EBioMedicine最新文献

Background: The Bari-SolidAct randomized controlled trial compared baricitinib with placebo in patients with severe COVID-19. A post hoc analysis revealed a higher incidence of serious adverse events (SAEs) among SARS-CoV-2-vaccinated participants who had received baricitinib. This sub-study aimed to investigate whether vaccination influences the safety profile of baricitinib in patients with severe COVID-19.

Methods: Biobanked samples from 146 participants (55 vaccinated vs. 91 unvaccinated) were analysed longitudinally for inflammation markers, humoral responses, tissue viral loads, and plasma viral antigens on days 1, 3, and 8. High-dimensional analyses, including RNA sequencing and flow cytometry, were performed on available samples. Mediation analyses were used to assess relationships between SAEs, baseline-adjusted biomarkers, and treatment-vaccination status.

Findings: Vaccinated participants were older, more frequently hospitalized, had more comorbidities, and exhibited higher nasopharyngeal viral loads. Baricitinib treatment did not affect antibody responses or viral clearance, but reduced markers of T-cell and monocyte activation compared to placebo (sCD25, sCD14, sCD163, sTIM-3). Age, baseline levels of plasma viral antigen, and several inflammatory markers, as well as IL-2, IL-6, Neopterin, CXCL16, sCD14, and suPAR on day 8 were associated with the occurrence of SAEs. However, mediation analyses of markers linked to SAEs, baricitinib treatment, or vaccination status did not reveal statistically significant interactions between vaccination status and SAEs.

Interpretation: This sub-study did not identify any virus- or host-related biomarkers significantly associated with the interaction between SARS-CoV-2 vaccination status and the safety of baricitinib. However, caution should be exercised due to the moderate sample size.

Funding: EU Horizon 2020 (grant number 101015736).

Background: We hypothesised that the implementation of an electronic medical record (EMR) embedded perioperative clinical decision support (CDS) application, Anesthesia Testing Guidelines (ATG), would result in at least a 10% reduction of unnecessary perioperative testing in patients undergoing elective surgeries.

Methods: The development and implementation of ATG occurred in several phases: 1) team development, 2) development of an embedded EMR application, 3) creation of ATG training and education toolkit, and 4) implementation involving promoting ATG through training and education, addressing challenges, and monitoring compliance. The proportions of patients with any overutilisation across 19 perioperative tests were compared between the baseline cohort and the ATG implementation cohort.

Findings: The overutilisation of perioperative tests was observed in 77.6% of the baseline cohort (n = 59,799) and 68.1% in the ATG cohort (n = 132,131), with a significant 12.2% reduction over two years of implementation (p < 0.0001). The two tests with the greatest amount of associated cost were reduced by 46% for chest X-rays and 39% for complete metabolic panels. The health system was able to reduce overall costs by 22% from baseline. Interrupted time series analysis estimated an immediate 7.5% decrease in monthly overutilisation when ATG initially launched, and it continued to decrease by 0.24% per month.

Interpretation: Our findings suggest CDS ATG is a successful tactic to reduce unnecessary preoperative testing while maintaining quality of care and improving cost avoidance. This type of CDS implementation approach transforms organisational behaviour and medical practices to follow defined guidelines, thereby improving the value of care.

Funding: CDS ATG was supported by UPMC Department of Anesthesiology and Perioperative Medicine and UPMC Department of Finance.

Background: Cancer is one of the leading causes of mortality worldwide, highlighting the urgent need for a deeper molecular understanding and the development of personalized treatments. The present study aims to establish a solid association between gene expression and patient survival outcomes to enhance the utility of the Human Pathology Atlas for cancer research.

Methods: In this updated analysis, we examined the expression profiles of 6918 patients across 21 cancer types. We integrated data from 10 independent cancer cohorts, creating a cross-validated, reliable collection of prognostic genes. We applied systems biology approach to identify the association between gene expression profiles and patient survival outcomes. We further constructed prognostic regulatory networks for kidney renal clear cell carcinoma (KIRC) and liver hepatocellular carcinoma (LIHC), which elucidate the molecular underpinnings associated with patient survival in these cancers.

Findings: We observed that gene expression during the transition from normal to tumorous tissue exhibited diverse shifting patterns in their original tissue locations. Significant correlations between gene expression and patient survival outcomes were identified in KIRC and LIHC among the major cancer types. Additionally, the prognostic regulatory network established for these two cancers showed the indicative capabilities of the Human Pathology Atlas and provides actionable insights for cancer research.

Interpretation: The updated Human Pathology Atlas provides a significant foundation for precision oncology and the formulation of personalized treatment strategies. These findings deepen our understanding of cancer biology and have the potential to advance targeted therapeutic approaches in clinical practice.

