diverse pilot projects were launched to provide key insights into important features of this emerging landscape and to determine the requirements for the development and adoption of cancer patient digital twins. Projects included exploring approaches to using a large cohort of digital twins to perform deep phenotyping and plan treatments at the individual level, prototyping self-learning digital twin platforms, using adaptive digital twin approaches to monitor treatment response and resistance, developing methods to integrate and fuse data and observations across multiple scales, and personalizing treatment based on cancer type. Collectively these efforts have yielded increased insights into the opportunities and challenges facing cancer patient digital twin approaches and helped define a path forward. Given the rapidly growing interest in patient digital twins, this manuscript provides a valuable early progress report of several CPDT pilot projects commenced in common, their overall aims, early progress, lessons learned and future directions that will increasingly involve the broader research community.

Correction: Zhang et al. Association between Body Mass Index and Immune-Related Adverse Events (irAEs) among Advanced-Stage Cancer Patients Receiving Immune Checkpoint Inhibitors: A Pan-Cancer Analysis. Cancers 2021, 13, 6109.

Subha Madhavan (S.M.) and Anas Belouali (A.B.) were not included as authors in the published article [...].

Real-World Studies Link Nonsteroidal Anti-inflammatory Drug Use to Improved Overall Lung Cancer Survival.

Inflammation is a cancer hallmark. Nonsteroidal anti-inflammatory drugs (NSAIDs) improve overall survival (OS) in certain cancers. Real-world studies explored here if NSAIDs improve non-small cell lung cancer (NSCLC) OS. Analyses independently interrogated clinical databases from The University of Texas MD Anderson Cancer Center (MDACC cohort, 1987 to 2015; 33,162 NSCLCs and 3,033 NSAID users) and Georgetown-MedStar health system (Georgetown cohort, 2000 to 2019; 4,497 NSCLCs and 1,993 NSAID users). Structured and unstructured clinical data were extracted from electronic health records (EHRs) using natural language processing (NLP). Associations were made between NSAID use and NSCLC prognostic features (tobacco use, gender, race, and body mass index, BMI). NSAIDs were statistically-significantly (P < 0.0001) associated with increased NSCLC survival (5-year OS 29.7% for NSAID users versus 13.1% for non-users) in the MDACC cohort. NSAID users gained 11.6 months over nonusers in 5-year restricted mean survival time. Stratified analysis by stage, histopathology and multicovariable assessment substantiated benefits. NSAID users were pooled independent of NSAID type and by NSAID type. Landmark analysis excluded immortal time bias. Survival improvements (P < 0.0001) were confirmed in the Georgetown cohort. Thus, real-world NSAID usage was independently associated with increased NSCLC survival in the MDACC and Georgetown cohorts. Findings were confirmed by landmark analyses and NSAID type. The OS benefits persisted despite tobacco use and did not depend on gender, race, or BMI (MDACC cohort, P < 0.0001). These real-world findings could guide future NSAID lung cancer randomized trials.

Enhancing the REMBRANDT MRI collection with expert segmentation labels and quantitative radiomic features.

Malignancy of the brain and CNS is unfortunately a common diagnosis. A large subset of these lesions tends to be high grade tumors which portend poor prognoses and low survival rates, and are estimated to be the tenth leading cause of death worldwide. The complex nature of the brain tissue environment in which these lesions arise offers a rich opportunity for translational research. Magnetic Resonance Imaging (MRI) can provide a comprehensive view of the abnormal regions in the brain, therefore, its applications in the translational brain cancer research is considered essential for the diagnosis and monitoring of disease. Recent years has seen rapid growth in the field of radiogenomics, especially in cancer, and scientists have been able to successfully integrate the quantitative data extracted from medical images (also known as radiomics) with genomics to answer new and clinically relevant questions. In this paper, we took raw MRI scans from the REMBRANDT data collection from public domain, and performed volumetric segmentation to identify subregions of the brain. Radiomic features were then extracted to represent the MRIs in a quantitative yet summarized format. This resulting dataset now enables further biomedical and integrative data analysis, and is being made public via the NeuroImaging Tools & Resources Collaboratory (NITRC) repository ( https://www.nitrc.org/projects/rembrandt_brain/ ).

A research agenda to support the development and implementation of genomics-based clinical informatics tools and resources.

