Prediction of medial knee contact force using multisource fusion recurrent neural network and transfer learning

Estimation of knee contact force (KCF) during gait provides essential information to evaluate knee joint function. Machine learning has been employed to estimate KCF because of the advantages of low computational cost and real-time. However, the existing machine learning models do not adequately consider gait-related data's temporal-dependent, multidimensional, and highly heterogeneous nature. This study is aimed at developing a multisource fusion recurrent neural network to predict the medial condyle KCF. First, a multisource fusion long short-term memory (MF-LSTM) model was established. Then, we developed a transfer learning strategy based on the MF-LSTM model for subject-specific medial KCF prediction. Four subjects with instrumented tibial prostheses were obtained from the literature. The results showed that the MF-LSTM model could predict medial KCF to a certain high level of accuracy (the mean of ρ = 0.970). The transfer learning model improved the prediction accuracy (the mean of ρ = 0.987). This study shows that the MF-LSTM model is a powerful and accurate computational tool for medial KCF prediction. Introducing transfer learning techniques could further improve the prediction performance for the target subject. This coupling strategy can help clinicians accurately estimate and track joint contact forces in real time.

[Application of three dimensional printed personalized guide plate assisted arthroscopic ankle arthrodesis in the treatment of ankle arthritis]

Objective: To compare the efficacy of conventional open ankle fusion and three dimensional(3D) printed guide plate assisted arthroscopic ankle fusion. Methods: A retrospective cohort study was performed on 256 patients with advanced traumatic ankle arthritis, who were admitted to the Department of Orthopaedics, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine from May 2018 to February 2023 and underwent ankle fusion procedures. The study cohort comprised 119 males and 137 females, with an age of (59.6±9.5) years (range: 37 to 83 years). Among them, 175 cases underwent internal fixation with plates and screws (58 cases through the combined medial and lateral approach, and 117 cases through the simple lateral approach), 48 cases underwent internal fixation with screws through the anterior approach (conventional open group), and 33 cases underwent minimally invasive arthroscopic ankle fusion assisted by 3D printed guide plate (3D printed guide plate arthroscopy group). Propensity score matching was employed to achieve a 1∶1 match(caliper value=0.02) between the baseline characteristics of patients in the 3D printed guide plate arthroscopy group and the conventional open group. Perioperative and follow-up data between the two groups were compared using the t-test, Mann-Whitney U test, Wilcoxon signed rank test,χ² test, or Fisher's exact probability method, as appropriate. Results: Matching was successfully achieved with 20 cases in both the 3D printed guide plate arthroscopy group and the conventional open group, and there were no statistically significant differences in baseline characteristics between the two groups (all P>0.05). The operation time in the 3D printed guide plate arthroscopy group was significantly longer than that in the conventional open group ((88.9±5.6) minutes vs. (77.9±11.7) minutes;t=-2.392, P=0.022), while the frequency of intraoperative fluoroscopies ((1.7±0.8) times vs. (5.2±1.2) times; t=10.604, P<0.01) and length of hospitalization ((5.5±0.9) days vs. (6.4±1.5) days;t=2.480, P=0.018) were significantly lower in the 3D printed guide plate arthroscopy group compared to the conventional open group. The fusion rate was 95.0% (19/20) in the 3D printed guide plate arthroscopy group and 85.0% (17/20) in the conventional open group, with no statistically significant difference between the two groups (χ²=1.111,P=0.605). The fusion time was (12.1±2.0) weeks in the conventional open group and (11.1±1.7) weeks in the 3D printed guide plate arthroscopy group, with no statistically significant difference between the two groups (t=1.607, P=0.116). At the final follow-up, the American Orthopedic Foot and Ankle Society ankle hindfoot scale was (72.6±5.5)points in the 3D printed guide plate arthroscopy group and (70.5±5.8)points in the conventional open group, with no statistically significant difference between the two groups (t=-1.003, P=0.322). The VAS score of the 3D printed guide plate arthroscopy group was (M(IQR)) 1.50 (1.00) points, lower than that of the conventional open group by 3.00 (1.00) points, with statistically significant differences (Z=-3.937, P<0.01). The complication rate was significantly higher in the conventional open group (25.0%(5/20) vs. 5.0%(1/20), P=0.182). Conclusion: 3D printed guide plate assisted arthroscopic ankle fusion exhibited several advantages, including reduced frequency of fluoroscopies, alleviation of postoperative pain, and decreased complications and length of hospitalization.

Ice sheet and precession controlled subarctic Pacific productivity and upwelling over the last 550,000 years

The polar oceans play a vital role in regulating atmospheric CO2 concentrations (pCO2) during the Pleistocene glacial cycles. However, despite being the largest modern reservoir of respired carbon, the impact of the subarctic Pacific remains poorly understood due to limited records. Here, we present high-resolution, 230Th-normalized export productivity records from the subarctic northwestern Pacific covering the last five glacial cycles. Our records display pronounced, glacial-interglacial cyclicity superimposed with precessional-driven variability, with warm interglacial climate and high boreal summer insolation providing favorable conditions to sustain upwelling of nutrient-rich subsurface waters and hence increased export productivity. Our transient model simulations consistently show that ice sheets and to a lesser degree, precession are the main drivers that control the strength and latitudinal position of the westerlies. Enhanced upwelling of nutrient/carbon-rich water caused by the intensification and poleward migration of the northern westerlies during warmer climate intervals would have led to the release of previously sequestered CO2 from the subarctic Pacific to the atmosphere. Our results also highlight the significant role of the subarctic Pacific in modulating pCO2 changes during the Pleistocene climate cycles, especially on precession timescale ( ~ 20 kyr).

Identification of two distinct head and neck squamous cell carcinoma subtypes based on fatty acid metabolism-related signatures: Implications for immunotherapy and chemotherapy

The dysregulation of lipid metabolism is a critical factor in the initiation and progression of tumors. In this investigation, we aim to characterize the molecular subtypes of head and neck squamous cell carcinoma (HNSCC) based on their association with fatty acid metabolism and develop a prognostic risk model. The transcriptomic and clinical data about HNSCC were obtained from public databases. Clustering analysis was conducted on fatty acid metabolism genes (FAMG) associated with prognosis, utilizing the non-negative matrix factorization algorithm. The immune infiltration, response to immune therapy, and drug sensitivity between molecular subtypes were evaluated. Differential expression genes were identified between subtypes, and a prognostic model was constructed using Cox regression analyses. A nomogram for HNSCC was constructed and evaluated. Thirty FAMGs have been found to exhibit differential expression in HNSCC, out of which three are associated with HNSCC prognosis. By performing clustering analysis on these 3 genes, 2 distinct molecular subtypes of HNSCC were identified that exhibit significant heterogeneity in prognosis, immune landscape, and treatment response. Using a set of 7778 genes that displayed differential expression between the 2 molecular subtypes, a prognostic risk model for HNSCC was constructed comprising 11 genes. This model has the ability to stratify HNSCC patients into high-risk and low-risk groups, which exhibit significant differences in prognosis, immune infiltration, and immune therapy response. Moreover, our data suggest that this risk model is negatively correlated with B cells and most T cells, but positively correlated with macrophages, mast cells, and dendritic cells. Ultimately, we constructed a nomogram incorporating both the risk signature and radiotherapy, which has demonstrated exceptional performance in predicting prognosis for HNSCC patients. A molecular classification system and prognostic risk models were developed for HNSCC based on FAMGs. This study revealed the potential involvement of FAMGs in modulating tumor immune microenvironment and response to treatment.

Comparative assessment of radiation therapy-induced vasculitis using [18F]FDG-PET/CT in patients with non-small cell lung cancer treated with proton versus photon radiotherapy

PURPOSE

To assess radiation therapy (RT)-induced vasculitis in patients with non-small cell lung cancer (NSCLC) by examining changes in the uptake of 18F-fluoro-D-deoxyglucose ([18F]FDG) by positron emission tomography/computed tomography (PET/CT) images of the ascending aorta (AA), descending aorta (DA), and aortic arch (AoA) before and after proton and photon RT.

METHOD

Thirty-five consecutive locally advanced NSCLC patients were definitively treated with proton (n = 27) or photon (n = 8) RT and concurrent chemotherapy. The patients were prospectively enrolled to undergo [18F]FDG-PET/CT imaging before and 3 months after RT. An adaptive contrast-oriented thresholding algorithm was applied to generate mean standardized uptake values (SUVmean) for regions of interest (ROIs) 3 mm outside and 3 mm inside the outer perimeter of the AA, DA, and AoA. These ROIs were employed to exclusively select the aortic wall and remove the influence of blood pool activity. SUVmeans before and after RT were compared using two-tailed paired t-tests.

RESULTS

RT treatments were associated with increased SUVmeans in the AA, DA, and AoA-1.9%, 0.3%, and 1.3% for proton and 15.8%, 9.5%, and 15.5% for photon, respectively. There was a statistically significant difference in the ∆SUVmean (post-RT SUVmean - pre-RT SUVmean) in patients treated with photon RT when compared to ∆SUVmean in patients treated with proton RT in the AA (p = 0.043) and AoA (p = 0.015). There was an average increase in SUVmean that was related to dose for photon patients (across structures), but that was not seen for proton patients, although the increase was not statistically significant.

CONCLUSION

Our results suggest that patients treated with photon RT for NSCLC may exhibit significantly more RT-induced inflammation (measured as ∆SUVmean) in the AA and AoA when compared to patients who received proton RT. Knowledge gained from further analyses in larger cohorts could aid in treatment planning and help prevent the significant morbidity and mortality associated with RT-induced vascular complications.

TRIAL REGISTRATION

NCT02135679.

A machine learning stacking model accurately estimating gastric fluid volume in patients undergoing elective sedated gastrointestinal endoscopy

BACKGROUND

The current point-of-care ultrasound (POCUS) assessment of gastric fluid volume primarily relies on the traditional linear approach, which often suffers from moderate accuracy. This study aimed to develop an advanced machine learning (ML) model to estimate gastric fluid volume more accurately.

