Emergency drug use in a pandemic: Harsh lessons from COVID-19.

The scientific and clinical communities have both experienced several harsh lessons on clinical care management and drug development during the COVID-19 pandemic. Here, we discuss several key lessons learned and describe a framework within which our two communities can work together and invest in to improve future pandemic responses.

TOP1 inhibition therapy protects against SARS-CoV-2-induced lethal inflammation.

The ongoing pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is currently affecting millions of lives worldwide. Large retrospective studies indicate that an elevated level of inflammatory cytokines and pro-inflammatory factors are associated with both increased disease severity and mortality. Here, using multidimensional epigenetic, transcriptional, in vitro, and in vivo analyses, we report that topoisomerase 1 (TOP1) inhibition suppresses lethal inflammation induced by SARS-CoV-2. Therapeutic treatment with two doses of topotecan (TPT), an FDA-approved TOP1 inhibitor, suppresses infection-induced inflammation in hamsters. TPT treatment as late as 4 days post-infection reduces morbidity and rescues mortality in a transgenic mouse model. These results support the potential of TOP1 inhibition as an effective host-directed therapy against severe SARS-CoV-2 infection. TPT and its derivatives are inexpensive clinical-grade inhibitors available in most countries. Clinical trials are needed to evaluate the efficacy of repurposing TOP1 inhibitors for severe coronavirus disease 2019 (COVID-19) in humans.

Outcomes of tricuspid regurgitation after lead extraction.

INTRODUCTION: Transvenous leads have been implicated in tricuspid valve (TV) dysfunction, but limited data are available regarding the effect of extracting leads across the TV on valve regurgitation. The aim of this study is to quantify tricuspid regurgitation (TR) before and after lead extraction and identify predictors of worsening TR. METHODS: We studied 321 patients who had echocardiographic data before and after lead extraction. TR was graded on a scale (0 = none/trivial, 1 = mild, 2 = moderate, 3 = severe). A change of >1 grade following extraction was considered significant. RESULTS: A total of 321 patients underwent extraction of a total of 338 leads across the TV (1.05 ± 0.31 leads across the TV per patient). There was no significant difference on average TR grade pre- and postextraction (1.18 ± 0.91 vs. 1.15 ± 0.87; p = 0.79). TR severity increased after extraction in 84 patients, but was classified as significantly worse (i.e., >1 grade change in severity) in only 8 patients (2.5%). Use of laser lead extraction was associated with a higher rate of worsening TR postextraction (44.0% vs. 31.6%, p = 0.04). CONCLUSION: In our single-center analysis, extraction of leads across the TV did not significantly affect the extent of TR in most patients. Laser lead extraction was associated with a higher rate of worsening TR after extraction.

Nuclear pCHK1 as a potential biomarker of increased sensitivity to ATR inhibition.

Excessive genomic instability coupled with abnormalities in DNA repair pathways induces high levels of 'replication stress' when cancer cells propagate. Rather than hampering cancer cell proliferation, novel treatment strategies are turning their attention towards targeting cell cycle checkpoint kinases (such as ATR, CHK1, WEE1, and others) along the DNA damage response and replicative stress response pathways, thereby allowing unrepaired DNA damage to be carried forward towards mitotic catastrophe and apoptosis. The selective ATR kinase inhibitor elimusertib (BAY 1895344) has demonstrated preclinical and clinical monotherapy activity; however, reliable predictive biomarkers of treatment benefit are still lacking. In this study, using gene expression profiling of 24 cell lines from different cancer types and in a panel of ovarian cancer cell lines, we found that nuclear-specific enrichment of checkpoint kinase 1 (CHK1) correlated with increased sensitivity to elimusertib. Using an advanced multispectral imaging system in subsequent cell line-derived xenograft specimens, we showed a trend between nuclear phosphorylated CHK1 (pCHK1) staining and increased sensitivity to the ATR inhibitor elimusertib, indicating the potential value of pCHK1 expression as a predictive biomarker of ATR inhibitor sensitivity. © 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Super-enhancer-driven TOX2 mediates oncogenesis in Natural Killer/T Cell Lymphoma.

