e20526 Background: The majority of early-phase clinical trials in oncology are single-arm trials. External comparator arms can provide important contextual information to interpret results and to inform the design of future trials. But patient-level trial data may be lacking, especially because oncology is characterized by rapidly evolving standards-of-care for increasingly granular subgroups. Real-world data (RWD) and summary trial results offer complementary but incomplete pictures of the effects of therapies on patients. The difficulty of combining these sources of information hampers optimal decision-making in trial planning, design, and analysis. Methods: We developed a machine learning (ML) model that integrates patient-level RWD with summary statistics from recent trial publications. The model generates patient-level predictions of outcomes under existing treatments that conform to the results of past RCTs in the aggregate. This provides a novel approach to integrating RWD into granular analyses while maintaining the gold-standard status of randomized trial evidence. The core of the model consists in a foundational pan-cancer transformer model that was trained on RWD from a collection of detailed clinical and genetic data from roughly 250k tumor biopsies [custom-processed versions of AACR GENIE, GENIE BPC NSCLC & CRC, and MSK-CHORD]. We demonstrate a novel calibration technique that allows us to conform the model’s survival predictions to match published results, incorporating one or many RCT results into its weights. Results: We demonstrate the capability of this approach to both interpolate between results of RCTs and to extrapolate to new trials. We calibrate the model to the full set of KEYNOTE trials in mNSCLC and show that it generates patient-level trial simulations across baseline cohorts and treatments within the KEYNOTE trial span while recapitulating observed outcomes at calibration points. We further demonstrate that the model recapitulates published control-arm results from the POSEIDON and LEAP-006 trials. Across PD-L1 expression strata, histologic subtype, KEAP1/STK11/KRAS mutation status, and the overall population, 90.9% (40/44) of median and 2–5-year OS estimates fell within reported 95% confidence intervals, with median absolute deviation 2.7% in the survival benchmarks. Conclusions: ML models integrating RWD with trial results enable accurate prediction of survival in a way that can simultaneously support granular analyses while conforming to gold-standard RCT results. Our approach enables the creation of external comparators to better evaluate the efficacy of new therapies. More broadly it supports data-driven decision-making in trial planning, trial analysis, and hypothesis generation.

Today's approach to medicine requires extensive trial and error to determine the proper treatment path for each patient. While many fields have benefited from technological breakthroughs in computer science, such as artificial intelligence (AI), the task of developing effective treatments is actually getting slower and more costly. With the increased availability of rich historical datasets from previous clinical trials and real-world data sources, one can leverage AI models to create holistic forecasts of future health outcomes for an individual patient in the form of an AI-generated digital twin. This could support the rapid evaluation of intervention strategies in silico and could eventually be implemented in clinical practice to make personalized medicine a reality. In this work, we focus on uses for AI-generated digital twins of clinical trial participants and contend that the regulatory outlook for this technology within drug development makes it an ideal setting for the safe application of AI-generated digital twins in healthcare. With continued research and growing regulatory acceptance, this path will serve to increase trust in this technology and provide momentum for the widespread adoption of AI-generated digital twins in clinical practice.

A patient's digital twin is a computational model that describes the evolution of their health over time. Digital twins have the potential to revolutionize medicine by enabling individual-level computer simulations of human health, which can be used to conduct more efficient clinical trials or to recommend personalized treatment options. Due to the overwhelming complexity of human biology, machine learning approaches that leverage large datasets of historical patients' longitudinal health records to generate patients' digital twins are more tractable than potential mechanistic models. In this manuscript, we describe a neural network architecture that can learn conditional generative models of clinical trajectories, which we call Digital Twin Generators (DTGs), that can create digital twins of individual patients. We show that the same neural network architecture can be trained to generate accurate digital twins for patients across 13 different indications simply by changing the training set and tuning hyperparameters. By introducing a general purpose architecture, we aim to unlock the ability to scale machine learning approaches to larger datasets and across more indications so that a digital twin could be created for any patient in the world.

This chapter argues that in an era of unprecedented environmental turbulence and uncertainty, there is no longer space for standard leadership approaches. Further, top-down leadership control in the age of sustainability is unworkable because it fails to appreciate that change occurs naturally and is intimately entwined with continuity. In a time when organizations must be capable of adapting to immense competition while maintaining new levels of environmental and ethical performance, change leadership must assume a new form. The chapter proposes that the change–continuity continuum moderates organizational performance. The ability to exploit and explore simultaneously comes at the price of new leadership dynamics. Sustainable leadership means accepting that organizational change has shifted in its focus and deployment. Successful change no longer equates with fast change. Sustainable leadership for change demands accepting a worldview where either/or choices such as flexibility or control are misleading. Change and continuity do not exist as opposite sides of the leadership see-saw but co-exist as dualities that can sit side-by-side without compromising one another.