Funding: The Knut and Alice Wallenberg Foundation (72110), the China Scholarship Council (Grant No. 202006940003).

Background: Preimplantation genetic testing for chromosomal structural rearrangements (PGT-SR) has been widely utilized to select euploid embryos in patients carrying balanced chromosomal rearrangements (BCRs) by chromosome copy number analysis. However, reliable and extensively validated PGT-SR methods for selecting embryos without BCRs in large-cohort studies are lacking.

Methods: In this prospective, multicenter, cohort study, carriers with BCRs undergoing PGT-SR were recruited across 12 academic fertility centers within China. PGT-SR was performed using genome-wide SNP genotyping and haplotyping approach. Parental haplotypes were phased by available genotypes from a close relative or an unbalanced embryo. The karyotypes of embryos were inferred from the haplotypes. Only a single embryo was transferred in each cycle.

Findings: Between April 2018 and March 2023, 1298 carriers we randomly enrolled. A total of 7867 blastocysts from 1603 PGT-SR cycles were biopsied, in which 7750 (98.51%) were successfully genotyped and analyzed. Overall, 75.98% (1218/1603) of cycles obtained euploid embryos and 53.15% (852/1603) generated non-carrier embryos. The proportion of carrier and non-carrier embryos was similar in different subgroups. A total of 1030 non-carrier and 439 carrier embryos were transferred, 817 healthy babies were delivered cumulatively. Our results demonstrate that SNP-haplotyping method is highly accurate (sensitivity 95% CI: 98.34%-100%, specificity 95% CI: 96.63%-100%, respectively), and can be applied universally to different BCR types. Moreover, the clinical outcomes were comparable between the carrier and non-carrier embryo groups.

Interpretation: This study demonstrates the effectiveness of preimplantation genetic genome-wide SNP-genotyping and haplotyping method, resulting in the delivery of more babies with a normal karyotype.

Funding: This study was funded by the National Key Research and Development Program of China (2022YFC2703200, 2021YFC2700600, 2021YFC2700500), National Natural Science Foundation of China (82201807, 82171639, 82071717). Shanghai Science and Technology Innovation Action Plan Program (18411953800), and the Municipal Human Resources Development Program for Outstanding Young Talents in Medical and Health Sciences in Shanghai (2022YQ075).

Background: Evidence from preclinical studies in small and large animal models has shown neuroprotective effects of intravenous administration of umbilical cord blood derived cells (UCBCs). This study aimed to evaluate the feasibility of umbilical cord blood (UCB) collection, extraction of UCBCs, and subsequent safety of intravenous autologous administration of UCBCs in extremely preterm infants (born <28 weeks gestation).

Methods: A single-centre, open-label, single-arm, safety and feasibility clinical intervention trial was conducted at Monash Medical Centre and Monash Children's Hospital, Melbourne, Australia. Participants were extremely preterm infants born at less than 28 weeks completed gestation, and exclusions included major congenital malformation, maternal blood-borne virus infection, and severe brain injury on postnatal cranial ultrasound. UCB was collected at birth, and UCBCs were characterised (total nucleated cell count (TNC), mononuclear cell count (MNC), CD34+ cell count) and cryopreserved. Infants were reinfused with autologous UCBCs (25-50 million MNCs/kg) intravenously in the second postnatal week. Primary outcomes included feasibility: sufficient UCB volume (>7 mL) and UCBC numbers following processing (>25 × 10⁶ TNCs/kg); and safety: absence of adverse events directly related to UCBC administration.

Findings: Forty-four UCB collections were attempted and sufficient UCB volume/UCBC extraction was demonstrated in 37 (84.1%) infants. Good Manufacturing Practice (GMP) grade cells were obtained in 31/44 (70.4%) of infants. Median (IQR) TNCs and MNCs collected were 130 (67-207) x 10⁶/kg and 60 (39-105) x 10⁶/kg, respectively. 23 infants with median (IQR) gestation of 26 (24-27) weeks and birth weight of 761 (650-946) grams were administered cells at a median (IQR) dose of 42.3 (31.1-62.3) x 10⁶ MNCs/kg). No serious adverse events were noted, and the infusions were well-tolerated.

Interpretation: This phase-1 clinical trial has shown UCBC collection and reinfusion was feasible in approximately 70% of extremely preterm infants and was well tolerated without any serious adverse events.

Funding: Funding to support this study was obtained from National Health and Medical Research Council of Australia, Cerebral Palsy Alliance, National Stem Cell Foundation of Australia, and Lions Cord Blood Foundation.