OBJECTIVE: The Genomic Medicine Working Group of the National Advisory Council for Human Genome Research virtually hosted its 13th genomic medicine meeting titled "Developing a Clinical Genomic Informatics Research Agenda". The meeting's goal was to articulate a research strategy to develop Genomics-based Clinical Informatics Tools and Resources (GCIT) to improve the detection, treatment, and reporting of genetic disorders in clinical settings. MATERIALS AND METHODS: Experts from government agencies, the private sector, and academia in genomic medicine and clinical informatics were invited to address the meeting's goals. Invitees were also asked to complete a survey to assess important considerations needed to develop a genomic-based clinical informatics research strategy. RESULTS: Outcomes from the meeting included identifying short-term research needs, such as designing and implementing standards-based interfaces between laboratory information systems and electronic health records, as well as long-term projects, such as identifying and addressing barriers related to the establishment and implementation of genomic data exchange systems that, in turn, the research community could help address. DISCUSSION: Discussions centered on identifying gaps and barriers that impede the use of GCIT in genomic medicine. Emergent themes from the meeting included developing an implementation science framework, defining a value proposition for all stakeholders, fostering engagement with patients and partners to develop applications under patient control, promoting the use of relevant clinical workflows in research, and lowering related barriers to regulatory processes. Another key theme was recognizing pervasive biases in data and information systems, algorithms, access, value, and knowledge repositories and identifying ways to resolve them.

Standardized evidence-based approach for assessment of oncogenic and clinical significance of NTRK fusions.

Gene fusions involving the neurotrophic receptor tyrosine kinase genes NTRK1, NTRK2, and NTRK3, are well established oncogenic drivers in a broad range of pediatric and adult tumors. These fusions are also important actionable markers, predicting often dramatic response to FDA approved kinase inhibitors. Accurate interpretation of the clinical significance of NTRK fusions is a high priority for diagnostic laboratories, but remains challenging and time consuming given the rapid pace of new data accumulation, the diversity of fusion partners and tumor types, and heterogeneous and incomplete information in variant databases and knowledgebases. The ClinGen NTRK Fusions Somatic Cancer Variant Curation Expert Panel (SC-VCEP) was formed to systematically address these challenges and create an expert-curated resource to support clinicians, researchers, patients and their families in making accurate interpretations and informed treatment decisions for NTRK fusion-driven tumors. We describe a system for NTRK fusion interpretation (including compilation of key elements and annotations) developed by the NTRK fusions SC-VCEP. We illustrate this stepwise process on examples of LMNA::NTRK1 and KANK1::NTRK2 fusions. Finally, we provide detailed analysis of current representation of NTRK fusions in public fusion databases and the CIViC knowledgebase, performed by the NTRK fusions SC-VCEP to determine existing gaps and prioritize future curation activities.

Standards for the classification of pathogenicity of somatic variants in cancer (oncogenicity): Joint recommendations of Clinical Genome Resource (ClinGen), Cancer Genomics Consortium (CGC), and Variant Interpretation for Cancer Consortium (VICC).

PURPOSE: Several professional societies have published guidelines for the clinical interpretation of somatic variants, which specifically address diagnostic, prognostic, and therapeutic implications. Although these guidelines for the clinical interpretation of variants include data types that may be used to determine the oncogenicity of a variant (eg, population frequency, functional, and in silico data or somatic frequency), they do not provide a direct, systematic, and comprehensive set of standards and rules to classify the oncogenicity of a somatic variant. This insufficient guidance leads to inconsistent classification of rare somatic variants in cancer, generates variability in their clinical interpretation, and, importantly, affects patient care. Therefore, it is essential to address this unmet need. METHODS: Clinical Genome Resource (ClinGen) Somatic Cancer Clinical Domain Working Group and ClinGen Germline/Somatic Variant Subcommittee, the Cancer Genomics Consortium, and the Variant Interpretation for Cancer Consortium used a consensus approach to develop a standard operating procedure (SOP) for the classification of oncogenicity of somatic variants. RESULTS: This comprehensive SOP has been developed to improve consistency in somatic variant classification and has been validated on 94 somatic variants in 10 common cancer-related genes. CONCLUSION: The comprehensive SOP is now available for classification of oncogenicity of somatic variants.

Automated Identification of Patients With Immune-Related Adverse Events From Clinical Notes Using Word Embedding and Machine Learning.

PURPOSE: Although immune checkpoint inhibitors (ICIs) have substantially improved survival in patients with advanced malignancies, they are associated with a unique spectrum of side effects termed immune-related adverse events (irAEs). To ensure treatment safety, research efforts are needed to comprehensively detect and understand irAEs. Retrospective analysis of data from electronic health records can provide knowledge to characterize these toxicities. However, such information is not captured in a structured format within the electronic health record and requires manual chart review. MATERIALS AND METHODS: In this work, we propose a natural language processing pipeline that can automatically annotate clinical notes and determine whether there is evidence that a patient developed an irAE. Seven hundred eighty-one cases were manually reviewed by clinicians and annotated for irAEs at the patient level. A dictionary of irAEs keywords was used to perform text reduction on clinical notes belonging to each patient; only sentences with relevant expressions were kept. Word embeddings were then used to generate vector representations over the reduced text, which served as input for the machine learning classifiers. The output of the models was presence or absence of any irAEs. Additional models were built to classify skin-related toxicities, endocrine toxicities, and colitis. RESULTS: The model for any irAE achieved an average F1-score = 0.75 and area under the receiver operating characteristic curve = 0.85. This outperformed a basic keyword filtering approach. Although the classifier of any irAEs achieved good accuracy, individual irAE classification still has room for improvement. CONCLUSION: We demonstrate that patient-level annotations combined with a machine learning approach using keywords filtering and word embeddings can achieve promising accuracy in classifying irAEs in clinical notes. This model may facilitate annotation and analysis of large irAEs data sets.

