Using Computer Vision to Improve Endoscopic Disease Quantification in Therapeutic Clinical Trials of Ulcerative Colitis.

BACKGROUND & AIMS: Endoscopic assessment of ulcerative colitis (UC) typically reports only the maximum severity observed. Computer vision methods may better quantify mucosal injury detail, which varies among patients. METHODS: Endoscopic video from the UNIFI clinical trial (A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction and Maintenance Therapy in Participants With Moderately to Severely Active Ulcerative Colitis) comparing ustekinumab and placebo for UC were processed in a computer vision analysis that spatially mapped Mayo Endoscopic Score (MES) to generate the Cumulative Disease Score (CDS). CDS was compared with the MES for differentiating ustekinumab vs placebo treatment response and agreement with symptomatic remission at week 44. Statistical power, effect, and estimated sample sizes for detecting endoscopic differences between treatments were calculated using both CDS and MES measures. Endoscopic video from a separate phase 2 clinical trial replication cohort was performed for validation of CDS performance. RESULTS: Among 748 induction and 348 maintenance patients, CDS was lower in ustekinumab vs placebo users at week 8 (141.9 vs 184.3; P < .0001) and week 44 (78.2 vs 151.5; P < .0001). CDS was correlated with the MES (P < .0001) and all clinical components of the partial Mayo score (P < .0001). Stratification by pretreatment CDS revealed ustekinumab was more effective than placebo (P < .0001) with increasing effect in severe vs mild disease (-85.0 vs -55.4; P < .0001). Compared with the MES, CDS was more sensitive to change, requiring 50% fewer participants to demonstrate endoscopic differences between ustekinumab and placebo (Hedges' g = 0.743 vs 0.460). CDS performance in the JAK-UC replication cohort was similar to UNIFI. CONCLUSIONS: As an automated and quantitative measure of global endoscopic disease severity, the CDS offers artificial intelligence enhancement of traditional MES capability to better evaluate UC in clinical trials and potentially practice.

An electrocardiogram-based AI algorithm for early detection of pulmonary hypertension.

BACKGROUND: Early diagnosis of pulmonary hypertension (PH) is critical for effective treatment and management. We aimed to develop and externally validate an artificial intelligence algorithm that could serve as a PH screening tool, based on analysis of a standard 12-lead electrocardiogram (ECG). METHODS: The PH Early Detection Algorithm (PH-EDA) is a convolutional neural network developed using retrospective ECG voltage-time data, with patients classified as "PH-likely" or "PH-unlikely" (controls) based on right heart catheterisation or echocardiography. In total, 39 823 PH-likely patients and 219 404 control patients from Mayo Clinic were randomly split into training (48%), validation (12%), and test (40%) sets. ECGs taken within 1 month of PH diagnosis (diagnostic dataset) were used to train the PH-EDA at Mayo Clinic. Performance was tested on diagnostic ECGs within the test sets from Mayo Clinic (n=16 175/87 998 PH-likely/controls) and Vanderbilt University Medical Center (VUMC; n=6045/24 256 PH-likely/controls). Performance was also tested on ECGs taken 6-18 months (pre-emptive dataset), and up to 5 years prior to a PH diagnosis at both sites. RESULTS: Performance testing yielded an area under the receiver operating characteristic curve (AUC) of 0.92 and 0.88 in the diagnostic test set at Mayo Clinic and VUMC, respectively, and 0.86 and 0.81, respectively, in the pre-emptive test set. The AUC remained a minimum of 0.79 at Mayo Clinic and 0.73 at VUMC up to 5 years before diagnosis. CONCLUSION: The PH-EDA can detect PH at diagnosis and 6-18 months prior, demonstrating the potential to accelerate diagnosis and management of this debilitating disease.

An infodemiological framework for tracking the spread of SARS-CoV-2 using integrated public data.

The outbreak of the SARS-CoV-2 novel coronavirus has caused a health crisis of immeasurable magnitude. Signals from heterogeneous public data sources could serve as early predictors for infection waves of the pandemic, particularly in its early phases, when infection data was scarce. In this article, we characterize temporal pandemic indicators by leveraging an integrated set of public data and apply them to a Prophet model to predict COVID-19 trends. An effective natural language processing pipeline was first built to extract time-series signals of specific articles from a news corpus. Bursts of these temporal signals were further identified with Kleinberg's burst detection algorithm. Across different US states, correlations for Google Trends of COVID-19 related terms, COVID-19 news volume, and publicly available wastewater SARS-CoV-2 measurements with weekly COVID-19 case numbers were generally high with lags ranging from 0 to 3 weeks, indicating them as strong predictors of viral spread. Incorporating time-series signals of these effective predictors significantly improved the performance of the Prophet model, which was able to predict the COVID-19 case numbers between one and two weeks with average mean absolute error rates of 0.38 and 0.46 respectively across different states.