Background: The standard malaria rapid diagnostic test (RDT) and newer ultra-sensitive RDT (uRDT) target Plasmodium falciparum histidine rich protein-2 (HRP2), which persists post-treatment. The duration of test positivity has not previously been studied in a low transmission setting.

Methods: We conducted a longitudinal cohort study in a low transmission setting in Namibia. RDT-positive individuals identified through passive and active case detection were treated and followed weekly for testing by RDT and uRDT, HRP2 quantification, quantitative PCR (qPCR) of parasitemia, and quantitative reverse transcriptase PCR (RT-PCR) of gametocytemia, until RDT and uRDT were negative for two consecutive weeks. Determinants of persistent positivity were identified using Cox proportional hazards models.

Findings: Among 137 participants with complete follow-up and no evidence of resurgence during follow-up, median duration of positivity was 42 days (range: 3-98 range) for RDT, compared to 67 days (range 12-105) for uRDT. In a sub-analysis of those with laboratory data before treatment (n = 60), drug resistance did not explain persistent positivity. Younger age (<15 years versus ≥15 years: aHR: 1.85, 95% CI 1.04-3.30, and 1.67, 95% CI 0.96-2.89, for RDT and uRDT, respectively), higher initial parasite density (highest versus lowest tertile: aHR 0.11, 95% CI 0.04-0.32 and 0.19, 95% CI 0.07-0.48 for RDT and uRDT, respectively), and persistent parasitemia (≥7 days versus reference of <7 days, aHR 0.39, 95% CI 0.20-0.76, and 0.40, 95% CI 0.21-0.76 for RDT and uRDT, respectively) were associated with longer duration of positivity.

Interpretation: Duration of RDT/uRDT positivity was more than double compared to reports from higher endemic settings, potentially due to lower population immunity to clear parasite DNA and antigen. Prolonged duration of positivity compromises their use to detect current infection, but increased detection of recent infection can facilitate surveillance and inform elimination efforts.

Funding: The project was funded by the Bill and Melinda Gates Foundation (A128488 and INV1135840), Horchow Family Fund (5300375400), and Chan Zuckerberg Biohub.

Background: Sepsis remains a leading cause of mortality in intensive care units. Understanding the dynamics of the plasma proteome of patients with sepsis is critical for improving prognostic and therapeutic strategies.

Methods: This prospective, multicentre observational cohort study included 363 patients with sepsis recruited from five university hospitals in Germany between March 2018 and April 2023. Plasma samples were collected on days 1 and 4 after sepsis diagnosis, and proteome analysis was performed using mass spectrometry. Classical statistical methods and machine learning (random forest) were employed to identify proteins associated with 30-day survival outcomes.

Findings: Out of 363 patients, 224 (62%) survived, and 139 (38%) did not survive the 30-day period. Proteomic analysis revealed significant differences in 87 proteins on day 1 and 95 proteins on day 4 between survivors and non-survivors. Additionally, 63 proteins were differentially regulated between day 1 and day 4 in the two groups. The identified protein networks were primarily related to blood coagulation, immune response, and complement activation. The random forest classifier achieved an area under the receiver operating characteristic curve of 0.75 for predicting 30-day survival. The results were compared and partially validated with an external sepsis cohort.

Interpretation: This study describes temporal changes in the plasma proteome associated with mortality in sepsis. These findings offer new insights into sepsis pathophysiology, emphasizing the innate immune system as an underexplored network, and may inform the development of targeted therapeutic strategies.

Funding: European Regional Development Fund of the European Union. The State of North Rhine-Westphalia, Germany.

Background: Understanding the mechanisms of algorithmic bias is highly challenging due to the complexity and uncertainty of how various unknown sources of bias impact deep learning models trained with medical images. This study aims to bridge this knowledge gap by studying where, why, and how biases from medical images are encoded in these models.

Methods: We systematically studied layer-wise bias encoding in a convolutional neural network for disease classification using synthetic brain magnetic resonance imaging data with known disease and bias effects. We quantified the degree to which disease-related information, as well as morphology-based and intensity-based biases were represented within the learned features of the model.

Findings: Although biases were encoded throughout the model, a stronger encoding did not necessarily lead to the model using these biases as a shortcut for disease classification. We also observed that intensity-based effects had a greater influence on shortcut learning compared to morphology-based effects when multiple biases were present.

Interpretation: We believe that these results constitute an important first step towards a deeper understanding of algorithmic bias in deep learning models trained using medical imaging data. This study also showcases the benefits of utilising controlled, synthetic bias scenarios for objectively studying the mechanisms of shortcut learning.

Funding: Alberta Innovates, Natural Sciences and Engineering Research Council of Canada, Killam Trusts, Parkinson Association of Alberta, River Fund at Calgary Foundation, Canada Research Chairs Program.