METHODS

We retrospectively analyzed the clinical data and POCUS data (D1: craniocaudal diameter, D2: anteroposterior diameter) of 1386 patients undergoing elective sedated gastrointestinal endoscopy (GIE) at Nanjing First Hospital to predict gastric fluid volume using ML techniques, including six different ML models and a stacking model. We evaluated the models using the adjusted Coefficient of Determination (R2), mean absolute error (MAE) and root mean square error (RMSE). The SHapley Additive exPlanations (SHAP) method was used to interpret the importance of the variables. Finally, a web calculator was constructed to facilitate its clinical application.

RESULTS

The stacking model (Linear regression + Multilayer perceptron) performed best, with the highest adjusted R2 of 0.718 (0.632 to 0.804). The mean prediction bias was 4 ml (MAE: 4.008 (3.68 to 4.336)), which is better than that of the linear model. D1 and D2 ranked high in the SHAP plot and performed better in the right lateral decubitus (RLD) than in the supine position. The web calculator can be accessed at https://cheason.shinyapps.io/Stacking_regressor/.

CONCLUSION

The stacking model and its web calculator can serve as practical tools for accurately estimating gastric fluid volume in patients undergoing elective sedated GIE. It is recommended that anesthesiologists measure D1 and D2 in the patient's RLD position.

Hypoxia-derived molecular subtype and gene signature characterize prognoses and therapeutic responses in head and neck squamous cell carcinoma

Intratumoral hypoxia is widely associated with the development of malignancy, treatment resistance, and worse prognoses. This study aims to investigate the role of hypoxia-related genes (HRG) in the immune landscape, treatment response, and prognosis of head and neck squamous cell carcinoma (HNSCC). The transcriptome and clinical data of HNSCC were downloaded from TCGA and GEO databases, and HNSCC molecular subtypes were identified using non-negative matrix factorization (NMF) clustering. Prognostic models were constructed using univariate, Lasso, and multivariate Cox regression analyses. The relationship between HRGs and immune cell infiltration, immune therapy response, and drug sensitivity was evaluated, and a nomogram was constructed. 47 HRGs were differentially expressed in HNSCC, among which 10 genes were significantly associated with HNSCC prognosis. Based on these 10 genes, 2 HNSCC molecular subtypes were identified, which showed significant heterogeneity in terms of prognosis, immune infiltration, and treatment response. A total of 3280 differentially expressed genes were identified between the subtypes. After univariate, Lasso, and multivariate Cox regression analysis, 18 genes were selected to construct a novel prognostic model, which showed a significant correlation with B cells, T cells, and macrophages. Using this model, HNSCC was classified into high-risk and low-risk groups, which exhibited significant differences in terms of prognosis, immune cell infiltration, immune therapy response, and drug sensitivity. Finally, a nomogram based on this model and radiotherapy was constructed, which showed good performance in predicting HNSCC prognosis and guiding personalized treatment strategies. The decision curve analysis demonstrated its better clinical applicability compared to other strategies. HRGs can identify 2 HNSCC molecular subtypes with significant heterogeneity, and the HRG-derived risk model has the potential for prognostic prediction and guiding personalized treatment strategies.

Development and validation of a novel nomogram model to assess the risk of gastric contents in outpatients undergoing elective sedative gastrointestinal endoscopy procedures

BACKGROUND

Gastric contents may contribute to patients' aspiration during anesthesia. Ultrasound can accurately assess the risk of gastric contents in patients undergoing sedative gastrointestinal endoscopy (GIE) procedures, but its efficiency is limited. Therefore, developing an accurate and efficient model to predict gastric contents in outpatients undergoing elective sedative GIE procedures is greatly desirable.

METHODS

This study retrospectively analyzed 1501 patients undergoing sedative GIE procedures. Gastric contents were observed under direct gastroscopic vision and suctioned through the endoscope. High-risk gastric contents were defined as having solid content or liquid volume > 25 ml and pH < 2.5; otherwise, they were considered low-risk gastric contents. Univariate analysis and multivariate analysis were used to select the independent risk factors to predict high-risk gastric contents. Based on the selected independent risk factors, we assigned values to each independent risk factor and established a novel nomogram. The performance of the nomogram was verified in the testing cohort by the metrics of discrimination, calibration, and clinical usefulness. In addition, an online accessible web calculator was constructed.

RESULTS

We found BMI, cerebral infarction, cirrhosis, male, age, diabetes, and gastroesophageal reflux disease were risk factors for gastric contents. The AUROCs were 0.911 and 0.864 in the development and testing cohort, respectively. Moreover, the nomogram showed good calibration ability. Decision curve analysis and Clinical impact curve demonstrated that the predictive nomogram was clinically useful. The website of the nomogram was https://medication.shinyapps.io/dynnomapp/.

CONCLUSIONS

This study demonstrates that clinical variables can be combined with algorithmic techniques to predict gastric contents in outpatients. Nomogram was constructed from routine variables, and the web calculator had excellent clinical applicability to assess the risk of gastric contents accurately and efficiently in outpatients, assist anesthesiologists in assessment and identify the most appropriate patients for ultrasound.

Online interpretable dynamic prediction models for postoperative delirium after cardiac surgery under cardiopulmonary bypass developed based on machine learning algorithms: A retrospective cohort study

OBJECTIVE

Postoperative delirium (POD) is strongly associated with poor early and long-term prognosis in cardiac surgery patients with cardiopulmonary bypass (CPB). This study aimed to develop dynamic prediction models for POD after cardiac surgery under CPB using machine learning (ML) algorithms.

METHODS

From July 2021 to June 2022, clinical data were collected from patients undergoing cardiac surgery under CPB at Nanjing First Hospital. A dataset from the same center (October 2022 to November 2022) was also used for temporal external validation. We used ML and deep learning to build models in the training set, optimized parameters in the test set, and finally validated the best model in the validation set. The SHapley Additive exPlanations (SHAP) method was introduced to explain the best models.

RESULTS

Of the 885 patients enrolled, 221 (25.0%) developed POD. 22 (22.0%) of 100 validation cohort patients developed POD. The preoperative and postoperative artificial neural network (ANN) models exhibited optimal performance. The validation results demonstrated satisfactory predictive performance of the ANN model, with area under the receiver operator characteristic curve (AUROC) values of 0.776 and 0.684 for the preoperative and postoperative models, respectively. Based on the ANN algorithm, we constructed dynamic, highly accurate, and interpretable web risk calculators for POD.

CONCLUSIONS

We successfully developed online interpretable dynamic ANN models as clinical decision aids to identify patients at high risk of POD before and after cardiac surgery to facilitate early intervention or care.

Machine learning-based risk prediction of hypoxemia for outpatients undergoing sedation colonoscopy: a practical clinical tool

OBJECTIVES

Hypoxemia as a common complication in colonoscopy under sedation and may result in serious consequences. Unfortunately, a hypoxemia prediction model for outpatient colonoscopy has not been developed. Consequently, the objective of our study was to develop a practical and accurate model to predict the risk of hypoxemia in outpatient colonoscopy under sedation.

METHODS

In this study, we included patients who received colonoscopy with anesthesia in Nanjing First Hospital from July to September 2021. Risk factors were selected through the least absolute shrinkage and selection operator (LASSO). Prediction models based on logistic regression (LR), random forest classifier (RFC), extreme gradient boosting (XGBoost), support vector machine (SVM), and stacking classifier (SCLF) model were implemented and assessed by standard metrics such as the area under the receiver operating characteristic curve (AUROC), sensitivity and specificity. Then choose the best model to develop an online tool for clinical use.

RESULTS

We ultimately included 839 patients. After LASSO, body mass index (BMI) (coefficient = 0.36), obstructive sleep apnea-hypopnea syndrome (OSAHS) (coefficient = 1.32), basal oxygen saturation (coefficient = -0.14), and remifentanil dosage (coefficient = 0.04) were independent risk factors for hypoxemia. The XGBoost model with an AUROC of 0.913 showed the best performance among the five models.

CONCLUSION

Our study selected the XGBoost as the first model especially for colonoscopy, with over 95% accuracy and excellent specificity. The XGBoost includes four variables that can be quickly obtained. Moreover, an online prediction practical tool has been provided, which helps screen high-risk outpatients with hypoxemia swiftly and conveniently.

A novel interpretative tool for early prediction of low cardiac output syndrome after valve surgery: online machine learning models

OBJECTIVE

Low cardiac output syndrome (LCOS) is a severe complication after valve surgery, with no uniform standard for early identification. We developed interpretative machine learning (ML) models for predicting LCOS risk preoperatively and 0.5 h postoperatively for intervention in advance.

METHODS

A total of 2218 patients undergoing valve surgery from June 2019 to Dec 2021 were finally enrolled to construct preoperative and postoperative models. Logistic regression, support vector machine (SVM), random forest classifier, extreme gradient boosting, and deep neural network were executed for model construction, and the performance of models was evaluated by area under the curve (AUC) of the receiver operating characteristic and calibration curves. Our models were interpreted through SHapley Additive exPlanations, and presented as an online tool to improve clinical operability.

RESULTS

The SVM algorithm was chosen for modeling due to better AUC and calibration capability. The AUCs of the preoperative and postoperative models were 0.786 (95% CI 0.729-0.843) and 0.863 (95% CI 0.824-0.902), and the Brier scores were 0.123 and 0.107. Our models have higher timeliness and interpretability, and wider coverage than the vasoactive-inotropic score, and the AUC of the postoperative model was significantly higher. Our preoperative and postoperative models are available online at http://njfh-yxb.com.cn:2022/lcos.

CONCLUSIONS

The first interpretable ML tool with two prediction periods for online early prediction of LCOS risk after valve surgery was successfully built in this study, in which the SVM model has the best performance, reserving enough time for early precise intervention in critical care.

A convenient machine learning model to predict full stomach and evaluate the safety and comfort improvements of preoperative oral carbohydrate in patients undergoing elective painless gastrointestinal endoscopy

BACKGROUND AND AIMS

Assessment of the patient's gastric contents is the key to avoiding aspiration incidents, however, there is no effective method to determine whether elective painless gastrointestinal endoscopy (GIE) patients have a full stomach or an empty stomach. And previous studies have shown that preoperative oral carbohydrates (POCs) can improve the discomfort induced by fasting, but there are different perspectives on their safety. This study aimed to develop a convenient, accurate machine learning (ML) model to predict full stomach. And based on the model outcomes, evaluate the safety and comfort improvements of POCs in empty- and full stomach groups.