BACKGROUND: Extranodal natural killer/T-cell lymphoma (NKTL) is an aggressive type of non-Hodgkin lymphoma with dismal outcome. A better understanding of disease biology and key oncogenic process is necessary for the development of targeted therapy. Super-enhancers (SEs) have been shown to drive pivotal oncogenes in various malignancies. However, the landscape of SEs and SE-associated oncogenes remain elusive in NKTL. METHODS: We used Nano-ChIP-seq of the active enhancer marker histone H3 lysine 27 acetylation (H3K27ac) to profile unique SEs NKTL primary tumor samples. Integrative analysis of RNA-seq and survival data further pinned down high value, novel SE oncogenes. We utilized shRNA knockdown, CRISPR-dCas9, luciferase reporter assay, ChIP-PCR to investigate the regulation of transcription factor (TF) on SE oncogenes. Multi-color immunofluorescence (mIF) staining was performed on an independent cohort of clinical samples. Various function experiments were performed to evaluate the effects of TOX2 on the malignancy of NKTL in vitro and in vivo. RESULTS: SE landscape was substantially different in NKTL samples in comparison with normal tonsils. Several SEs at key transcriptional factor (TF) genes, including TOX2, TBX21(T-bet), EOMES, RUNX2, and ID2, were identified. We confirmed that TOX2 was aberrantly overexpressed in NKTL relative to normal NK cells and high expression of TOX2 was associated with worse survival. Modulation of TOX2 expression by shRNA, CRISPR-dCas9 interference of SE function impacted on cell proliferation, survival and colony formation ability of NKTL cells. Mechanistically, we found that RUNX3 regulates TOX2 transcription by binding to the active elements of its SE. Silencing TOX2 also impaired tumor formation of NKTL cells in vivo. Metastasis-associated phosphatase PRL-3 has been identified and validated as a key downstream effector of TOX2-mediated oncogenesis. CONCLUSIONS: Our integrative SE profiling strategy revealed the landscape of SEs, novel targets and insights into molecular pathogenesis of NKTL. The RUNX3-TOX2-SE-TOX2-PRL-3 regulatory pathway may represent a hallmark of NKTL biology. Targeting TOX2 could be a valuable therapeutic intervene for NKTL patients and warrants further study in clinic.

Novel use of an irrigated ablation catheter to monitor real-time hemodynamics during ablation.

INTRODUCTION: Hemodynamic decompensation during catheter ablation occurs due to prolonged procedure time and irrigant delivery directly into the cardiac chambers. Real-time hemodynamic monitoring of patients undergoing catheter ablation procedures may identify patients at risk of decompensation; we set out to assess the feasibility of a novel, real-time, intracardiac pressure monitoring system using a standard irrigated ablation catheter. METHODS: We studied 13 consecutive who underwent pressure measurement of the left atrium (LA) and left ventricle (LV) via transeptal access with a Swan Ganz (SG) catheter followed by two commercially available irrigated ablation catheters. Pressure waveform data was extracted to compare LA peak pressure, LV peak systolic pressure, LV end-diastolic pressure, and waveform analysis. RESULTS: Comparison between the SG and ablation catheters (AblA; AblB) demonstrated that LV systolic pressure (0.61-16.8 mmHg; 1.32-18.2 mmHg), and LV end-diastolic pressure (-3.4 to 2.8 mmHg; -3.0 to 3.35 mmHg) were well correlated and had accepted repeatability. Ablation waveforms demonstrated an 89.9 ± 6.4% correlation compared to SG waveforms. CONCLUSION: Pressure measurements derived from an irrigated ablation catheter are accurate and reliable when compared to an SG catheter. Further studies are needed to determine how real-time pressure monitoring can improve outcomes during ablation procedures.

Prospective validation of a risk score to predict pacemaker implantation after transcatheter aortic valve replacement.