Cryptophane-folate biosensor for (129)xe NMR.

Folate-conjugated cryptophane was developed for targeting cryptophane to membrane-bound folate receptors that are overexpressed in many human cancers. The cryptophane biosensor was synthesized in 20 nonlinear steps, which included functionalization with folate recognition moiety, solubilizing peptide, and Cy3 fluorophore. Hyperpolarized (129)Xe NMR studies confirmed xenon binding to the folate-conjugated cryptophane. Cellular internalization of biosensor was monitored by confocal laser scanning microscopy and quantified by flow cytometry. Competitive blocking studies confirmed cryptophane endocytosis through a folate receptor-mediated pathway. Flow cytometry revealed 10-fold higher cellular internalization in KB cancer cells overexpressing folate receptors compared to HT-1080 cells with normal folate receptor expression. The biosensor was determined to be nontoxic in HT-1080 and KB cells by MTT assay at low micromolar concentrations typically used for hyperpolarized (129)Xe NMR experiments.

Measurement of radon and xenon binding to a cryptophane molecular host.

Xenon and radon have many similar properties, a difference being that all 35 isotopes of radon ((195)Rn-(229)Rn) are radioactive. Radon is a pervasive indoor air pollutant believed to cause significant incidence of lung cancer in many geographic regions, yet radon affinity for a discrete molecular species has never been determined. By comparison, the chemistry of xenon has been widely studied and applied in science and technology. Here, both noble gases were found to bind with exceptional affinity to tris-(triazole ethylamine) cryptophane, a previously unsynthesized water-soluble organic host molecule. The cryptophane-xenon association constant, K(a)=42,000 ± 2,000 M(-1) at 293 K, was determined by isothermal titration calorimetry. This value represents the highest measured xenon affinity for a host molecule. The partitioning of radon between air and aqueous cryptophane solutions of varying concentration was determined radiometrically to give the cryptophane-radon association constant K(a)=49,000 ± 12,000 M(-1) at 293 K.

Creation of an ustekinumab external control arm for Crohn's disease using electronic health records data: A pilot study.

BACKGROUND: Randomized trials are the gold-standard for clinical evidence generation, but they can sometimes be limited by infeasibility and unclear generalizability to real-world practice. External control arm (ECA) studies may help address this evidence gaps by constructing retrospective cohorts that closely emulate prospective ones. Experience in constructing these outside the context of rare diseases or cancer is limited. We piloted an approach for developing an ECA in Crohn's disease using electronic health records (EHR) data. METHODS: We queried EHR databases and manually screened records at the University of California, San Francisco to identify patients meeting the eligibility criteria of TRIDENT, a recently completed interventional trial involving an ustekinumab reference arm. We defined timepoints to balance missing data and bias. We compared imputation models by their impacts on cohort membership and outcomes. We assessed the accuracy of algorithmic data curation against manual review. Lastly, we assessed disease activity following treatment with ustekinumab. RESULTS: Screening identified 183 patients. 30% of the cohort had missing baseline data. Nonetheless, cohort membership and outcomes were robust to the method of imputation. Algorithms for ascertaining non-symptom-based elements of disease activity using structured data were accurate against manual review. The cohort consisted of 56 patients, exceeding planned enrollment in TRIDENT. 34% of the cohort was in steroid-free remission at week 24. CONCLUSION: We piloted an approach for creating an ECA in Crohn's disease from EHR data by using a combination of informatics and manual methods. However, our study reveals significant missing data when standard-of-care clinical data are repurposed. More work will be needed to improve the alignment of trial design with typical patterns of clinical practice, and thereby enable a future of more robust ECAs in chronic diseases like Crohn's disease.

Durability of the Single-Dose Ad26.COV2.S Vaccine in the Prevention of COVID-19 Infections and Hospitalizations in the US Before and During the Delta Variant Surge.

