Predicting sepsis in-hospital mortality with machine learning: a multi-center study using clinical and inflammatory biomarkers.

BACKGROUND: This study aimed to develop and validate an interpretable machine-learning model that utilizes clinical features and inflammatory biomarkers to predict the risk of in-hospital mortality in critically ill patients suffering from sepsis. METHODS: We enrolled all patients diagnosed with sepsis in the Medical Information Mart for Intensive Care IV (MIMIC-IV, v.2.0), eICU Collaborative Research Care (eICU-CRD 2.0), and the Amsterdam University Medical Centers databases (AmsterdamUMCdb 1.0.2). LASSO regression was employed for feature selection. Seven machine-learning methods were applied to develop prognostic models. The optimal model was chosen based on its accuracy, F1 score and area under curve (AUC) in the validation cohort. Moreover, we utilized the SHapley Additive exPlanations (SHAP) method to elucidate the effects of the features attributed to the model and analyze how individual features affect the model's output. Finally, Spearman correlation analysis examined the associations among continuous predictor variables. Restricted cubic splines (RCS) explored potential non-linear relationships between continuous risk factors and in-hospital mortality. RESULTS: 3535 patients with sepsis were eligible for participation in this study. The median age of the participants was 66 years (IQR, 55-77 years), and 56% were male. After selection, 12 of the 45 clinical parameters collected on the first day after ICU admission remained associated with prognosis and were used to develop machine-learning models. Among seven constructed models, the eXtreme Gradient Boosting (XGBoost) model achieved the best performance, with an AUC of 0.94 and an F1 score of 0.937 in the validation cohort. Feature importance analysis revealed that Age, AST, invasive ventilation treatment, and serum urea nitrogen (BUN) were the top four features of the XGBoost model with the most significant impact. Inflammatory biomarkers may have prognostic value. Furthermore, SHAP force analysis illustrated how the constructed model visualized the prediction of the model. CONCLUSIONS: This study demonstrated the potential of machine-learning approaches for early prediction of outcomes in patients with sepsis. The SHAP method could improve the interoperability of machine-learning models and help clinicians better understand the reasoning behind the outcome.

MTBHsp70-exFPR1-pulsed Dendritic Cells Enhance the Immune Response against Cervical Cancer.

Cervical cancer is the most common malignancy of the female reproductive system. Dendritic cell (DC)-based immunological therapy is a novel treatment for this cancer. DCs are specialized antigen-presenting cells (APCs) in the human immune system, and they can activate the T cells used in tumor immunological therapy. In this study, we developed a novel immunotherapeutic peptide by linking the Mycobacterium tuberculosis (MTB) heat shock protein 70 (Hsp70) functional peptide to the extracellular domain of FPR1, a protein overexpressed in cervical cancer, to obtain an MTBHsp70-exFPR1 fusion protein. Our experiments confirmed that the MTBHsp70-exFPR1 protein could promote DC maturation and induce the secretion of IL-12p70, IL-1ß, and TNF-α. The antitumor effect of human cytotoxic T lymphocytes (CTLs) activated by autologous DCs was assessed in NOG mice. These results indicate that DCs pulsed with MTBHsp70-exFPR1 can enhance antitumor immunity against cervical cancer, providing a novel immune therapeutic strategy.

The prognosis impact of hyperthermic intraperitoneal chemotherapy (HIPEC) plus cytoreductive surgery (CRS) in advanced ovarian cancer: the meta-analysis.

BACKGROUND AND OBJECTIVE: Previous studies about the prognostic value of the HIPEC have yielded controversial results. Therefore, this study aims to assess the impact of HIPEC on patients with ovarian cancer. RESULTS: We included 13 comparative studies, and found that the overall survival (OS) and progression-free survival (PFS) in HIPEC groups were superior to groups without HIPEC treatment in the all total population (HR = 0.54,95% CI:0.45 to 0.66, HR = 0.45, 95% CI: 0.32 to 0.62). Additionally, the subgroup analysis showed that patients with advanced primary ovarian cancers also gained improved OS and PFS benefit from HIPEC (HR = 0.59,95% CI:0.46 to 0.75, HR = 0.41,95% CI:0.32 to 0.54). With regard to recurrent ovarian cancer, HIPEC was associated with improved OS (HR = 0.45,95% CI:0.24 to 0.83), but for the PFS, no correlation was observed between HIPC group and the non-HIPEC group (HR = 0.55,95% CI:0.27 to 1.11). HIPEC also led to favorable clinical outcome (HR = 0.64,95% CI:0.50 to 0.82, HR = 0.36,95% CI:0.20 to 0.65) for stage III or IV ovarian cancer with initial diagnosis. CONCLUSION: The review indicated that HIPEC-based regimens was correlated with better clinical prognosis for patients with primary ovarian cancers. For recurrent ovarian cancers, HIPEC only improved the OS but did not elicit significant value on the PFS.