INTRODUCTION: The need for pacemaker is a common complication after transcatheter aortic valve replacement (TAVR). We previously described the Emory Risk Score (ERS) to predict the need for new pacemaker implant (PPM) after TAVR. Metrics included in the score are a history of syncope, pre-existing RBBB, QRS duration ≥140 ms, and prosthesis oversizing ≥16%. To prospectively validate the previously described risk score. METHODS: We prospectively evaluated all patients without pre-existing pacemakers, ICD, or pre-existing indications for pacing undergoing TAVR with the Edwards SAPIEN 3 prosthesis at our institution from March 2019 to December 2020 (n = 661). Patients were scored prospectively; however, results were blinded from clinical decision-making. The primary endpoint was PPM at 30 days after TAVR. Performance of the ERS was evaluated using logistic regression, a calibration curve to prior performance, and receiver operating characteristic (ROC) analysis. RESULTS: A total of 48 patients (7.3%) had PPM after TAVR. A higher ERS predicted an increased likelihood of PPM (OR 2.61, 95% CI: 2.05-3.25 per point, p < 0.001). There was good correlation between observed and expected values on the calibration curve (slope = 1.04, calibration at large = 0.001). The area under the ROC curve was 0.81 (95% CI [0.74-0.88], p < 0.001). CONCLUSIONS: The ERS prospectively predicted the need for PPM in a serial, real-world cohort of patients undergoing TAVR with a balloon-expandable prosthesis, confirming findings previously described in retrospective cohorts. Notably, the prospective performance of the score was comparable with that of the initial cohorts. The risk score could serve as a framework for preprocedural risk stratification for PPM after TAVR.

Synchronization and coexistence in quantum networks.

We investigate the coexistence of clock synchronization protocols with quantum signals in a common single-mode optical fiber. By measuring optical noise between 1500 nm to 1620 nm we demonstrate a potential for up to 100 quantum, 100 GHz wide channels coexisting with the classical synchronization signals. Both "White Rabbit" and pulsed laser-based synchronization protocols were characterized and compared. We establish a theoretical limit of the fiber link length for coexisting quantum and classical channels. The maximal fiber length is below approximately 100 km for off-the-shelf optical transceivers and can be significantly improved by taking advantage of quantum receivers.

Long-term performance of single-connector (DF4) implantable defibrillator leads.

AIMS: Single-connector (DF4) defibrillator leads have become the predominantly implanted transvenous implantable cardioverter-defibrillator lead. However, data on their long-term performance are derived predominantly from manufacturer product performance reports. METHODS AND RESULTS: We reviewed medical records in 5289 patients with DF4 leads between 2011 and 2023 to determine the frequency of lead-related abnormalities. We defined malfunction as any single or combination of electrical abnormalities requiring revision including a sudden increase (≥2×) in stimulation threshold, a discrete jump in high-voltage impedance, or sensing of non-physiologic intervals or noise. We documented time to failure, predictors of failure, and management strategies. Mean follow-up after implant was 4.15 ± 3.6 years (median = 3.63), with 37% of leads followed for >5 years. A total of 80 (1.5%) leads demonstrated electrical abnormalities requiring revision with an average time to failure of 4 ± 2.8 years (median = 3.5). Of the leads that malfunctioned, 62/80 (78%) were extracted and replaced with a new lead and in the other 18 cases, malfunctioned DF4 leads were abandoned, and a new lead implanted. In multivariable models, younger age at implant (OR 1.03 per year; P < 0.001) and the presence of Abbott/St. Jude leads increased the risk of malfunction. CONCLUSION: DF4 defibrillator leads demonstrate excellent longevity with >98.3% of leads followed for at least 5 years still functioning normally. Younger age at implant and lead manufacturer are associated with an increased risk of DF4 lead malfunction. The differences in lead survival between manufacturers require further investigation.

Complex repolarization dynamics in ex vivo human ventricles are independent of the restitution properties.

AIMS: The mechanisms of transition from regular rhythms to ventricular fibrillation (VF) are poorly understood. The concordant to discordant repolarization alternans pathway is extensively studied; however, despite its theoretical centrality, cannot guide ablation. We hypothesize that complex repolarization dynamics, i.e. oscillations in the repolarization phase of action potentials with periods over two of classic alternans, is a marker of electrically unstable substrate, and ablation of these areas has a stabilizing effect and may reduce the risk of VF. To prove the existence of higher-order periodicities in human hearts. METHODS AND RESULTS: We performed optical mapping of explanted human hearts obtained from recipients of heart transplantation at the time of surgery. Signals recorded from the right ventricle endocardial surface were processed to detect global and local repolarization dynamics during rapid pacing. A statistically significant global 1:4 peak was seen in three of six hearts. Local (pixel-wise) analysis revealed the spatially heterogeneous distribution of Periods 4, 6, and 8, with the regional presence of periods greater than two in all the hearts. There was no significant correlation between the underlying restitution properties and the period of each pixel. CONCLUSION: We present evidence of complex higher-order periodicities and the co-existence of such regions with stable non-chaotic areas in ex vivo human hearts. We infer that the oscillation of the calcium cycling machinery is the primary mechanism of higher-order dynamics. These higher-order regions may act as niduses of instability and may provide targets for substrate-based ablation of VF.