As global regulators of eukaryotic homeostasis, arginyltransferases (ATE1s) have essential functions within the cell. Thus, the regulation of ATE1 is paramount. It was previously postulated that ATE1 was a hemoprotein and that heme was an operative cofactor responsible for enzymatic regulation and inactivation. However, we have recently shown that ATE1 instead binds an iron-sulfur ([Fe-S]) cluster that appears to function as an oxygen sensor to regulate ATE1 activity. As this cofactor is oxygen-sensitive, purification of ATE1 in the presence of O2 results in cluster decomposition and loss. Here, we describe an anoxic chemical reconstitution protocol to assemble the [Fe-S] cluster cofactor in Saccharomyces cerevisiae ATE1 (ScATE1) and Mus musculus ATE1 isoform 1 (MmATE1-1).

The previous decade has seen a surge in the use of nanoparticles for numerous biotechnological applications, a direct result in the expansion of new nanoparticle materials and recent advances in their synthesis. However, many applications require the conjugation of biomolecules or small molecules on nanoparticle surfaces for activity. Traditional chemical ligation methods have been extensively used, however, they suffer from several disadvantages when conjugating biomolecules. Despite substantial improvements on this front, enzymatic bioconjugation offers a powerful and attractive alternative. In this review, we highlight several of the most prevalent and successful enzymatic bioconjugation systems and introduce newer and emerging technologies.

Synthetic biology is touted as the next industrial revolution as it promises access to greener biocatalytic syntheses to replace many industrial organic chemistries. Here, it is shown to what synthetic biology can offer in the form of multienzyme cascades for the synthesis of the most basic of new materials-chemicals, including especially designer chemical products and their analogs. Since achieving this is predicated on dramatically expanding the chemical space that enzymes access, such chemistry will probably be undertaken in cell-free or minimalist formats to overcome the inherent toxicity of non-natural substrates to living cells. Laying out relevant aspects that need to be considered in the design of multi-enzymatic cascades for these purposes is begun. Representative multienzymatic cascades are critically reviewed, which have been specifically developed for the synthesis of compounds that have either been made only by traditional organic synthesis along with those cascades utilized for novel compound syntheses. Lastly, an overview of strategies that look toward exploiting bio/nanomaterials for accessing channeling and other nanoscale materials phenomena in vitro to direct novel enzymatic biosynthesis and improve catalytic efficiency is provided. Finally, a perspective on what is needed for this field to develop in the short and long term is presented.

It is widely acknowledged that calibrating and evaluating hydrological models only against streamflow may lead to inconsistencies of internal model states and large parameter uncertainties. Soil moisture is a key variable for the energy and water balance, which affects the partitioning of solar radiation into latent and sensible heat as well as the partitioning of precipitation into direct runoff and catchment storage. In contrast to ground-based measurements, satellite-derived soil moisture (SDSM) data are widely available and new data products benefit from improved spatio-temporal resolutions. Here we use a soil water index product based on data fusion of microwave data from METOP ASCAT and Sentinel 1 CSAR for calibrating the process-based ecohydrological model EcH2O-iso in the 66 km² Demnitzer Millcreek catchment in NE Germany. Available field measurements in and close to this intensively monitored catchment include soil moisture data from 74 sensors and water stable isotopes in precipitation, stream and soil water. Water stable isotopes provide information on flow pathways, storage dynamics, and the partitioning of evapotranspiration into evaporation and transpiration. Accounting for water stable isotopes in the ecohydrologic model therefore provides further insights regarding the consistency of internal processes. We first compare the SDSM data to the ground-based measurements. Based on a Monte Carlo approach, we then investigate the trade-off between model performance in terms of soil moisture and streamflow. In situ soil moisture and water stable isotopes are further consulted to evaluate the internal consistency of the model. Overall, we find relatively good agreements between satellite-derived and ground based soil moisture dynamics. Preliminary results suggest that including SDSM in the model calibration can improve the simulation of internal processes, but uncertainties of the SDSM data should be accounted for. The findings of this study are relevant for reliable ecohydrological modelling in catchments that lack detailed field measurements for model evaluation.