METHODS

We enrolled 1386 painless GIE patients between October 2022 and January 2023 in Nanjing First Hospital, and 1090 patients without POCs were used to construct five different ML models to identify full stomach. The metrics of discrimination and calibration validated the robustness of the models. For the best-performance model, we further interpreted it through SHapley Additive exPlanations (SHAP) and constructed a web calculator to facilitate clinical use. We evaluated the safety and comfort improvements of POCs by propensity score matching (PSM) in the two groups, respectively.

RESULTS

Random Forest (RF) model showed the greatest discrimination with the area under the receiver operating characteristic curve (AUROC) 0.837 [95% confidence interval (CI): 79.1-88.2], F1 71.5%, and best calibration with a Brier score of 15.2%. The web calculator can be visited at https://medication.shinyapps.io/RF_model/. PSM results demonstrated that POCs significantly reduced the full stomach incident in empty stomach group (p < 0.05), but no differences in full stomach group (p > 0.05). Comfort improved in both groups and was more significant in empty stomach group.

CONCLUSIONS

The developed convenient RF model predicted full stomach with high accuracy and interpretability. POCs were safe and comfortably improved in both groups, with more benefit in empty stomach group. These findings may guide the patients' gastrointestinal preparation.

Dynamic prediction of hypoxemia risk at different time points based on preoperative and intraoperative features: machine learning applications in outpatients undergoing esophagogastroduodenoscopy

BACKGROUND

Hypoxemia often occurs in outpatients undergoing anesthesia-assisted esophagogastroduodenoscopy (EGD). However, there is a scarcity in tools to predict the hypoxemia risk. We aimed to solve this problem by developing and validating machine learning (ML) models based on preoperative and intraoperative features.

METHODS

All data were retrospectively collected from June 2021 to February 2022. The most appropriate predictive features were selected by the least absolute shrinkage and selection operator, which were incorporated and modelled by 4 ML algorithms. The area under the precision-recall curve (AUPRC) was used as the main evaluation metric to select the best models, and the selected models were compared with the STOP-BANG score. Their predictive performance was visually interpreted by SHapley Additive exPlanations. The primary endpoint of this study was hypoxemia during the procedure, defined as at least one reading of pulse oximetry < 90% without probes misplacement from the anesthesia induction beginning to the end of EGD, while the secondary endpoint was hypoxemia during induction, from the induction beginning to the start of endoscopic intubation.

RESULTS

Of 1160 patients in the derivation cohort, 112 patients (9.6%) developed intraoperative hypoxemia, of which 102 (8.8%) occurred during the induction period. In temporal and external validation, no matter whether based on preoperative variables or still based on preoperative plus intraoperative variables, our models showed excellent predictive performance for the two endpoints, significantly better than STOP-BANG score. In the model interpretation section, preoperative variables (airway assessment indicators, pulse oximeter oxygen saturation and BMI) and intraoperative variables (the induced propofol dose) made the highest contribution to the predictions.

To our knowledge, our ML models were the first to predict hypoxemia risk, which achieved excellent overall predictive ability integrating various clinical indicators. These models have the potential to become an effective tool for adjusting sedation strategies flexibly and reducing the workload of anesthesiologists.KEY MESSAGESThis study is the first model employing ML methods based on preoperative and preoperative plus intraoperative variables for predicting the risk of hypoxemia during induction and the whole EGD procedure respectively.Our four models achieved satisfactory predictive performance and outperformed STOP-BANG score in terms of AUPRC in the temporal and external validation cohorts respectively.We found that the relevant variables of airway assessment should be fully taken into account when analyzing the risk factor of hypoxemia, and the effect of patients' age on their hypoxemia risk should be considered in conjunction with the propofol dose.

A risk prediction model based on machine learning for postoperative cognitive dysfunction in elderly patients with non-cardiac surgery

BACKGROUND

Early identification of elderly patients undergoing non-cardiac surgery who may be at high risk for postoperative cognitive dysfunction (POCD) can increase the chances of prevention for them, as extra attention and limited resources can be allocated more to these patients.

AIM

We performed this analysis with the aim of developing a simple, clinically useful machine learning (ML) model to predict the probability of POCD at 3 months in elderly patients after non-cardiac surgery.

METHODS

We collected information on patients who received surgical treatment at Nanjing First Hospital from May 2020 to May 2021. We used LASSO regression to select key features and built 5 ML models to assess the risk of POCD at 3 months in elderly patients after non-cardiac surgery. The Shapley Additive exPlanations (SHAP) and methods were introduced to interpret the best model.

RESULTS

A total of 415 patients with non-cardiac surgery were included. The support vector machine (SVM) was the best-performing model of the five ML models. The model showed excellent performance compared to the other four models. The SHAP results showed that VAS score, age, intraoperative hypotension, and preoperative hemoglobin were the four most important features, indicating that the SVM model had good interpretability and reliability. The website of the web-based calculator was https://modricreagan-non-3-pocd-9w2q78.streamlit.app/ .

CONCLUSION

Based on six important perioperative variables, we successfully established a series of ML models for predicting POCD occurrence at 3 months after surgery in elderly non-cardiac patients, with SVM model being the best-performing model. Our models are expected to serve as decision aids for clinicians to monitor screened high-risk patients more closely or to consider further interventions.

Dynamic and visual nomograms to online predict unfavorable outcome of mechanical thrombectomy for acute basilar artery occlusion

BACKGROUND

The evidence of mechanical thrombectomy (MT) in basilar artery occlusion (BAO) was limited. This study aimed to develop dynamic and visual nomogram models to predict the unfavorable outcome of MT in BAO online.

METHODS

BAO patients treated with MT were screened. Preoperative and postoperative nomogram models were developed based on clinical parameters and imaging features. An independent dataset was collected to perform external validation. Web-based calculators were constructed to provide convenient access.

RESULTS

A total of 127 patients were included in the study, and 117 of them were eventually included in the analysis. The nomogram models showed robust discrimination, with an area under the receiver operating characteristic (ROC) of 0.841 (preoperative) and 0.916 (postoperative). The calibration curves showed good agreement. The preoperative predictors of an unfavorable outcome were previous stroke, the National Institutes of Health Stroke Scale (NIHSS) at admission, and the posterior circulation Alberta Stroke Program Early Computed Tomography Score (pc-ASPECTS). The postoperative predictors were previous stroke, NIHSS at 24 h, and pc-ASPECTS.

CONCLUSION

Dynamic and visual nomograms were constructed and validated for the first time for BAO patients treated with MT, which provided precise predictions for the risk of an unfavorable outcome. The preoperative model may assist clinicians in selecting eligible patients, and the postoperative model may facilitate individualized poststroke management.

[Progress in intelligent antibacterial materials for dental caries]

Anti-caries strategies that based on the regulation of oral micro-ecology have recently drawn broad attention. Intelligent antibacterial materials have shown great potential for ecological anti-caries strategies, which can response to microenvironment of dental caries or external stimuli and inhibit cariogenic biofilms precisely. This technology could improve local anti-caries effect and help maintain oral micro-eubiosis. Here, we reviewed recent progress in intelligent anti-bacterial materials for dental caries. The future research direction was also prospected. We hope that by discussing about this new technology of prevention and treatment for dental caries, this review could provide ideas for the research on novel anti-caries materials.

Incorporating intraoperative blood pressure time-series variables to assist in prediction of acute kidney injury after type a acute aortic dissection repair: an interpretable machine learning model

BACKGROUND

Acute kidney injury (AKI) is a common and serious complication after the repair of Type A acute aortic dissection (TA-AAD). However, previous models have failed to account for the impact of blood pressure fluctuations on predictive performance. This study aims to develop machine learning (ML) models combined with intraoperative medicine and blood pressure time-series data to improve the accuracy of early prediction for postoperative AKI risk.

METHODS

Indicators reflecting the duration and depth of hypotension were obtained by analyzing continuous mean arterial pressure (MAP) monitored intraoperatively with multiple thresholds (<65, 60, 55, 50) set in the study. The predictive features were selected by logistic regression and the least absolute shrinkage and selection operator (LASSO), and 4 ML models were built based on the above features. The performance of the models was evaluated by area under receiver operating characteristic curve (AUROC), calibration curve and decision curve analysis (DCA). Shapley additive interpretation (SHAP) was used to explain the prediction models.

RESULTS

Among the indicators reflecting intraoperative hypotension, 65 mmHg showed a statistically superior difference to other thresholds in patients with or without AKI (p < .001). Among 4 models, the extreme gradient boosting (XGBoost) model demonstrated the highest AUROC: 0.800 (95% 0.683-0.917) and sensitivity: 0.717 in the testing set and was verified the best-performing model. The SHAP summary plot indicated that intraoperative urine output, cumulative time of mean arterial pressure lower than 65 mmHg outside cardiopulmonary bypass (OUT_CPB_MAP_65 time), autologous blood transfusion, and smoking were the top 4 features that contributed to the prediction model.

CONCLUSION

With the introduction of intraoperative blood pressure time-series variables, we have developed an interpretable XGBoost model that successfully achieve high accuracy in predicting the risk of AKI after TA-AAD repair, which might aid in the perioperative management of high-risk patients, particularly for intraoperative hemodynamic regulation.

Clinical Validation of HPV ctDNA for Early Detection of Residual Disease Following Chemoradiation in Cervical Cancer

PURPOSE/OBJECTIVE(S)

Despite chemoradiation (CRT), 30-40% of patients with locally advanced cervical cancer relapse. Most cases are caused by human papilloma virus (HPV), and HPV circulating tumor DNA (ctDNA) may identify patients at highest risk of relapse. Our previous pilot study showed that detectable HPV ctDNA at the end of CRT is associated with inferior progression-free survival (PFS) using digital polymerase chain reaction (dPCR), and that a next generation sequencing approach (HPV-seq) may outperform dPCR. We hypothesized that HPV ctDNA may identify cervical cancer patients at increased risk of relapse following CRT and aimed to prospectively validate HPV ctDNA as a tool for early detection of residual disease.