Importance: Vaccination against the SARS-CoV-2 virus is critical to control the pandemic. Randomized clinical trials demonstrated efficacy of the single-dose Ad26.COV2.S COVID-19 vaccine, but data on longer-term protection in clinical practice and effectiveness against variants are needed. Objective: To assess the association between receiving the Ad26.COV2.S vaccine and COVID-19-related infections and hospitalizations before and during the Delta variant surge. Design, Setting, and Participants: This cohort study included adults aged 18 years and older who were newly Ad26.COV2.S-vaccinated matched to as many as 10 unvaccinated individuals by date, location, age, sex, and comorbidity index. This was followed by 1:4 propensity score matching on COVID-19 risk factors. Data were collected from US insurance claims data from March 1, 2020, through August 31, 2021. Exposures: Vaccination with Ad26.COV2.S vs no vaccination. Main Outcomes and Measures: Vaccine effectiveness (VE) was estimated for recorded COVID-19 infection and COVID-19-related hospitalization, nationwide and in subgroups by age, high-risk factors, calendar time, and states with high incidences of the Delta variant. VE estimates were corrected for underrecording of vaccinations in insurance data. Results: Among 422 034 vaccinated individuals (mean [SD] age, 54.7 [17.4] years; 236 437 [56.0%] women) and 1 645 397 matched unvaccinated individuals (mean [SD] age, 54.5 [17.5] years; 922 937 [56.1%] women), VE was 76% (95% CI, 75%-77%) for COVID-19 infections and 81% (95% CI, 78%-82%) for COVID-19-related hospitalizations. VE was stable for at least 180 days after vaccination and over calendar time. Among states with high Delta variant incidence, VE during June to August 2021 was 74% (95% CI, 71%-77%) for infections and 81% (95% CI, 75%-86%) for hospitalizations. VE for COVID-19 was higher in individuals younger than 65 years (78%; 95% CI, 77%-79%) and lower in immunocompromised patients (64%; 95% CI, 59%-68%). All estimates were corrected for vaccination underrecording; uncorrected VE, which served as a lower bound, was 66% (95% CI, 64%-67%) for any recorded COVID-19 infection and 72% (95% CI, 69%-74%) for COVID-19-related hospitalization. Conclusions and Relevance: This cohort study in US clinical practice showed stable VE of Ad26.COV2.S for at least 6 months before as well as during the time the Delta variant emerged and became dominant.

Durability of Protection Post-Primary COVID-19 Vaccination in the United States.

The durability of immune responses after COVID-19 vaccination will drive long-term vaccine effectiveness across settings and may differ by vaccine type. To determine durability of protection of COVID-19 vaccines (BNT162b2, mRNA-1273, and Ad26.COV2.S) following primary vaccination in the United States, a matched case-control study was conducted in three cohorts between 1 January and 7 September 2021 using de-identified data from a database covering 168 million lives. Odds ratios (ORs) for developing outcomes of interest (breakthrough SARS-CoV-2 infection, hospitalization, or intensive care unit admission) were determined for each vaccine (no direct comparisons). In total, 17,017,435 individuals were identified. Relative to the baseline, stable protection was observed for Ad26.COV2.S against infections (OR [95% confidence interval (CI)], 1.31 [1.18-1.47]) and hospitalizations (OR [95% CI], 1.25 [0.86-1.80]). Relative to the baseline, protection waned over time against infections for BNT162b2 (OR [95% CI], 2.20 [2.01-2.40]) and mRNA-1273 (OR [95% CI], 2.07 [1.87-2.29]) and against hospitalizations for BNT162b2 (OR [95% CI], 2.38 [1.79-3.17]). Baseline protection remained stable for intensive care unit admissions for all three vaccines. Calculated baseline VE was consistent with published literature. This study suggests that the three vaccines in three separate populations may have different durability profiles.

Multiple hindered rotators in a gyroscope-inspired tribenzylamine hemicryptophane.

A gyroscope-inspired tribenzylamine hemicryptophane provides a vehicle for exploring the structure and properties of multiple p-phenylene rotators within one molecule. The hemicryptophane was synthesized in three steps in good overall yield using mild conditions. Three rotator-forming linkers were cyclized to form a rigid cyclotriveratrylene (CTV) stator framework, which was then closed with an amine. The gyroscope-like molecule was characterized by (1)H NMR and (13)C NMR spectroscopy, and the structure was solved by X-ray crystallography. The rigidity of the two-component CTV-trismethylamine stator was investigated by (1)H variable-temperature (VT) NMR experiments and molecular dynamics simulations. These techniques identified gyration of the three p-phenylene rotators on the millisecond time scale at -93 °C, with more dynamic but still hindered motion at room temperature (27 °C). The activation energy for the p-phenylene rotation was determined to be ~10 kcal mol(-1). Due to the propeller arrangement of the p-phenylenes, their rotation is hindered but not strongly correlated. The compact size, simple synthetic route, and molecular motions of this gyroscope-inspired tribenzylamine hemicryptophane make it an attractive starting point for controlling the direction and coupling of rotators within molecular systems.

Knowledge synthesis of 100 million biomedical documents augments the deep expression profiling of coronavirus receptors.