Design and characteristics of the prophylactic intra-operative ventricular arrhythmia ablation in high-risk LVAD candidates (PIVATAL) trial.

BACKGROUND: The use of a Left Ventricular Assist Device (LVAD) in patients with advanced heart failure refractory to optimal medical management has progressed steadily over the past two decades. Data have demonstrated reduced LVAD efficacy, worse clinical outcome, and higher mortality for patients who experience significant ventricular tachyarrhythmia (VTA). We hypothesize that a novel prophylactic intra-operative VTA ablation protocol at the time of LVAD implantation may reduce the recurrent VTA and adverse events postimplant. METHODS: We designed a prospective, multicenter, open-label, randomized-controlled clinical trial enrolling 100 patients who are LVAD candidates with a history of VTA in the previous 5 years. Enrolled patients will be randomized in a 1:1 fashion to intra-operative VTA ablation (n = 50) versus conventional medical management (n = 50) with LVAD implant. Arrhythmia outcomes data will be captured by an implantable cardioverter defibrillator (ICD) to monitor VTA events, with a uniform ICD programming protocol. Patients will be followed prospectively over a mean of 18 months (with a minimum of 9 months) after LVAD implantation to evaluate recurrent VTA, adverse events, and procedural outcomes. Secondary endpoints include right heart function/hemodynamics, healthcare utilization, and quality of life. CONCLUSION: The primary aim of this first-ever randomized trial is to assess the efficacy of intra-operative ablation during LVAD surgery in reducing VTA recurrence and improving clinical outcomes for patients with a history of VTA.

A Multicenter Observational Study Comparing Virtual with In-Person Morning Reports during the COVID-19 Pandemic.

OBJECTIVES: The coronavirus disease 2019 (COVID-19) pandemic disrupted how educational conferences were delivered, leaving programs to choose between in-person and virtual morning report formats. The objective of our study was to describe morning reports during the COVID-19 pandemic, including the use of virtual formats, attendance, leadership, and content. METHODS: A prospective observational study of morning reports was conducted at 13 Internal Medicine residency programs between September 1, 2020 and March 30, 2021, including a follow-up survey of current morning report format in January 2023. RESULTS: In total, 257 reports were observed; 74% used virtual formats, including single hospital, multiple hospital, and a hybrid format with both in-person and virtual participants. Compared with in-person reports, virtual reports had more participants, with increased numbers of learners (median 21 vs 7; P < 0.001) and attendings (median 4 vs 2; P < 0.001), and they were more likely to involve medical students (83% vs 40%; P < 0.001), interns (99% vs 53%; P < 0.001), and program directors (68% vs 32%; P < 0.001). Attendings were less likely to lead virtual reports (3% vs 28%, P < 0.001). Virtual reports also were more likely to be case based (88% vs 69%; P < 0.001) and to use digital presentation slides (91% vs 36%; P < 0.001). There was a marked increase in the number of slides (median 20 vs 0; P < 0.001). As of January 2023, all 13 programs had returned to in-person reports, with only 1 program offering an option to participate virtually. CONCLUSIONS: During the COVID-19 pandemic, virtual morning report formats predominated. Compared with traditional in-person reports, virtual report increased attendance, favored resident leadership, and approached a similar range of patient diagnoses with a greater number of case-based presentations and slides. In spite of these characteristics, all programs returned to an in-person format for morning report as pandemic restrictions waned.

What internal medicine attendings talk about at morning report: a multicenter study.