MATERIALS/METHODS

This prospective, multicenter validation study accrued 70 patients with HPV+ stage IB-IVA cervical cancer treated with definitive CRT from 2017-2022. Patients underwent phlebotomy at baseline, end of, 4-6 weeks and 3 months post CRT for HPV ctDNA levels. HPV genotyping was performed on the baseline plasma sample using HPV-seq. HPV genotype-specific DNA levels in plasma were quantified using both dPCR and HPV-seq. PFS was estimated using the Kaplan-Meier method and compared using the log rank test. Multivariable Cox regression analyses incorporating stage and HPV ctDNA detectability assessed independent prognostic factors associated with PFS.

RESULTS

At the time of abstract, results for 67 patients were available. The majority had squamous histology (84%) and stage IIB (36%) or IIIC1 (25%) disease. HPV genotyping using HPV-seq revealed 54% (36/67) of cases harboring HPV-16, and 46% harboring other HPV types: 15 HPV-18; 5 HPV-59; 2 HPV-31; 2 HPV-33; 2 HPV-52; 1 each HPV-39, HPV-45, HPV-53, HPV-58, and HPV-82. With a median follow up of 2.2 (range 0.4 - 5.2) years, there were 21 PFS events. Most recurrences (14/21) were distant and/or paraaortic; 4 local and nodal/distant; 2 pelvic nodal; and 1 local. Patients with detectable HPV ctDNA on dPCR at the end of, 4-6 weeks and 3 months post CRT had significantly worse 2-year PFS compared to those with undetectable HPV ctDNA (78 vs 52%, p = 0.04; 82 vs 26%, p < 0.001; and 80 vs 23%, p = < 0.001, respectively). HPV-seq showed similar results (87 vs 55%, p = 0.02; 81 vs 45%, p = 0.003; and 84 vs 31%, p = < 0.001, respectively). On multivariable analyses, detectable HPV ctDNA on dPCR and HPV-seq remained independently associated with inferior PFS (see table).

CONCLUSION

HPV-seq enables HPV genotyping directly from plasma in locally advanced cervical cancer. Persistent HPV ctDNA following CRT is independently associated with inferior PFS in this prospective validation study. HPV ctDNA testing can be used to identify, as early as at the end of CRT, patients at high risk of recurrence in future treatment intensification trials.

Effectiveness of Cisplatin in P16+ Oropharyngeal Cancer According to Relative Risk for Cancer Events: Ancillary Analysis of RTOG 1016

PURPOSE/OBJECTIVE(S)

To test the hypothesis that the effectiveness of cisplatin in p16+ oropharyngeal squamous cell carcinoma (OPSCC) increases with patients' relative risk for cancer events.

MATERIALS/METHODS

Ancillary analysis of 805 patients enrolled on RTOG 1016 accessed via Project DataSphere. Eligible patients had p16+ OPSCC, AJCC 7th T1-T2 N2a-N3 or T3-T4 N0-N3 M0, ECOG PS 0-1. Patients were randomized to RT with concurrent cisplatin vs. cetuximab. Relative risk for competing events was quantified using the Head and Neck Cancer Intergroup predictive classifier (omega score). Higher scores indicate higher relative risk for cancer events (LRF or distant metastasis) vs. competing mortality. We compared this to favorable, unfavorable/low, and unfavorable/intermediate risk groups using standard criteria: NRG HN005 eligible/low RTOG risk (Ang et al.), HN005 ineligible/low RTOG risk, and intermediate RTOG risk. Omega score cutoffs were selected to match numbers in standard risk strata. HRs for the effect of cisplatin vs. cetuximab on PFS and OS were compared for standard vs. relative risk strata. 1-tailed interaction tests were used to test whether cisplatin effectiveness increased within risk strata.

RESULTS

There were 354, 219, and 232 patients in standard favorable, unfavorable/low, and unfavorable/intermediate risk groups. Omega score cutoffs were 0.80 and 0.84 to define low, intermediate, and high relative risk groups. Discordant standard vs. relative risk classifications occurred in 559 patients (69.4%). Increasing omega score was associated with significantly higher relative HR (rHR) for cancer events (3.40, 95% CI: 1.66-6.96) and increasing effectiveness of cisplatin vs. cetuximab (Table), but standard risk grouping was not (rHR 0.80, 95% CI: 0.49-1.32). The effect of cisplatin on PFS significantly increased with higher omega score (interaction -0.30, p = .046), but decreased with increasing standard risk strata (interaction +0.27, p = NS).

CONCLUSION

The effectiveness of cisplatin in p16+ OPSCC increased with higher omega score but not with standard risk group. Relative risk for cancer events should be taken into account when designing deintensification strategies.

Analysis of Institutional DIBH Coaching Program for Surface Guided-DIBH Patients

PURPOSE/OBJECTIVE(S)

Our institute has implemented a surface guided-DIBH (SG-DIBH) coaching program which involves consultation, pre-treatment (CT-sim) and treatment. We would like to analyze the effectiveness of the program.

MATERIALS/METHODS

A total of 72 left breast cancer patients between 1st Apr 2022 to 9th Dec 2022 were registered for radiation treatment. During consultation, oncologist selected suitable patients based on the following criteria: a) age of 18-70; b) left breast cancer, right breast cancer with internal mammary nodes irradiation or dextrocardia; c) no lung/cardiac disease history & d) volunteer for SG-DIBH technique. The eligible patients were then trained by the coaching therapist using a teaching video and practiced at home. During CT simulation, patients were assessed according to the DIBH evaluation form. The evaluation components included patient's compliancy and understanding, the differences of lateral skin marking (free breathing, FB vs DIBH), duration of breath hold and reproducibility. Patients who passed the evaluation were scanned under both FB and DIBH for SG-DIBH treatment. IMRT-FFF 6 to 7 fields were planned. During SG-DIBH treatment, first 3 fractions and weekly CBCT were taken. Patients were encouraged to continue DIBH practice at home throughout whole course of the treatment and they were given 3 identical survey forms (5 questions) at the beginning, middle and end of treatment. The measure for the success of this coaching program would be number of breath-holds, duration of treatment time, treatment accuracy (CBCT matching) and survey results.

RESULTS

There were 48 patients who were eligible for DIBH coaching program, however, only 24 patients had passed the coaching evaluation and 20 patients were treated with SG-DIBH technique successfully. The mean of number of breath-hold and treatment time was 7 times and 7.9 minutes. Total of 123 CBCT images were studied. The setup errors were (0.242±0.180) cm, (0.152±0.137) cm, (0.202±0.165) cm, (0.684±0.640) degrees, (0.816±0.767) degrees, (0.912±0.707) degrees in lateral, longitudinal, vertical, pitch, roll and yaw directions. According to the survey analysis, the number of times to practice at home decreased as the treatment went by. An improvement was seen in patients' self-evaluation in mastering DIBH technique with proper coaching program (from 60% to 90%). Patients' anxiety in performing DIBH were alleviated greatly towards the end of the treatment (from 47% to 15%). 100% of the patients were willing to go for DIBH treatment if given a second chance and additional suggestions claimed that professional clinical teams and coaching program were important for their DIBH treatment journey.

CONCLUSION

A comprehensive DIBH coaching program can effectively identify SG-DIBH patient's suitability. Patient compliancy, treatment accuracy and treatment experience can be enhanced with good coaching program. The involvement of clinical team from consultation to pre-treatment and treatment stage is essential for a successful SG-DIBH treatment.

Comparison of Standard vs. Relative Risk Models to Define Candidates for Deintensification in Locoregionally Advanced P16+ Oropharyngeal Cancer

PURPOSE/OBJECTIVE(S)

Various methods to identify candidates for treatment deintensification with p16+ oropharyngeal squamous cell carcinoma (OPSCC) have been used, but the optimal approach is unknown.

MATERIALS/METHODS

Multi-institutional cohort study of 385 patients with previously untreated p16+ OPSCC undergoing definitive radiotherapy (RT) with or without systemic therapy between 2009-2020. Chemotherapy intensity was categorized as high (bolus cisplatin and/or induction chemotherapy), medium (weekly cisplatin), or low (non-cisplatin or RT alone). Standard favorable vs. unfavorable risk was defined using NRG HN005 eligibility criteria. High vs. low relative risk (RR) group was defined using the HNCIG omega score (≥ 0.80 vs. < 0.80), which quantifies the proportion of a patient's overall event risk due to cancer. We used multivariable ordinal logistic regression to estimate effects of age (yrs), sex, performance status (PS), Charlson comorbidity index (CCI), T/N (AJCC 8th), current smoking, and pack-years (> 10 vs. ≤ 10) on treatment allocation. Effects on relative event hazards were estimated using generalized competing event regression.

RESULTS

Median follow-up time was 44.2 months. Chemotherapy intensity was high in 206 (54%), medium in 108 (28%), and low in 71 (18%). 280 patients (73%) were unfavorable risk and 197 (51%) were high RR. 178 patients (46%) had discordant risk classification. On univariable analysis, significant predictors of higher intensity chemotherapy (normalized odds ratio (OR)) were CCI 0-1 (OR 1.49, 95% CI: 1.23-1.79), high omega score (OR 1.46; 1.20-1.77), decreased age (OR 1.43; 1.18-1.74), and PS 0 (OR 1.22; 1.01-1.48). Controlling for CCI, higher omega score was associated with significantly higher odds of intensive chemotherapy (OR 1.35; 1.10-1.65, but unfavorable risk (HN005 ineligibility) was not (OR 1.19; 0.98-1.44). Higher omega score was also associated with significantly higher RR for cancer recurrence (Rec) vs. competing mortality (CM) events (relative HR (rHR) 1.76; 1.12-2.75), but unfavorable risk was not (rHR 1.05; 0.63-1.75). Among patients receiving cisplatin, 50 favorable risk patients (58%) had high RR; all of their event risk was due to cancer recurrence (Table). The 110 unfavorable risk patients (48%) with low omega score had significantly lower RR for cancer events compared to the high omega score group (rHR 0.49; 0.29-0.84).

CONCLUSION

Many patients with favorable risk p16+ OPSCC have high relative risk for cancer events, which correlates with a benefit of intensive treatment. The HNCIG omega score is a strong predictor of allocation to intensive chemotherapy and may help identify candidates for deintensification.