The COVID-19 pandemic demands assimilation of all biomedical knowledge to decode mechanisms of pathogenesis. Despite the recent renaissance in neural networks, a platform for the real-time synthesis of the exponentially growing biomedical literature and deep omics insights is unavailable. Here, we present the nferX platform for dynamic inference from over 45 quadrillion possible conceptual associations from unstructured text, and triangulation with insights from single-cell RNA-sequencing, bulk RNA-seq and proteomics from diverse tissue types. A hypothesis-free profiling of ACE2 suggests tongue keratinocytes, olfactory epithelial cells, airway club cells and respiratory ciliated cells as potential reservoirs of the SARS-CoV-2 receptor. We find the gut as the putative hotspot of COVID-19, where a maturation correlated transcriptional signature is shared in small intestine enterocytes among coronavirus receptors (ACE2, DPP4, ANPEP). A holistic data science platform triangulating insights from structured and unstructured data holds potential for accelerating the generation of impactful biological insights and hypotheses.

Inference from longitudinal laboratory tests characterizes temporal evolution of COVID-19-associated coagulopathy (CAC).

Temporal inference from laboratory testing results and triangulation with clinical outcomes extracted from unstructured electronic health record (EHR) provider notes is integral to advancing precision medicine. Here, we studied 246 SARS-CoV-2 PCR-positive (COVIDpos) patients and propensity-matched 2460 SARS-CoV-2 PCR-negative (COVIDneg) patients subjected to around 700,000 lab tests cumulatively across 194 assays. Compared to COVIDneg patients at the time of diagnostic testing, COVIDpos patients tended to have higher plasma fibrinogen levels and lower platelet counts. However, as the infection evolves, COVIDpos patients distinctively show declining fibrinogen, increasing platelet counts, and lower white blood cell counts. Augmented curation of EHRs suggests that only a minority of COVIDpos patients develop thromboembolism, and rarely, disseminated intravascular coagulopathy (DIC), with patients generally not displaying platelet reductions typical of consumptive coagulopathies. These temporal trends provide fine-grained resolution into COVID-19 associated coagulopathy (CAC) and set the stage for personalizing thromboprophylaxis.

Shorter synthesis of trifunctionalized cryptophane-A derivatives.

Efficient syntheses of trisubstituted cryptophane-A derivatives that are versatile host molecules for many applications are reported. Trihydroxy cryptophane was synthesized in six or seven steps with yields as high as 9.5%. By a different route, trihydroxy cryptophane modified with three propargyl, allyl, or benzyl protecting groups was synthesized with yields of 4.1-5.8% in just six steps. Hyperpolarized (129)Xe NMR chemical shifts of 57-65 ppm were measured for these trisubstituted cryptophanes.

Cell-compatible, integrin-targeted cryptophane-129Xe NMR biosensors.

Peptide-modified cryptophane enables sensitive detection of protein analytes using hyperpolarized 129Xe NMR spectroscopy. Here we report improved targeting and delivery of cryptophane to cells expressing αvß3 integrin receptor, which is overexpressed in many human cancers. Cryptophane was functionalized with cyclic RGDyK peptide and Alexa Fluor 488 dye, and cellular internalization was monitored by confocal laser scanning microscopy. Competitive blocking assays confirmed cryptophane endocytosis through an αvß3 integrin receptor-mediated pathway. The peptide-cryptophane conjugate was determined to be nontoxic in normal human lung fibroblasts by MTT assay at the micromolar cryptophane concentrations typically used for hyperpolarized 129Xe NMR biosensing experiments. Flow cytometry revealed 4-fold higher cellular internalization in cancer cells overexpressing the integrin receptor compared to normal cells. Nanomolar inhibitory concentrations (IC50 = 20-30 nM) were measured for cryptophane biosensors against vitronectin binding to αvß3 integrin and fibrinogen binding to αIIbß3 integrin. Functionalization of the conjugate with two propionic acid groups improved water solubility for hyperpolarized 129Xe NMR spectroscopic studies, which revealed a single resonance at 67 ppm for the 129Xe-cryptophane-cyclic RGDyK biosensor. Introduction of αIIbß3 integrin receptor in detergent solution generated a new "bound" 129Xe biosensor peak that was shifted 4 ppm downfield from the "free" 129Xe biosensor.

Crystallographic observation of 'induced fit' in a cryptophane host-guest model system.

Cryptophane-A, comprised of two cyclotriguaiacylenes joined by three ethylene linkers, is a prototypal organic host molecule that binds reversibly to neutral small molecules via London forces. Of note are trifunctionalized, water-soluble cryptophane-A derivatives, which exhibit exceptional affinity for xenon in aqueous solution. In this paper, we report high-resolution X-ray structures of cryptophane-A and trifunctionalized derivatives in crown-crown and crown-saddle conformations, as well as in complexes with water, methanol, xenon or chloroform. Cryptophane internal volume varied by more than 20% across this series, which exemplifies 'induced fit' in a model host-guest system.