Ferrous iron (Fe2+) transport is an essential process that supports the growth, intracellular survival, and virulence of several drug-resistant pathogens, and the ferrous iron transport (Feo) system is the most important and widespread protein complex that mediates Fe2+ transport in these organisms. The Feo system canonically comprises three proteins (FeoA/B/C). FeoA and FeoC are both small, accessory proteins localized to the cytoplasm, and their roles in the Fe2+ transport process have been of great debate. FeoB is the only wholly-conserved component of the Feo system and serves as the inner membrane-embedded Fe2+ transporter with a soluble G-protein-like N-terminal domain. In vivo studies have underscored the importance of Feo during infection, emphasizing the need to better understand Feo-mediated Fe2+ uptake, although a paucity of research exists on intact FeoB. To surmount this problem, we designed an overproduction and purification system that can be applied generally to a suite of intact FeoBs from several organisms. Importantly, we noted that FeoB is extremely sensitive to excess salt while in the membrane of a recombinant host, and we designed a workflow to circumvent this issue. We also demonstrated effective protein extraction from the lipid bilayer through small-scale solubilization studies. We then applied this approach to the large-scale purifications of Escherichia coli and Pseudomonas aeruginosa FeoBs to high purity and homogeneity. Lastly, we show that our protocol can be generally applied to various FeoB proteins. Thus, this workflow allows for isolation of suitable quantities of FeoB for future biochemical and biophysical characterization.

Valvular heart disease contributes to a large burden of morbidity and mortality in the United States. During the last decade there has been a paradigm shift in the management of valve disease, primarily driven by the emergence of novel transcatheter technologies. In this article, the latest update of the American College of Cardiology/American Heart Association valve heart disease guidelines is reviewed.

Introduction: Coronary Computed Tomography Angiography (CCTA) and SPECT myocardial perfusion imaging (SPECT-MPI) are equally safe in evaluating patients presenting to the emergency department with acute chest pain. However, no study has compared the use of CCTA with Fractional Flow Reserve (FFR-CT) to SPECT-MPI in a chest pain observation unit (OU). We compared the performance of CCTA (with FFR-CT as needed) to SPECT-MPI for change in coronary artery disease (CAD) specific treatment (statins, antianginals, antiplatelets, and coronary revascularization) as well its impact on length of hospital stay (LOS). Methods: In 2020, our institution implemented an algorithm-based protocol in the OU to select patients appropriate for CCTA with explicit guidelines for management based on the results. We performed a retrospective analysis of 105 patients (52 SPECT-MPI patients from the period prior to initiation of the OU CCTA protocol who met criteria for CCTA, and 53 CCTA patients after CCTA protocol initiation). We performed two-tailed t-tests and chi-squared analyses to compare changes in treatment and LOS in both groups. Results: Patients in the CCTA group were younger (61 vs. 67 years, p=0.007), but there were no significant differences in CAD risk factors amongst the two groups. CCTA led to significant increase in statin, antianginal, and antiplatelet therapies, cardiac catheterization and revascularization within 6 months. There was no increase in LOS. Finally, CCTA did not increase acute kidney injury at follow up (Table 1). Conclusion: CCTA with FFR-CT significantly improved patient care by increasing appropriate preventative CAD treatment compared to SPECT-MPI while increasing the rate of revascularization without compromising length of stay or kidney function. In conclusion, protocol-based use of CCTA with FFR-CT is a more effective tool when compared to SPECT-MPI for chest pain patients admitted to the OU.

Abstract. Significance: Peripheral pitting edema is a clinician-administered measure for grading edema. Peripheral edema is graded 0, 1  +  , 2  +  , 3  +  , or 4  +  , but subjectivity is a major limitation of this technique. A pilot clinical study for short-wave infrared (SWIR) molecular chemical imaging (MCI) effectiveness as an objective, non-contact quantitative peripheral edema measure is underway. Aim: We explore if SWIR MCI can differentiate populations with and without peripheral edema. Further, we evaluate the technology for correctly stratifying subjects with peripheral edema. Approach: SWIR MCI of shins from healthy subjects and heart failure (HF) patients was performed. Partial least squares discriminant analysis (PLS-DA) was used to discriminate the two populations. PLS regression (PLSR) was applied to assess the ability of MCI to grade edema. Results: Average spectra from edema exhibited higher water absorption than non-edema spectra. SWIR MCI differentiated healthy volunteers from a population representing all pitting edema grades with 97.1% accuracy (N  =  103 shins). Additionally, SWIR MCI correctly classified shin pitting edema levels in patients with 81.6% accuracy. Conclusions: Our study successfully achieved the two primary endpoints. Application of SWIR MCI to monitor patients while actively receiving HF treatment is necessary to validate SWIR MCI as an HF monitoring technology.