Hepatic retinaldehyde deficiency is involved in diabetes deterioration by enhancing PCK1- and G6PC-mediated gluconeogenesis

Type 2 diabetes (T2D) is often accompanied with an induction of retinaldehyde dehydrogenase 1 (RALDH1 or ALDH1A1) expression and a consequent decrease in hepatic retinaldehyde (Rald) levels. However, the role of hepatic Rald deficiency in T2D progression remains unclear. In this study, we demonstrated that reversing T2D-mediated hepatic Rald deficiency by Rald or citral treatments, or liver-specific Raldh1 silencing substantially lowered fasting glycemia levels, inhibited hepatic glucogenesis, and downregulated phosphoenolpyruvate carboxykinase 1 (PCK1) and glucose-6-phosphatase (G6PC) expression in diabetic db/db mice. Fasting glycemia and Pck1/G6pc mRNA expression levels were strongly negatively correlated with hepatic Rald levels, indicating the involvement of hepatic Rald depletion in T2D deterioration. A similar result that liver-specific Raldh1 silencing improved glucose metabolism was also observed in high-fat diet-fed mice. In primary human hepatocytes and oleic acid-treated HepG2 cells, Rald or Rald + RALDH1 silencing resulted in decreased glucose production and downregulated PCK1/G6PC mRNA and protein expression. Mechanistically, Rald downregulated direct repeat 1-mediated PCK1 and G6PC expression by antagonizing retinoid X receptor α, as confirmed by luciferase reporter assays and molecular docking. These results highlight the link between hepatic Rald deficiency, glucose dyshomeostasis, and the progression of T2D, whilst also suggesting RALDH1 as a potential therapeutic target for T2D.

Pre- and Post-Operative Online Prediction of Outcome in Patients Undergoing Endovascular Coiling after Aneurysmal Subarachnoid Hemorrhage: Visual and Dynamic Nomograms

BACKGROUND

Aneurysmal subarachnoid hemorrhage (aSAH) causes long-term functional dependence and death. Early prediction of functional outcomes in aSAH patients with appropriate intervention strategies could lower the risk of poor prognosis. Therefore, we aimed to develop pre- and post-operative dynamic visualization nomograms to predict the 1-year functional outcomes of aSAH patients undergoing coil embolization.

METHODS

Data were obtained from 400 aSAH patients undergoing endovascular coiling admitted to the People's Hospital of Hunan Province in China (2015-2019). The key indicator was the modified Rankin Score (mRS), with 3-6 representing poor functional outcomes. Multivariate logistic regression (MLR)-based visual nomograms were developed to analyze baseline characteristics and post-operative complications. The evaluation of nomogram performance included discrimination (measured by C statistic), calibration (measured by the Hosmer-Lemeshow test and calibration curves), and clinical usefulness (measured by decision curve analysis).

RESULTS

Fifty-nine aSAH patients (14.8%) had poor outcomes. Both nomograms showed good discrimination, and the post-operative nomogram demonstrated superior discrimination to the pre-operative nomogram with a C statistic of 0.895 (95% CI: 0.844-0.945) vs. 0.801 (95% CI: 0.733-0.870). Each was well calibrated with a Hosmer-Lemeshow p-value of 0.498 vs. 0.276. Moreover, decision curve analysis showed that both nomograms were clinically useful, and the post-operative nomogram generated more net benefit than the pre-operative nomogram. Web-based online calculators have been developed to greatly improve the efficiency of clinical applications.

CONCLUSIONS

Pre- and post-operative dynamic nomograms could support pre-operative treatment decisions and post-operative management in aSAH patients, respectively. Moreover, this study indicates that integrating post-operative variables into the nomogram enhanced prediction accuracy for the poor outcome of aSAH patients.

Telomere maintenance genes-derived prognosis signature characterizes immune landscape and predicts prognosis of head and neck squamous cell carcinoma

Telomere dysfunction has been identified as a biological marker of cancer progression in several types of cancer, including Head and Neck Squamous Cell Carcinoma (HNSCC). This study aimed to characterize the telomere maintenance genes (TMG)-related signature in prognosis and treatment response in HNSCC. The transcriptome and clinical data of HNSCC were obtained from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus databases, respectively. Non-negative matrix factorization (NMF) was used to identify molecular subtypes derived from TMG. Gene set enrichment analysis (GSEA) was performed to analyze the differentially expressed pathways between subtypes, and a risk score model derived from TMG was established. Kaplan-Meier survival analysis was used to evaluate inter-group prognostic features, and the correlation between TMG-derived molecular subtypes and risk score model with immune infiltration, immunotherapy, and chemosensitivity was assessed. Two HNSCC subtypes were identified based on 59 TMG-related genes, which exhibit significant heterogeneity in prognosis, immune cell infiltration, and treatment response. Additionally, a TMG-derived risk signature containing 9 genes was developed to assess the prognosis of HNSCC patients. The signature had significant predictive ability for HNSCC prognosis and was significantly correlated with immune cell infiltration and immunotherapy response. A nomogram integrating the risk signature, N stage and radiotherapy was constructed to predict 1-, 3-, and 5-year overall survival (OS) of HNSCC patients, which had better performance than other prognostic models and included TMG-derived risk score, radiotherapy, and N stage. This study identified TMG-derived molecular subtypes in HNSCC and developed a novel prognostic score model, highlighting the potential value of TMG in HNSCC prognosis and immunotherapy.

An Adhesion-based Method for Rapid and Low-cost Isolation of Circulating Tumor Cells

BACKGROUND

Noninvasive monitoring of cancer through circulating tumor cells (CTCs) is hampered long by unsatisfactory CTCs testing techniques. Efficient isolation of CTCs in a rapid and price-favorable way from billions of leukocytes is crucial for testing.

METHODS

We developed a new method based on the stronger adhesive power of CTCs versus leukocytes to sensitively isolate CTCs. Using a BSA-coated microplate and low-speed centrifuge, this method could easily separate cancer cells within 20 min at a very low cost.

RESULT

The capture ratio can reach 70.7∼86.6% in various cancer cell lines (breast/lung/liver/cervical/colorectal cancer) covering different EMT phenotypes and cell sizes, demonstrating the potential for efficient pan-cancer CTCs detection. Moreover, the label-free process can well preserve cell viability (∼99%) to fit downstream DNA/RNA sequencing.

CONCLUSIONS

A novel technique for non-destructive and rapid enrichment of CTCs has been devised. It has enabled the successful isolation of rare tumor cells in the patient blood sample and pleural effusion, highlighting a promising future of this method in clinical translation.

Assessing hypertension and diabetes knowledge, attitudes and practices among residents in Akatsi South District, Ghana using the KAP questionnaire

Objective

Low awareness of hypertension and diabetes is a public health concern in Ghana. Assessing the general population's behaviour via knowledge, attitude, and practice (KAP) will be invaluable in these diseases, where prevention and control need a lifelong commitment to a healthy lifestyle. Hence, our goal was to assess the behaviour of Akatsi South residents towards the diseases to assist health providers in implementing tailored intervention programs.

Methods

This was a population-based cross-sectional study with 150 adults (18-70 years) from November to December 2021. A semi-structured questionnaire with face-to-face interviews was used to obtain data. All variables in the model had descriptive statistics. The Chi-square (χ²) test was used to examine correlations between variables, and a value of p < 0.05 was considered statistically significant. The factors associated with checking blood sugar levels and blood pressure were determined using binary logistic regression.

Results

The respondents' mean age and BMI were 32.40 years (± 12.07) and 24.98 kg/m² (± 2.36), respectively. Only 46.67% of the respondents frequently monitor their blood pressure and 17.33% their blood glucose (at least once a year). Less than half of those surveyed had a good knowledge of hypertension (42.7%) and diabetes (32.0%), whereas nearly 3/4 had poor attitudes regarding both conditions. A binary logistic regression analysis revealed that having a good attitude toward hypertension (exp B = 2.479, p = 0.036) and diabetes (exp B = 4.547, p = 0.009) were the participants' strongest predictor of blood pressure and sugar level checks. However, being overweight (exp B = 0.046, p = 0.002,) or obese (exp B = 0.144, p = 0.034) negatively influenced the frequency with which our respondents checked their blood glucose levels.

Conclusion

In the study, we found that the population generally has poor knowledge, which affects their behaviour (attitudes and practices) towards the diseases. To enable healthcare practitioners to reduce disease-associated mortality and morbidity in the future, frequent public health education and promotion about the conditions is critical to closing the knowledge gap.

Novel dual-emission sulfur quantum dot sensing platform for quantitative monitoring of pesticide 2,4-dichlorophenoxyacetic acid

In this work, a novel environment-friendly dual-emission Rhodamine B modified sulfur quantum dots (RhB-SQDs) sensing platform was established to economically monitor organochlorine pesticide 2,4-dichlorophenoxyacetic acid (2,4-D) through regulating the activity of alkaline phosphatase (ALP). This dual emission RhB-SQDs exhibited excellent fluorescence and high photostability with emission wavelengths of 455 nm and 580 nm. ALP catalyzed the hydrolysis of the substrate p-nitrophenyl phosphate to p-nitrophenol, which quenched RhB-SQDs fluorescence at 455 nm due to the internal filtration effect, but had no effect the fluorescence intensity of RhB-SQDs at 580 nm. When 2,4-D was present, the activity of ALP was specifically inhibited and enzymatic reaction was interrupted, leading to the reduction of p-nitrophenol production, so the fluorescence of RhB-SQDs at 455 nm was restored. It demonstrated a good linear relationship between the concentration of 2,4-D and F₄₅₅/F₅₈₀ in the range of 0.050-0.500 μg mL^-1, with a detection limit of 17.3 ng mL^-1. The dual-emission fluorescent probe was successfully realized in the identification of 2,4-D in natural water samples and vegetables with the advantages of exceptional accuracy, immunity to interference, and selectivity. The platform offers a fresh look at pesticide monitoring and has the potential to prevent pesticide-related health issues.

An optimization for postpartum depression risk assessment and preventive intervention strategy based machine learning approaches

BACKGROUND

Postpartum depression (PPD) is one of the most common psychiatric disorders for women after delivery. The establishment of an effective PPD prediction model helps to distinguish high-risk groups, and verifying whether such high-risk groups can benefit from drug intervention is very important for clinical guidance.