High mRNA expression of LY6 gene family is associated with overall survival outcome in pancreatic ductal adenocarcinoma.

Pancreatic cancer ranks one of the worst in overall survival outcome with a 5 year survival rate being less than 10%. Pancreatic cancer faces unique challenges in its diagnosis and treatment, such as the lack of clinically validated biomarkers and the immensely immunosuppressive tumor microenvironment. Recently, the LY6 gene family has received increasing attention for its multi-faceted roles in cancer development, stem cell maintenance, immunomodulation, and association with more aggressive and hard-to-treat cancers. A detailed study of mRNA expression of LY6 gene family and its association with overall survival (OS) outcome in pancreatic cancers is lacking. We used publicly available clinical datasets to analyze the mRNA expression of a set of LY6 genes and its effect on OS outcome in the context of the tumor microenvironment and immunomodulation. We used web-based tools Kaplan-Meier Plotter, cBioPortal, Oncomine and R-programming to analyze copy number alterations, mRNA expression and its association with OS outcome in pancreatic cancer. These analyses demonstrated that high expression of LY6 genes is associated with OS and disease free survival (DFS) outcome. High expression of LY6 genes and their association with OS outcome is dependent on the composition of tumor microenvironment. Considering that LY6 proteins are anchored to the outer cell membrane or secreted, making them readily accessible, these findings highlight the potential of LY6 family members in the future of pancreatic cancer diagnosis and treatment.

Recommendations for the safe, effective use of adaptive CDS in the US healthcare system: an AMIA position paper.

The development and implementation of clinical decision support (CDS) that trains itself and adapts its algorithms based on new data-here referred to as Adaptive CDS-present unique challenges and considerations. Although Adaptive CDS represents an expected progression from earlier work, the activities needed to appropriately manage and support the establishment and evolution of Adaptive CDS require new, coordinated initiatives and oversight that do not currently exist. In this AMIA position paper, the authors describe current and emerging challenges to the safe use of Adaptive CDS and lay out recommendations for the effective management and monitoring of Adaptive CDS.

Use of electronic health records to support a public health response to the COVID-19 pandemic in the United States: a perspective from 15 academic medical centers.

Our goal is to summarize the collective experience of 15 organizations in dealing with uncoordinated efforts that result in unnecessary delays in understanding, predicting, preparing for, containing, and mitigating the COVID-19 pandemic in the US. Response efforts involve the collection and analysis of data corresponding to healthcare organizations, public health departments, socioeconomic indicators, as well as additional signals collected directly from individuals and communities. We focused on electronic health record (EHR) data, since EHRs can be leveraged and scaled to improve clinical care, research, and to inform public health decision-making. We outline the current challenges in the data ecosystem and the technology infrastructure that are relevant to COVID-19, as witnessed in our 15 institutions. The infrastructure includes registries and clinical data networks to support population-level analyses. We propose a specific set of strategic next steps to increase interoperability, overall organization, and efficiencies.

De novo assembly and annotation of transcriptomes from two cultivars of Cannabis sativa with different cannabinoid profiles.

Cannabis has been cultivated for millennia for medicinal, industrial and recreational uses. Our long-term goal is to compare the transcriptomes of cultivars with different cannabinoid profiles for therapeutic purposes. Here we describe the de novo assembly, annotation and initial analysis of two cultivars of Cannabis, a high THC variety and a CBD plus THC variety. Cultivars were grown under different lighting conditions; flower buds were sampled over 71 days. Cannabinoid profiles were determined by ESI-LC/MS. RNA samples were sequenced using the HiSeq4000 platform. Transcriptomes were assembled using the DRAP pipeline and annotated using the BLAST2GO pipeline and other tools. Each transcriptome contained over twenty thousand protein encoding transcripts with ORFs and flanking sequence. Identification of transcripts for cannabinoid pathway and related enzymes showed full-length ORFs that align with the draft genomes of the Purple Kush and Finola cultivars. Two transcripts were found for olivetolic acid cyclase (OAC) that mapped to distinct locations on the Purple Kush genome suggesting multiple genes for OAC are expressed in some cultivars. The ability to make high quality annotated reference transcriptomes in Cannabis or other plants can promote rapid comparative analysis between cultivars and growth conditions in Cannabis and other organisms without annotated genome assemblies.