BACKGROUND: Morning report is a core educational activity in internal medicine resident education. Attending physicians regularly participate in morning report and influence the learning environment, though no previous study has described the contribution of attending physicians to this conference. This study aims to describe attending comments at internal medicine morning reports. METHODS: We conducted a prospective, observational study of morning reports conducted at 13 internal medicine residency programs between September 1, 2020, and March 30, 2021. Each attending comment was described including its duration, whether the comment was teaching or non-teaching, teaching topic, and field of practice of the commenter. We also recorded morning report-related variables including number of learners, report format, program director participation, and whether report was scripted (facilitator has advance knowledge of the case). A regression model was developed to describe variables associated with the number of attending comments per report. RESULTS: There were 2,344 attending comments during 250 conferences. The median number of attendings present was 3 (IQR, 2-5). The number of comments per report ranged across different sites from 3.9 to 16.8 with a mean of 9.4 comments/report (SD, 7.4). 66% of comments were shorter than one minute in duration and 73% were categorized as teaching by observers. The most common subjects of teaching comments were differential diagnosis, management, and testing. Report duration, number of general internists, unscripted reports, and in-person format were associated with significantly increased number of attending comments. CONCLUSIONS: Attending comments in morning report were generally brief, focused on clinical teaching, and covered a wide range of topics. There were substantial differences between programs in terms of the number of comments and their duration which likely affects the local learning environment. Morning report stakeholders that are interested in increasing attending involvement in morning report should consider employing in-person and unscripted reports. Additional studies are needed to explore best practice models of attending participation in morning report.

Biochar and fungi as bioretention amendments for bacteria and PAH removal from stormwater.

Bioretention has been widely used to mitigate hydrologic impacts of stormwater runoff and is increasingly being relied upon to treat chemical and biological pollutants transported by stormwater. Despite this reliance, we still lack an understanding of treatment performance for certain organic and biological contaminants which may interact with biotic and abiotic components of bioretention systems. We evaluated the treatment of fecal indicator bacteria (FIB) and polycyclic aromatic hydrocarbons (PAHs) in stormwater runoff by bioretention. We compared treatment performance by Washington's standard bioretention mix of 60% sand: 40% compost (by volume), and by three other mixtures amended with biochar, fungi (Stropharia rugosoannulata), or both. All bioretention columns were conditioned with clean water and then dosed with collected roadway runoff at a rate equivalent to a 6 month, 24 h storm in this region during 8 events over a 14-month period. Effluents for each column were analyzed for 23 PAHs, Escherichia coli, fecal coliform, dissolved organic carbon (DOC), and total suspended solids (TSS). The fate and transport of PAHs within the bioretention columns was tracked by measuring soil PAHs in media cores taken from the columns. ΣPAH were almost completely removed by all treatments across all storms, with removal rates ranging from 97 to 100% for 94 out of 96 samples. Compost appeared to be a source of PAHs in bioretention media, as biochar-amended media initially contained half the ΣPAHs as treatments with the standard 60:40 sand:compost mixture. We observed a net loss of ΣPAHs (19-73%) in bioretention media across the study, which could not be explained by PAHs in the effluent, suggesting that bioremediation by microbes and/or plants attenuated media PAHs. E. coli and fecal coliform were exported in the first dosing event, but all columns achieved some treatment in subsequent dosing events. Overall, these findings suggest that PAHs in stormwater can be remediated with bioretention, are unlikely to accumulate in bioretention media, and that biochar amendments can improve the treatment of E. coli.

Differential Fairness: An Intersectional Framework for Fair AI.

We propose definitions of fairness in machine learning and artificial intelligence systems that are informed by the framework of intersectionality, a critical lens from the legal, social science, and humanities literature which analyzes how interlocking systems of power and oppression affect individuals along overlapping dimensions including gender, race, sexual orientation, class, and disability. We show that our criteria behave sensibly for any subset of the set of protected attributes, and we prove economic, privacy, and generalization guarantees. Our theoretical results show that our criteria meaningfully operationalize AI fairness in terms of real-world harms, making the measurements interpretable in a manner analogous to differential privacy. We provide a simple learning algorithm using deterministic gradient methods, which respects our intersectional fairness criteria. The measurement of fairness becomes statistically challenging in the minibatch setting due to data sparsity, which increases rapidly in the number of protected attributes and in the values per protected attribute. To address this, we further develop a practical learning algorithm using stochastic gradient methods which incorporates stochastic estimation of the intersectional fairness criteria on minibatches to scale up to big data. Case studies on census data, the COMPAS criminal recidivism dataset, the HHP hospitalization data, and a loan application dataset from HMDA demonstrate the utility of our methods.