METHODS

We collected data of parturients that underwent a cesarean delivery. The Control group was divided into a training cohort and a testing cohort. Six different ML models were constructed and we compared their prediction performance in the testing cohort. For model interpretation, we introduced SHapley Additive exPlanations (SHAP). Then, training cohort, ketamine group and dexmedetomidine (DEX) group were classified as high or low risk for PPD by the model. A 1:1 propensity score matching (PSM) was performed to compare the incidence of PPD between two groups in different risk cohorts.

RESULTS

Extreme gradient enhancement (XGB) had the best recognition effect, with an area under the receiver operating characteristic curve (AUROC) of 0.789 (95 % CI 0.742-0.836) in the training cohort and 0.744 (95 % CI 0.655-0.823) in the testing cohort, respectively. A threshold of 21.5 % PPD risk probability was determined. After PSM, the results showed that the incidence of PPD in the two intervention groups was significantly different from the control group in the high-risk cohort (P < 0.001) but not in the low-risk cohort (P > 0.001).

CONCLUSION

Our study demonstrated that the XGB algorithm provided a more accurate in prediction of PPD risk, and it was beneficial to receive early intervention for the high-risk groups distinguished by the model.

Pyrotinib alone or in combination with docetaxel in refractory HER2-positive gastric cancer: A dose-escalation phase I study

AIM

Pyrotinib (an irreversible pan-ErbB small-molecular tyrosine kinase inhibitor) was approved in human epidermal growth factor receptor 2 (HER2)-positive breast cancer and showed great antitumor activity in preclinical studies of gastric cancer (GC). This study was first designed to prospectively assess pyrotinib in pretreated HER2-positive GC.

METHODS

This multicenter, phase I study followed a standard "3 + 3" design and included two parts. In the pyrotinib part, pyrotinib was administered orally, once per day at dose levels of 240, 320, 400, and 480 mg. In the pyrotinib plus docetaxel part, patients received pyrotinib (qd, d1-21, q3W) combined with docetaxel (60 mg/m² , d1, q3W) at dose levels of 240, 320, and 400 mg. Primary endpoints were to determine the maximum tolerated dose (MTD) and recommended phase II dose (RP2D) of pyrotinib as monotherapy or coadministered with docetaxel.

RESULTS

A total of 25 patients were enrolled and received pyrotinib (n = 15) or pyrotinib plus docetaxel (n = 10). One DLT was observed in pyrotinib monotherapy part (Grade 3 uncontrolled diarrhea after supportive care) and pyrotinib plus docetaxel part (Grade 4 neutropenia and leukopenia). In the pyrotinib monotherapy part, MTD was not reached. Diarrhea, anemia, neutropenia, and leukopenia were the most common treatment-related adverse events (TRAEs). The RP2D for pyrotinib monotherapy was recommended as 400 mg. After combining with docetaxel, the risk of leukopenia and neutropenia was increased. Grade ≥3 TRAEs were reported for four patients in the monotherapy part and for eight patients in the combination part. Mean t_1/2 was approximately 20 h. Pyrotinib exposure was dose-dependent with a nonlinear relationship versus dose. There were five patients who had confirmed partial response (monotherapy: one each at 240, 400, and 480 mg dose cohort; combination therapy: two at 240 mg dose cohort), resulting in an objective response rate of 21% and 20%, respectively.

CONCLUSIONS

Pyrotinib alone and combined with docetaxel showed acceptable toxicities in patients with pretreated HER2-positive GC.

TRIAL REGISTRATION

This study was registered with ClinicalTrials.gov, NCT02378389.

Interpretable Machine Learning Model Predicting Early Neurological Deterioration in Ischemic Stroke Patients Treated with Mechanical Thrombectomy: A Retrospective Study

Early neurologic deterioration (END) is a common and feared complication for acute ischemic stroke (AIS) patients treated with mechanical thrombectomy (MT). This study aimed to develop an interpretable machine learning (ML) model for individualized prediction to predict END in AIS patients treated with MT. The retrospective cohort of AIS patients who underwent MT was from two hospitals. ML methods applied include logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost). The area under the receiver operating characteristic curve (AUC) was the main evaluation metric used. We also used Shapley Additive Explanation (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME) to interpret the result of the prediction model. A total of 985 patients were enrolled in this study, and the development of END was noted in 157 patients (15.9%). Among the used models, XGBoost had the highest prediction power (AUC = 0.826, 95% CI 0.781–0.871). The Delong test and calibration curve indicated that XGBoost significantly surpassed those of the other models in prediction. In addition, the AUC in the validating set was 0.846, which showed a good performance of the XGBoost. The SHAP method revealed that blood glucose was the most important predictor variable. The constructed interpretable ML model can be used to predict the risk probability of END after MT in AIS patients. It may help clinical decision making in the perioperative period of AIS patients treated with MT.

A Novel Phenazine Analog, CPUL1, Suppresses Autophagic Flux and Proliferation in Hepatocellular Carcinoma: Insight from Integrated Transcriptomic and Metabolomic Analysis

BACKGROUND

CPUL1, a phenazine analog, has demonstrated potent antitumor properties against hepatocellular carcinoma (HCC) and indicates a promising prospect in pharmaceutical development. However, the underlying mechanisms remain largely obscure.

METHODS

Multiple HCC cell lines were used to investigate the in vitro effects of CPUL1. The antineoplastic properties of CPUL1 were assessed in vivo by establishing a xenograft nude mice model. After that, metabolomics, transcriptomics, and bioinformatics were integrated to elucidate the mechanisms underlying the therapeutic efficacy of CPUL1, highlighting an unanticipated involvement of autophagy dysregulation.

RESULTS

CPUL1 suppressed HCC cell proliferation in vitro and in vivo, thereby endorsing the potential as a leading agent for HCC therapy. Integrative omics characterized a deteriorating scenario of metabolic debilitation with CPUL1, presenting an issue in the autophagy contribution of autophagy. Subsequent observations indicated that CPUL1 treatment could impede autophagic flow by suppressing autophagosome degradation rather than its formation, which supposedly exacerbated cellular damage triggered by metabolic impairment. Moreover, the observed late autophagosome degradation may be attributed to lysosome dysfunction, which is essential for the final stage of autophagy and cargo disposal.

CONCLUSIONS

Our study comprehensively profiled the anti-hepatoma characteristics and molecular mechanisms of CPUL1, highlighting the implications of progressive metabolic failure. This could partially be ascribed to autophagy blockage, which supposedly conveyed nutritional deprivation and intensified cellular vulnerability to stress.

Programmable CRISPR-Cas12a and self-recruiting crRNA assisted dual biosensing platform for simultaneous detection of lung cancer biomarkers hOGG1 and FEN1

Human 8-oxoguanine DNA glycosylase (hOGG1) and flap endonuclease 1 (FEN1) are recognized as potential biomarkers in lung cancer investigations. Developing analytical platforms of simultaneously targeting hOGG1 and FEN1 with high selectivity, sensitivity, especially programmability and universality is highly valuable for clinical research. Herein, we established a signal-amplified platform for simultaneously detecting hOGG1 and FEN1 on the basis of cleavage-induced ligation of DNA dumbbell probes, rolling circle transcription (RCT) and CRISPR-Cas12a. A hOGG1 cleavable site and FEN1 cleavable flap were dexterously designed at the 5' end of DNA flapped dumbbell probes (FDP) for hOGG1 and FEN1. After cleavage, the resulting nick sites with juxtaposition of 5' phosphate and 3' hydroxyl terminus could be linked to closed DNA dumbbell probes (CDP) by DNA ligase. The CDP served as a template for RCT, producing plentiful crRNA repeats to activate the trans-cleavage activity of CRISPR-Cas12a which could cleave fluorophores (TAMRA and FAM) and quenchers (BHQ2 and BHQ1) double-labeled ssDNA reporters. Then, hOGG1 and FEN1 could be detected by the recovered fluorescence signal, allowing for the highly sensitive calculated detection limits of 0.0013 and 0.0052 U/mL, respectively. Additionally, this method made it possible to evaluate the inhibitory effects, even to measure hOGG1 and FEN1 activities at the single-cell level. This novel target enzyme-initiated, circles-transcription without promoters, real-time generation, and self-assembly features of FDP-RCT-Cas12a system suppressed nonspecific background remarkably and relieved rigorous requirement of protospacer adjacent motif site. Hence, the universality of FDP-RCT-Cas12a system toward various disease-related non-nucleic acid targets which are tested without using aptamers was extremely improved.

In vitro co-culture systems of hepatic and intestinal cells for cellular pharmacokinetic and pharmacodynamic studies of capecitabine against colorectal cancer

BACKGROUND

As a prodrug of 5-fluorouracil (5-FU), orally administrated capecitabine (CAP) undergoes preliminary conversion into active metabolites in the liver and then releases 5-FU in the gut to exert the anti-tumor activity. Since metabolic changes of CAP play a key role in its activation, a single kind of intestinal or hepatic cell can never be used in vitro to evaluate the pharmacokinetics (PK) and pharmacodynamics (PD) nature. Hence, we aimed to establish a novel in vitro system to effectively assess the PK and PD of these kinds of prodrugs.

METHODS

Co-culture cellular models were established by simultaneously using colorectal cancer (CRC) and hepatocarcinoma cell lines in one system. Cell Counting Kit-8 (CCK-8) and flow cytometric analysis were used to evaluate cell viability and apoptosis, respectively. Apoptosis-related protein expression levels were measured using western blot analysis. A selective liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was developed for cellular PK in co-culture models.

RESULTS

CAP had little anti-proliferative effect on the five monolayer CRC cell lines (SW480, LoVo, HCT-8, HCT-116 and SW620) or the hepatocarcinoma cell line (HepG2). However, CAP exerted marked anti-tumor activities on each of the CRC cell lines in the co-culture models containing both CRC and hepatocarcinoma cell lines, although its effect on the five CRC cell lines varied. Moreover, after pre-incubation of CAP with HepG2 cells, the culture media containing the active metabolites of CAP also showed an anti-tumor effect on the five CRC cell lines, indicating the crucial role of hepatic cells in the activation of CAP.

CONCLUSION

The simple and cost‑effective co-culture models with both CRC and hepatocarcinoma cells could mimic the in vivo process of a prodrug dependent on metabolic conversion to active metabolites in the liver, providing a valuable strategy for evaluating the PK and PD characteristics of CAP-like prodrugs in vitro at the early stage of drug development.