Machine learning models can leverage historical data to forecast disease progression. These predictions can be integrated in clinical trial design to reduce sample size or increase power, speeding up the evaluation of new drugs. This is especially critical in AD where trials face challenges with enrollment and where no new therapies have emerged in 18 years. Recently, there has been a shift towards evaluating drugs in earlier stages of the disease using Clinical Dementia Rating Sum-of-Boxes (CDR-SB) as the primary endpoint. We present a model that includes CDR-SB and spans a broad range of baseline disease severity from MCI to mild-to-moderate AD. We used nearly 7,000 clinical records from placebo arms of AD clinical trials (in the C-Path Online Data Repository for AD) and from observational studies (in the AD Neuroimaging Initiative) to train Conditional Restricted Boltzmann Machines (CRBMs). A CRBM is a generative machine learning model that learns a multivariate distribution over the relevant variables, and is particularly suited to model clinical data. A CRBM generates Digital Twins - synthetic clinical records with baseline characteristics matched to those of actual trial subjects describing their likely progression under standard-of-care (SOC) with/without placebo. These forecasts include all components of CDR-SB, Alzheimer’s Disease Assessment Scale - Cognitive Subscale (ADAS-Cog11) and Mini Mental State Examination (MMSE), along with other variables including labs, vitals, and biomarker status. We evaluated Digital Twins on a held-out portion of the dataset (not included in the training dataset), stratified by baseline ADAS-Cog11. For each variable, the mean change from baseline +/- 95% confidence interval error at 3-month intervals was estimated by drawing 100 Digital Twins for each subject and averaging over the population. CRBM-generated predictions of ADAS-Cog11, MMSE, and CDR-SB progression up to 18 months were consistent within 95% confidence intervals with actual data across all scores, timepoints, and cohorts. Digital Twins accurately model MCI and AD subjects’ progression on SOC with/without placebo across a broad range of baseline disease severity, and can be integrated into clinical trials as prognostic scores to increase power or reduce sample size required to achieve desired power.

Functional phenotypic cancer cell heterogeneity limits the efficacy of targeted and immuno-therapies. The transcription factor MITF is known to regulate melanoma cell plasticity and, consequently, response to drugs. However, the underlying mechanisms of this phenomenon remain incompletely understood. Here, we show that MITF negatively regulates peroxidasin and fine-tunes the ability to contract the extracellular matrix, the maturation of focal adhesions and ROCK-mediated melanoma cell contractility. This, in turn, results in control of functional melanoma cell heterogeneity through spatio-temporal downregulation of p27kip. Modulation of MITF expression alters extracellular matrix organization, melanoma cell morphology and solid stress in three-dimensional melanoma spheroids, thereby accounting for spatial differences in cell cycle dynamics. Together, our data identify MITF as a master regulator of the melanoma micro-architecture and point towards novel targeting strategies for cancer cell heterogeneity. Significance: Development of drug resistance is a major cause of melanoma therapy failure. The role of MITF in melanoma response to therapy has been discussed controversially, which can be explained, at least in part, through the rheostat model linking MITF activity to cell proliferation. Heterogeneity is widely associated with therapy resistance, however, whether cell phenotype switching, mediated by MITF, is responsible for treatment resistance is not known. Our findings provide an in-depth mechanistic understanding of the MITF-mediated regulation of cell cycle behavior and physical regulation of the tumor architecture. As MITF is not amenable to direct drug targeting, the identification of mediators of MITF-triggered functional heterogeneity reveals novel targets that can be deployed to control this phenomenon.

Arginyl-tRNATransferase 1 (ATE1) is a eukaryotic enzyme that arginylates cellular proteins, is an essential regulator of eukaryotic homeostasis via its involvement in the N-degron pathway, and regulates essential physiological processes such as embryogenesis, aging, cell migration, muscle contraction, and stress. Despite its importance, the structure, mechanism of action, and regulation of ATE1 has yet to be elucidated. The objective of this work is to model the three-dimensional fold of an ATE1, and to determine essential conserved residues that are involved in substrate recognition and enzymatic arginylation. In this study, we have modeled the structure of Saccharomyces cerevisiae ATE1 (ScATE1) and identified a putative location of the GCN5-related N-acetyl transferase (GNAT) fold, which is structurally conserved amongst amino-acid transferases. Additionally, we have performed a large-scale sequence alignment across eukaryotic ATE1s to map conserved residues onto our three-dimensional model. Notably, we have identified a highly-conserved His residue that would be consistent with the recognition of a negatively charged substrate and is located in a structural homologous site in bacterial amino-acid transferases. We hypothesize that this highly conserved (>90%) His residue is essential for substrate recognition, and site-directed mutagenesis is underway to verify this hypothesis. These results provide the first structural insights into the enzyme active site and mechanism of substrate recognition for ScATE1, and the strong sequence conservation suggest a common mode of action across eukaryotic homologs.

Heart failure (HF) has a large impact on patient outcomes and health care costs. Objective monitoring of the pitting edema level of a HF patient may help clinicians reduce the amount of readmissions. ChemImage is developing a Molecular Chemical Imaging (MCI) device for monitoring HF patients that will non-invasively quantify peripheral edema. Results from a completed in-human clinical trial will be presented demonstrating ability to discriminate between healthy volunteers and HF patients with all levels of pitting edema and correct prediction of peripheral edema grade across the patient population. Follow-on clinical trials will address monitoring patients during treatment.