Approximating Functions with Approximate Privacy for Applications in Signal Estimation and Learning.

Large corporations, government entities and institutions such as hospitals and census bureaus routinely collect our personal and sensitive information for providing services. A key technological challenge is designing algorithms for these services that provide useful results, while simultaneously maintaining the privacy of the individuals whose data are being shared. Differential privacy (DP) is a cryptographically motivated and mathematically rigorous approach for addressing this challenge. Under DP, a randomized algorithm provides privacy guarantees by approximating the desired functionality, leading to a privacy-utility trade-off. Strong (pure DP) privacy guarantees are often costly in terms of utility. Motivated by the need for a more efficient mechanism with better privacy-utility trade-off, we propose Gaussian FM, an improvement to the functional mechanism (FM) that offers higher utility at the expense of a weakened (approximate) DP guarantee. We analytically show that the proposed Gaussian FM algorithm can offer orders of magnitude smaller noise compared to the existing FM algorithms. We further extend our Gaussian FM algorithm to decentralized-data settings by incorporating the CAPE protocol and propose capeFM. Our method can offer the same level of utility as its centralized counterparts for a range of parameter choices. We empirically show that our proposed algorithms outperform existing state-of-the-art approaches on synthetic and real datasets.

Machine-learning model derived gene signature predictive of paclitaxel survival benefit in gastric cancer: results from the randomised phase III SAMIT trial.

OBJECTIVE: To date, there are no predictive biomarkers to guide selection of patients with gastric cancer (GC) who benefit from paclitaxel. Stomach cancer Adjuvant Multi-Institutional group Trial (SAMIT) was a 2×2 factorial randomised phase III study in which patients with GC were randomised to Pac-S-1 (paclitaxel +S-1), Pac-UFT (paclitaxel +UFT), S-1 alone or UFT alone after curative surgery. DESIGN: The primary objective of this study was to identify a gene signature that predicts survival benefit from paclitaxel chemotherapy in GC patients. SAMIT GC samples were profiled using a customised 476 gene NanoString panel. A random forest machine-learning model was applied on the NanoString profiles to develop a gene signature. An independent cohort of metastatic patients with GC treated with paclitaxel and ramucirumab (Pac-Ram) served as an external validation cohort. RESULTS: From the SAMIT trial 499 samples were analysed in this study. From the Pac-S-1 training cohort, the random forest model generated a 19-gene signature assigning patients to two groups: Pac-Sensitive and Pac-Resistant. In the Pac-UFT validation cohort, Pac-Sensitive patients exhibited a significant improvement in disease free survival (DFS): 3-year DFS 66% vs 40% (HR 0.44, p=0.0029). There was no survival difference between Pac-Sensitive and Pac-Resistant in the UFT or S-1 alone arms, test of interaction p<0.001. In the external Pac-Ram validation cohort, the signature predicted benefit for Pac-Sensitive (median PFS 147 days vs 112 days, HR 0.48, p=0.022). CONCLUSION: Using machine-learning techniques on one of the largest GC trials (SAMIT), we identify a gene signature representing the first predictive biomarker for paclitaxel benefit. TRIAL REGISTRATION NUMBER: UMIN Clinical Trials Registry: C000000082 (SAMIT); ClinicalTrials.gov identifier, 02628951 (South Korean trial).

Epigenetic promoter alterations in GI tumour immune-editing and resistance to immune checkpoint inhibition.

OBJECTIVES: Epigenomic alterations in cancer interact with the immune microenvironment to dictate tumour evolution and therapeutic response. We aimed to study the regulation of the tumour immune microenvironment through epigenetic alternate promoter use in gastric cancer and to expand our findings to other gastrointestinal tumours. DESIGN: Alternate promoter burden (APB) was quantified using a novel bioinformatic algorithm (proActiv) to infer promoter activity from short-read RNA sequencing and samples categorised into APBhigh, APBint and APBlow. Single-cell RNA sequencing was performed to analyse the intratumour immune microenvironment. A humanised mouse cancer in vivo model was used to explore dynamic temporal interactions between tumour kinetics, alternate promoter usage and the human immune system. Multiple cohorts of gastrointestinal tumours treated with immunotherapy were assessed for correlation between APB and treatment outcomes. RESULTS: APBhigh gastric cancer tumours expressed decreased levels of T-cell cytolytic activity and exhibited signatures of immune depletion. Single-cell RNAsequencing analysis confirmed distinct immunological populations and lower T-cell proportions in APBhigh tumours. Functional in vivo studies using 'humanised mice' harbouring an active human immune system revealed distinct temporal relationships between APB and tumour growth, with APBhigh tumours having almost no human T-cell infiltration. Analysis of immunotherapy-treated patients with GI cancer confirmed resistance of APBhigh tumours to immune checkpoint inhibition. APBhigh gastric cancer exhibited significantly poorer progression-free survival compared with APBlow (median 55 days vs 121 days, HR 0.40, 95% CI 0.18 to 0.93, p=0.032). CONCLUSION: These findings demonstrate an association between alternate promoter use and the tumour microenvironment, leading to immune evasion and immunotherapy resistance.