An interpretable artificial neural network model for predicting hypoxemia via an online tool in adult (18-64) patients during esophagogastroduodenoscopy

Background

The hypoxemia risk in adult (18-64) patients treated with esophagogastroduodenoscopy (EGD) under sedation often poses a dilemma for anesthesiologists. We aimed to establish an artificial neural network (ANN) model to solve this problem, and introduce the Shapley additive explanations (SHAP) algorithm to further improve the interpretability.

Methods

The relevant data of patients underwent routine anesthesia-assisted EGD were collected. Elastic network was used to filter the optimal features. Airway-ANN and Basic-ANN models were established based on all collected indicators and remaining variables excluding airway assessment indicators, respectively. The performance of Basic-ANN, Airway-ANN and STOP-BANG was evaluated by the area under the precision-recall curve (AUPRC) on temporal validation set. The SHAP was used for revealing the predictive behavior of our best model.

Results

999 patients were eventually included. The AUPRC value of Airway-ANN model was significantly higher than Basic-ANN model in the temporal validation set (0.532 vs 0.429, P < 0.05). And the performance of both two ANN models was significantly better than that of STOP-BANG score (both P < 0.05). The Airway-ANN model was deployed to the cloud (http://njfh-yxb.com.cn:2022/airway_ann).

Conclusion

Our online interpretable Airway-ANN model achieved satisfying ability in identifying the hypoxemia risk in adult (18-64) patients undergoing EGD.

Pulmonary blastoma treatment response to anti-PD-1 therapy: a rare case report and literature review

Pulmonary blastoma (PB) is a rare and invasive malignancy of the lungs with a poor prognosis. Although the mainstay treatment of PB is surgery, and radiotherapy and chemotherapy have been reported, no standard therapy exists for patients inoperable in advanced stages. Moreover, little is known about driver mutation status and immunotherapy efficacy. This paper presents a male patient diagnosed with classic biphasic PB using CT-guided lung biopsy pathology and immunohistochemistry. The patient's symptoms included cough, chest pain, shortness of breath, hemoptysis, and hypodynamia. The primary focus of this paper is to discuss the impact of anti-PD-1 immunotherapy on PB. The patient experienced progression-free survival (PFS) of over 27 months following sintilimab second-line anti-PD-1 therapy. The patient has currently survived for nearly 40 months with a satisfactory quality of life.

Neoadjuvant pyrotinib, trastuzumab, and docetaxel for HER2-positive breast cancer (PHEDRA): a double-blind, randomized phase 3 trial

BACKGROUND

Pyrotinib (an irreversible pan-ErbB inhibitor) plus capecitabine has survival benefits and acceptable tolerability in patients with HER2-positive metastatic breast cancer. We further assessed addition of pyrotinib to trastuzumab and docetaxel in the neoadjuvant setting.

METHODS

In this multicenter, double-blind, phase 3 study (PHEDRA), treatment-naive women with HER2-positive early or locally advanced breast cancer were randomly assigned (1:1) to receive four neoadjuvant cycles of oral pyrotinib or placebo (400 mg) once daily, plus intravenous trastuzumab (8 mg/kg loading dose, followed by 6 mg/kg) and docetaxel (100 mg/m²) every 3 weeks. The primary endpoint was the total pathological complete response (tpCR; ypT0/is and ypN0) rate per independent central review.

RESULTS

Between Jul 23, 2018, and Jan 8, 2021, 355 patients were randomly assigned, 178 to the pyrotinib group and 177 to the placebo group. The majority of patients completed four cycles of neoadjuvant treatment as planned (92.7% and 97.7% in the pyrotinib and placebo groups, respectively). The tpCR rate was 41.0% (95% CI 34.0 to 48.4) in the pyrotinib group compared with 22.0% (95% CI 16.6 to 28.7) in the placebo group (difference, 19.0% [95% CI 9.5 to 28.4]; one-sided P < 0.0001). The objective response rate per investigator was 91.6% (95% CI 86.6 to 94.8) in the pyrotinib group and 81.9% (95% CI 75.6 to 86.9) in the placebo group after the neoadjuvant treatment, resulting in an increase of 9.7% (95% CI 2.7 to 16.6). The most common grade 3 or worse adverse events were diarrhea (79 [44.4%] in the pyrotinib group and nine [5.1%] in the placebo group), neutropenia (33 [18.5%] and 36 [20.3%]), and decreased white blood cell count (29 [16.3%] and 24 [13.6%]). No deaths were reported during neoadjuvant treatment.

CONCLUSIONS

The primary endpoint of the study was met. Neoadjuvant pyrotinib, trastuzumab, and docetaxel significantly improved the tpCR rate compared with placebo, trastuzumab, and docetaxel, with manageable toxicity, providing a new option for HER2-positive early or locally advanced breast cancer.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT03588091.

SHR-1701, a Bifunctional Fusion Protein Targeting PD-L1 and TGFβ, for Recurrent or Metastatic Cervical Cancer: A Clinical Expansion Cohort of a Phase I Study

PURPOSE

Patients with recurrent or metastatic cervical cancer have limited treatment options after platinum-containing treatment. We initiated a phase I study to assess SHR-1701, a novel bifunctional fusion protein composed of a mAb against programmed death ligand 1 (PD-L1) fused with the extracellular domain of TGFβ receptor II, in solid tumors (NCT03774979). Here, results from the cervical cancer cohort are presented.

PATIENTS AND METHODS

Patients with recurrent or metastatic cervical cancer who progressed during or after platinum-based therapy were enrolled to receive SHR-1701 at 30 mg/kg every 3 weeks. Primary endpoint was objective response rate (ORR) per RECIST v1.1.

RESULTS

In total, 32 patients were recruited. ORR was 15.6% [95% confidence interval (CI), 5.3-32.8], and disease control rate was 50.0% (95% CI, 31.9-68.1). Responses were still ongoing in 80.0% of the responders; 6-month duration of response rate was 80.0% (95% CI, 20.4-96.9). Median progression-free survival (PFS) was 2.7 months (95% CI, 1.4-4.1). Of note, as assessed by immune-modified RECIST, median PFS was 4.1 months (95% CI, 1.6-4.3). Overall survival rate at 12 months was 54.6% (95% CI, 31.8-72.7). Treatment-related adverse events of grade 3 or 4 were reported in 11 (34.4%) patients. No treatment-related deaths occurred. No difference in ORR was found between patients with PD-L1 combined positive score ≥1 or <1; patients with high phosphorylated SMAD2 level in immune cells or tumor cells had numerically higher ORR.

CONCLUSIONS

SHR-1701 exhibits encouraging antitumor activity and controllable safety in patients with recurrent or metastatic cervical cancer after platinum-based regimens, and therefore might provide another treatment option for this population. See related commentary by Miller and Friedman, p. 5238.

Influence of tirofiban on stroke outcome after mechanical thrombectomy in acute vertebrobasilar artery occlusion

BACKGROUND

Even undergoing mechanical thrombectomy (MT), patients with acute vertebrobasilar artery occlusion (AVBAO) still have a high rate of mortality. Tirofiban is a novel antiplatelet agent which is now widely empirically used in acute ischemic stroke (AIS). In this study, we aimed to evaluate the safety and efficacy of tirofiban as adjunctive therapy for MT in AVBAO.

METHODS

From October 2016 to July 2021, consecutive AVBAO patients receiving MT were included in the prospective stroke registry. The short-term outcomes were (1) symptomatic intracerebral hemorrhage (sICH); (2) in-hospital death; (3) National Institute of Health Stroke Scale (NIHSS) at discharge. The Long-term outcomes were: (1) modified Rankin Scale (mRS) at 3 months; (2) death at 3 months.

RESULTS

A total of 130 eligible patients were included in the study, 64 (49.2%) patients received tirofiban. In multivariate regression analysis, no significant differences were observed in all outcomes between the tirofiban and non-tirofiban group [sICH (adjusted OR 0.96; 95% CI, 0.12-7.82, p = 0.97), in-hospital death (adjusted OR 0.57; 95% CI, 0.17-1.89, p = 0.36), NIHSS at discharge (95% CI, -2.14-8.63, p = 0.24), mRS (adjusted OR 1.20; 95% CI, 0.40-3.62, p = 0.75), and death at 3 months (adjusted OR 0.83; 95% CI, 0.24-2.90, p = 0.77)].

CONCLUSIONS

In AVBAO, tirofiban adjunctive to MT was not associated with an increased risk of sICH. Short-term (in-hospital death, NIHSS at discharge) and long-term outcomes (mRS and death at 3 months) seem not to be influenced by tirofiban use.

Multiplexed imaging of tumor immune microenvironmental markers in locally advanced or metastatic non-small-cell lung cancer characterizes the features of response to PD-1 blockade plus chemotherapy

BACKGROUND

Although programmed cell death 1 (PD-1) blockade plus chemotherapy can significantly prolong the progression-free survival (PFS) and overall survival (OS) in first-line settings in patients with driver-negative advanced non-small-cell lung cancer (NSCLC), the predictive biomarkers remain undetermined. Here, we investigated the predictive value of tumor immune microenvironmental marker expression to characterize the response features to PD-1 blockade plus chemotherapy.

METHODS

Tumor tissue samples at baseline were prospectively collected from 144 locally advanced or metastatic NSCLC patients without driver gene alterations who received camrelizumab plus chemotherapy or chemotherapy alone. Tumor immune microenvironmental markers, including PD-1 ligand (PD-L1), CD8, CD68, CD4 and forkhead box P3, were assessed using multiplex immunofluorescence (mIF) assays. Kaplan-Meier curves were used to determine treatment outcome differences according to their expression status. Mutational profiles were compared between tumors with distinct expression levels of these markers and their combinations.