Combinatorial therapy of immune checkpoint and cancer pathways provides a novel perspective on ovarian cancer treatment.

An increasing number of studies have reported that immunotherapy serves a significant role in ovarian cancer treatment. In recent years, blockade of checkpoint pathways, including programmed death-ligand 1 (PD-L1)/programmed death-1 and cytotoxic T-lymphocyte-associated protein 4, has demonstrated significant clinical and preclinical benefits in the treatment of ovarian cancer. Additionally, tumor-associated angiogenesis and homologous recombination deficiency frequently occurs in patients with high-grade ovarian cancer, which makes cancer cells more susceptible to targeted therapies, including therapies targeting poly (ADP-ribose) polymerase inhibitor, and anti-angiogenic approaches. Additionally, targeted therapy has been associated with elevated PD-L1 expression in tumor cells, increased T-cell infiltration in tumors and dendritic cell stimulation. This synergistic effect provides the rationale for the joint application of targeted therapy and immunotherapy. Checkpoint blockades are able to elicit durable antitumor immune reactions and complement the transient antitumor effect of targeted therapies. The current review discusses the underlying mechanism of these therapies and novel developments in combined therapy for the treatment of ovarian cancer.

pH/redox sensitive nanoparticles with platinum(iv) prodrugs and doxorubicin enhance chemotherapy in ovarian cancer.

pH/redox sensitive, dual drug loaded nanoparticles were prepared from poly(ethylene glycol)-block-poly(l-lysine) (PEG-b-PLL) for improving cancer therapy. Platinum(iv) and cis-aconitic anhydride-doxorubicin (CAD) were anchored to lysine residual amine groups of PLL to form polymer prodrug conjugates, which then self-assembled into nanoparticles with hydrophobic platinum(iv) prodrugs and CAD as the core. The nanoparticles were stable in neutral environments, but once under acidic and reductive conditions, the drugs were rapidly released. The dual-loaded nanoparticles had comparable intracellular toxicity to the regimen of combined application of free cisplatin and doxorubicin.

Simultaneous removal of Cr(VI) and acid orange 7 from water solution by dielectric barrier discharge plasma.

A feasibility study was conducted for simultaneous removal of hexavalent chromium (Cr(VI)) and azo dye acid orange 7 (AO7) by the dielectric barrier discharge (DBD) plasma. The results showed that there was a synergistic effect between Cr(VI) reduction and AO7 degradation. The presence of Cr(VI) enhanced the degradation efficiency of AO7. Meanwhile, the removal efficiency of Cr(VI) also increased in the presence of AO7. Under acidic conditions (pH = 3.0), the reduction efficiency of Cr(VI) was higher (approximately 94%). However, the presence of Cr(VI) diminished the effect of pH on the AO7 degradation efficiency. By increasing the input voltage from 80 to 120 V, the removal efficiencies of Cr(VI) and AO7 were observably increased from 54% to 88% and 62% to 89%, respectively. Adding organic matters inhibited the degradation of AO7 and promoted the reduction of Cr(VI). The addition of Cu(II), Co(II), Ni(II), Mn(II) and Fe(III) could inhibit the Cr(VI) reduction, but not significantly affect the degradation of AO7. The degradation intermediates of AO7 were identified by LC-MS/MS system and a possible degradation pathway was proposed. This study showed that the DBD plasma can simultaneously remove Cr(VI) and AO7, which provided a new idea for the actual wastewater treatment.

Decomposition of acetaminophen in water by a gas phase dielectric barrier discharge plasma combined with TiO2-rGO nanocomposite: Mechanism and degradation pathway.

Acetaminophen (APAP) served as the model pollutant to evaluate the feasibility of pollutant removal by gas phase dielectric barrier discharge plasma combined with the titanium dioxide-reduced Graphene Oxide (TiO2-rGO) nanocomposite. TiO2-rGO nanocomposite was prepared using the modified hydrothermal method and characterized by TEM and XPS before and after plasma process. The results indicated that the APAP degradation efficiency was significantly improved to 92% after 18min of discharge plasma treatment coupling 0.25gL-1 TiO2-rGO 5%wt at 18kV, compared with the plasma alone and plasma combined with P25 TiO2. The degradation mechanism for APAP in this system was studied by investigating the effects of the operational variables (e.g. discharge voltage and pH value) and the amount of the generated active species; and the results showed that O3 and H2O2 yields were influenced notably by adding TiO2-rGO. Also, it was observed that, compared with unused TiO2-rGO, the photocatalytic performance of used TiO2-rGO declined after several recirculation times due to the further reduction of Graphene Oxide in plasma system. Finally, intermediate products were analyzed by UV-vis spectrometry and HPLC/MS, and possible transformation pathways were identified with the support of theoretically calculating the frontier electron density of APAP.