Privacy-preserving quality control of neuroimaging datasets in federated environments.

Privacy concerns for rare disease data, institutional or IRB policies, access to local computational or storage resources or download capabilities are among the reasons that may preclude analyses that pool data to a single site. A growing number of multisite projects and consortia were formed to function in the federated environment to conduct productive research under constraints of this kind. In this scenario, a quality control tool that visualizes decentralized data in its entirety via global aggregation of local computations is especially important, as it would allow the screening of samples that cannot be jointly evaluated otherwise. To solve this issue, we present two algorithms: decentralized data stochastic neighbor embedding, dSNE, and its differentially private counterpart, DP-dSNE. We leverage publicly available datasets to simultaneously map data samples located at different sites according to their similarities. Even though the data never leaves the individual sites, dSNE does not provide any formal privacy guarantees. To overcome that, we rely on differential privacy: a formal mathematical guarantee that protects individuals from being identified as contributors to a dataset. We implement DP-dSNE with AdaCliP, a method recently proposed to add less noise to the gradients per iteration. We introduce metrics for measuring the embedding quality and validate our algorithms on these metrics against their centralized counterpart on two toy datasets. Our validation on six multisite neuroimaging datasets shows promising results for the quality control tasks of visualization and outlier detection, highlighting the potential of our private, decentralized visualization approach.

Immune pathway upregulation and lower genomic instability distinguish EBV-positive nodal T/NK-cell lymphoma from ENKTL and PTCL-NOS.

Primary Epstein-Barr virus (EBV)-positive nodal T/NK-cell lymphoma (PTCL-EBV) is a poorly understood disease which shows features resembling extranodal NK/T-cell lymphoma (ENKTL) and is currently not recognized as a distinct entity but categorized as a variant of primary T-cell lymphoma not otherwise specified (PTCL-NOS). Herein, we analyzed copynumber aberrations (n=77) with a focus on global measures of genomic instability and homologous recombination deficiency and performed gene expression (n=84) and EBV miRNA expression (n=24) profiling as well as targeted mutational analysis (n=16) to further characterize PTCL-EBV in relation to ENKTL and PTCL-NOS. Multivariate analysis revealed that patients with PTCL-EBV had a significantly worse outcome compared to patients with PTCL-NOS (P=0.002) but not to those with ENKTL. Remarkably, PTCL-EBV exhibited significantly lower genomic instability and homologous recombination deficiency scores compared to ENKTL and PTCL-NOS. Gene set enrichment analysis revealed that many immune-related pathways, interferon α/γ response, and IL6_JAK_STAT3 signaling were significantly upregulated in PTCLEBV and correlated with lower genomic instability scores. We also identified that NFκB-associated genes, BIRC3, NFKB1 (P50) and CD27, and their proteins are upregulated in PTCL-EBV. Most PTCL-EBV demonstrated a type 2 EBV latency pattern and, strikingly, exhibited downregulated expression of most EBV miRNA compared to ENKTL and their target genes were also enriched in immune-related pathways. PTCL-EBV also showed frequent mutations of TET2, PIK3CD and STAT3, and are characterized by microsatellite stability. Overall, poor outcome, low genomic instability, upregulation of immune pathways and downregulation of EBV miRNA are distinctive features of PTCL-EBV. Our data support the concept that PTCL-EBV could be considered as a distinct entity, provide novel insights into the pathogenesis of the disease and offer potential new therapeutic targets for this tumor.