RESULTS

Responders had significantly higher CD8/PD-L1 (P = 0.015) or CD68/PD-L1 co-expression levels (P = 0.021) than non-responders in the camrelizumab plus chemotherapy group, while no difference was observed in the chemotherapy group. Patients with high CD8/PD-L1 or CD68/PD-L1 co-expression level was associated with significantly longer PFS (P = 0.002, P = 0.024; respectively) and OS (P = 0.006, P = 0.026; respectively) than those with low co-expression in camrelizumab plus chemotherapy group. When comparing survival in the camrelizumab plus chemotherapy with chemotherapy by CD8/PD-L1 co-expression stratification, significantly better PFS (P = 0.003) and OS (P = 0.032) were observed in high co-expression subgroups. The predictive value of CD8/PD-L1 and CD68/PD-L1 co-expression remained statistically significant for PFS and OS when adjusting clinicopathological features. Although the prevalence of TP53 or KRAS mutations was similar between patients with and without CD8/PD-L1 or CD68/PD-L1 co-expression, the positive groups had a significantly higher proportion of TP53/KRAS co-mutations than the negative groups (both 13.0% vs. 0.0%, P = 0.023). Notably, enriched PI3K (P = 0.012) and cell cycle pathway (P = 0.021) were found in the CD8/PD-L1 co-expression group.

CONCLUSION

Tumor immune microenvironmental marker expression, especially CD8/PD-L1 or CD68/PD-L1 co-expression, was associated with the efficacy of PD-1 blockade plus chemotherapy as first-line treatment in patients with advanced NSCLC.

Prediction on the medial knee contact force in patients with knee valgus using transfer learning approaches: Application to rehabilitation gaits

The Knee contact force (KCF) is a key factor in evaluating knee joint function of patients with knee osteoarthritis. In vivo measurement of KCF based on the instrumented implants is limited due to the ethical issues and technical complexities. Machine learning can be used to predict tibiofemoral compartment contact forces. However, anthropometric differences between individuals make the accurate predictions challenging. The purpose of this study was to develop transfer learning models to predict the medial KCF of patients with knee valgus in rehabilitation gaits. Four subjects with instrumented tibial prostheses were considered, including one with knee valgus and three with normal knee joint alignment. Two transfer learning models were proposed: a fine-tuning model and an adaptive model. In particular, a synchronization method for extracting experimental data in a complete gait cycle was developed, since different types of experimental data have different sampling frequencies. The transfer learning models were pre-trained by the experiment data of patients with normal knee joint alignment, and re-trained by the data of the patient with knee valgus. Predictions of the transfer learning models and traditional machine learning model were validated against the in vivo measurements. The proposed transfer learning models were tested within two levels: the single subject (Level 1) and multiple subjects (Level 2). The results show that the two transfer learning models could more accurately predict the medial KCF of patients with knee valgus than the traditional machine learning model. The performance of the fine-tuning model is better than that of the adaptive model. Compared with the traditional machine learning and inverse dynamics analysis, transfer learning represents a much easier and more accurate method. It can be introduced to help clinicians validate and adjust the rehabilitation gait for specific patients.

Novel Bedside Dynamic Nomograms to Predict the Probability of Postoperative Cognitive Dysfunction in Elderly Patients Undergoing Noncardiac Surgery: A Retrospective Study

Purpose

Early and accurate prediction of elderly patients at high risk of postoperative cognitive dysfunction (POCD) after non-cardiac surgery will provide favorable evidence for rational perioperative management and long-term postoperative recovery. This study aimed to develop bedside dynamic nomograms to provide accurately an individualized prediction of the risk of POCD at 6-month postoperatively with patients undergoing non-cardiac surgery and to guide clinical decision-making and postoperative management.

Patients and Methods

We retrospectively collected patients undergoing surgical treatment at the Nanjing First Hospital between May 2020 and May 2021. We collected the data on preoperative, intraoperative, and postoperative variables. Clinical and laboratory data on admission and intraoperative variables and postoperative variables were used. We measured the performances of the nomograms using sensitivity, specificity of the receiver operating characteristic (ROC), the area under the ROC curves (AUC), the 10-fold cross-validation, and decision curve analysis (DCA).

Results

POCD was observed in 23 of 415 patients (5.6%) at 6-month postoperatively. The preoperative and postoperative models obtained 91.6% and 94.0% accuracy rates on the data. Compared to the preoperative model, the postoperative model had an area under the receiver characteristic curve (AUC) of 0.973 vs 0.947, corresponding to a specificity of 0.941 vs 0.918 and a sensitivity of 0.913 vs 0.870. The overall performance of the postoperative model was better than the preoperative model.

Conclusion

In this study, we developed novel bedside dynamic nomograms with reasonable clinical utility that can provide individualized prediction of POCD risk at 6-month postoperatively in elderly patients undergoing non-cardiac surgery at different time points based on patient admission and postoperative data. External validations are needed to ensure their value in predicting POCD in elderly patients.

Machine learning to predict futile recanalization of large vessel occlusion before and after endovascular thrombectomy

Background and purpose

Futile recanalization occurs when the endovascular thrombectomy (EVT) is a technical success but fails to achieve a favorable outcome. This study aimed to use machine learning (ML) algorithms to develop a pre-EVT model and a post-EVT model to predict the risk of futile recanalization and to provide meaningful insights to assess the prognostic factors associated with futile recanalization.

Methods

Consecutive acute ischemic stroke patients with large vessel occlusion (LVO) undergoing EVT at the National Advanced Stroke Center of Nanjing First Hospital (China) between April 2017 and May 2021 were analyzed. The baseline characteristics and peri-interventional characteristics were assessed using four ML algorithms. The predictive performance was evaluated by the area under curve (AUC) of receiver operating characteristic and calibration curve. In addition, the SHapley Additive exPlanations (SHAP) approach and partial dependence plot were introduced to understand the relative importance and the influence of a single feature.

Results

A total of 312 patients were included in this study. Of the four ML models that include baseline characteristics, the “Early” XGBoost had a better performance {AUC, 0.790 [95% confidence intervals (CI), 0.677–0.903]; Brier, 0.191}. Subsequent inclusion of peri-interventional characteristics into the “Early” XGBoost showed that the “Late” XGBoost performed better [AUC, 0.910 (95% CI, 0.837–0.984); Brier, 0.123]. NIHSS after 24 h, age, groin to recanalization, and the number of passages were the critical prognostic factors associated with futile recanalization, and the SHAP approach shows that NIHSS after 24 h ranks first in relative importance.

Conclusions

The “Early” XGBoost and the “Late” XGBoost allowed us to predict futile recanalization before and after EVT accurately. Our study suggests that including peri-interventional characteristics may lead to superior predictive performance compared to a model based on baseline characteristics only. In addition, NIHSS after 24 h was the most important prognostic factor for futile recanalization.

Transcriptome dissection of candidate genes associated with lentil seed quality traits

Lentils provide a rich plant-based protein source and staple food in many parts of the world. Despite numerous nutritional benefits, lentil seeds also possess undesirable elements, such as anti-nutritional factors. Understanding the genetic networks of seed metabolism is of great importance for improving the seed nutritional profile. We applied RNA sequencing analysis to survey the transcriptome of developing lentil seeds and compared this with that of the pod shells and leaves. In total, we identified 2622 genes differentially expressed among the tissues examined. Genes preferentially expressed in seeds were enriched in the Gene Ontology (GO) terms associated with development, nitrogen and carbon (N/C) metabolism and lipid synthesis. We further categorized seed preferentially expressed genes based on their involvement in storage protein production, starch accumulation, lipid and suberin metabolism, phytate, saponin and phenylpropanoid biosynthesis. The availability of transcript profile datasets on lentil seed metabolism and a roadmap of candidate genes presented here will be of great value for breeding strategies towards further improvement of lentil seed quality traits.

SHR2554, an EZH2 inhibitor, in relapsed or refractory mature lymphoid neoplasms: a first-in-human, dose-escalation, dose-expansion, and clinical expansion phase 1 trial

BACKGROUND

Dysregulation of EZH2 has a crucial role in lymphomagenesis. We did a first-in-human study to assess the safety, pharmacokinetics, pharmacodynamics, and preliminary clinical activity of SHR2554, an oral EZH2 inhibitor, in patients with relapsed or refractory mature lymphoid neoplasms, including B-cell lymphomas, T-cell lymphomas, and classical Hodgkin lymphoma.

METHODS

This was a multicentre, dose-escalation, dose-expansion, and clinical expansion phase 1 study done at 13 hospitals in China. Eligible patients had histologically or cytologically confirmed mature lymphoid neoplasms that had relapsed or were refractory to standard systemic therapies or had no standard-of-care. The study included a dose-escalation phase, at doses of SHR2554 from 50 mg to 800 mg twice daily; a dose-expansion phase, at two selected doses; and a subsequent clinical expansion phase at the recommended phase 2 dose in selected tumours. Primary endpoints were the safety, maximum tolerated dose, and recommended phase 2 dose. Objective response rate was a secondary endpoint. Safety and activity were assessed in all patients who received at least one dose of SHR2554 and had at least one post-baseline evaluation. This study is registered with ClinicalTrials.gov, NCT03603951, and follow-up is ongoing.

FINDINGS

Between Aug 14, 2018, and July 13, 2021, 113 patients received SHR2554. At data cutoff (Sept 10, 2021), the median follow-up duration was 7·0 months (IQR 3·7-12·0). 71 (63%) patients were men and 42 (37%) were women, 110 (97%) were of Han ethnicity and 3 (3%) of other ethnicities, and 53 (47%) had received three or more lines of previous anticancer therapies. Dose-limiting toxicities occurred in two (67%) of three patients who received 400 mg SHR2554 twice daily and one (17%) of six patients who received 350 mg SHR2554 twice daily. The maximum tolerated dose and recommended phase 2 dose was determined to be 350 mg twice daily. The most common grade 3 or 4 treatment-related adverse events in all 113 patients were decreased platelet count (20 [18%]), decreased neutrophil count (ten [9%]), decreased white blood cell count (nine [8%]), and anaemia (seven [6%]). 18 (16%) patients had serious treatment-related adverse events. Two patients (2%) died due to treatment-related adverse events: one (1%) due to skin infection and toxic epidermal necrolysis and one (1%) due to respiratory failure. 107 (95%) of the 113 enrolled patients had post-baseline assessments for tumour response and were included in the activity analysis. 46 (43%; 95% CI 33-53) of these 107 patients had an overall response.

INTERPRETATION

SHR2554 showed an acceptable safety profile and promising antitumour activity in patients with relapsed or refractory lymphomas, providing evidence for future investigations.

FUNDING

Jiangsu Hengrui Pharmaceuticals.

TRANSLATION

For the Chinese translation of the abstract see Supplementary Materials section.