Simple Clinical Characteristics Research Articles

How well can simple clinical features predict long-term language recovery after left-hemisphere glioma surgery?

Long-term language recovery after left-hemisphere glioma surgery varies substantially across patients. We investigated how well it can be predicted using clinical variables such as the postoperative decline in language processing, tumor grade, resection volume and location, extent of resection, and intraoperative language mapping. Beyond predicting the overall recovery, we examined which domains of language processing are most prone to persistent deficits. Fifty-nine patients with left-hemisphere gliomas completed the Russian Aphasia Test (RAT) before surgery, immediately after surgery, and at follow-up three to seventeen months after surgery. We modeled their average language score (Generalized Aphasia Quotient, GAQ) at follow-up using a cross-validated multiple linear regression and calculated the number of patients showing persistent deficits in each subtest of the RAT. The difference between GAQ scores at follow-up and before surgery was not significant at the group level but varied substantially across patients (mean -1.3%, range -34.2-9.2%). Our best-performing model predicted the follow-up GAQ scores with the mean absolute error of 3.5% (cross-validated R2 = 0.15). A greater decline in language processing immediately after surgery predicted worse recovery, whereas intraoperative language mapping predicted better recovery. Deficits in sentence repetition, verb production, verb and sentence comprehension, and object naming most often persisted at follow-up. The postoperative decline in language processing and intraoperative language mapping explain a substantial amount of variability in long-term language recovery. Verbal working memory and lexical retrieval, particularly that of verbs, are most prone to persistent deficits.

Differential diagnosis of pediatric cervical lymph node lesions based on simple clinical features.

A discriminant analysis model based on simple clinical features can be effective in differential diagnosis of cervical histiocytic necrotizing lymphadenitis and malignant lymphoma in children. Peak body temperature, white blood cell count, and short diameter of the largest mass are significant predictors. • Several multivariate diagnostic models for HNL and ML have been proposed based on B-ultrasound or CT features in adults. • The differences between children and adults are nonnegligible in the clinical featues of HNL. • The study firstly report a large-sample diagnostic model between the HNL and MLin pediatric patients. • Non-imaging clinical features has also been proven with quite good diagnostic value.

Early pregnancy associations with Gestational Diabetes: methods and cohort results of the Hoosier Moms Cohort

Background Gestational Diabetes (GDM) raises the risk of adverse perinatal outcomes and long-term risk of type 2 diabetes. There is currently a lack of comprehensive GDM prediction models based on more than simple clinical features. Objective The objective of this study was to collect a comprehensive set of clinical, sociodemographic, biobehavioral, and genomic features in a prospective high-risk cohort for GDM, to discover novel predictive and therapeutic targets for GDM during early pregnancy. Study design The Hoosier Moms Cohort was a prospective observational study of pregnant individuals, with a singleton gestation <20 weeks. The study protocol included 2 visits during pregnancy and one at delivery. Psychosocial, dietary, social, and demographic characteristics were collected in addition to maternal and newborn samples. Developing GDM was the primary outcome. Univariate associations with GDM for continuous variables were analyzed using either two-sample t-test or Wilcoxon Rank Sum test, and categorical variables using either chi-square or Fishers exact test. Multiple logistic regression was performed for independent associations with GDM. Results A total of 411 participants were recruited, with complete data available for 391. Patients were on average 30 years of age, had a mean body mass index (BMI) of 28, and 17% were of Hispanic ethnicity. Additionally, 54% reported a family history of diabetes, with 4% reporting a personal prior history of GDM. A total of 39 participants (10.0%) developed GDM. Compared to those that did not, participants who developed GDM had a significantly higher baseline BMI (31.6 vs 27.2, p=0.003), HbA1c (5.24 vs 5.07, p<0.001), triglycerides (156.8 vs 134.2, p=0.022), and random blood glucose (85.90 vs 79.96, p=0.025) at the initial visit. Those with GDM were more likely to have a prior history of gestational diabetes (28.21% vs 1.96%, p<0.001), and current chronic hypertension (12.82% vs 1.9%, p=0.003). Additionally, they scored higher on a validated insomnia questionnaire (9.62 vs 7.80, p=0.028). A significant association was found between GDM and 3 previously reported genetic markers (p<0.01). Individuals with high polygenic risk scores for type 2 diabetes were not more likely to have a GDM diagnosis. Through stepwise logistic regression, prior history of GDM, current diagnosis of hypertension, insomnia, and BMI were independently associated with GDM (odds ratio, 95% confidence intervals: 14.98, 4.49-50.02; 10.94, 2.32-51.69; 1.11, 1.01-1.22; 1.09, 1.03-1.16, respectively). Conclusion The Hoosier Moms Cohort identified that participants with a previous GDM diagnosis, chronic hypertension, elevated BMI, and insomnia have significantly increased odds of developing GDM in a diverse cohort of participants. These factors will be integrated into a machine learning model with multi-omics data to develop a comprehensive predictor for GDM.

Prediction of prognosis in patients with systemic sclerosis based on a machine-learning model.

The clinical manifestations of systemic sclerosis (SSc) are highly variable, resulting in varied outcomes and complications. Diverse fibrosis of the skin and internal organs, vasculopathy, and dysregulated immune system lead to poor and varied prognoses in patients with SSc subtypes. Therefore, this study aimed to develop a personalized tool for predicting the prognosis of patients with SSc. A cohort of 517 patients with SSc were recruited between January 2009 and November 2021 at Xijing Hospital in China, and 266 patients completed the follow-up and performed in the survival analysis. Risk factors for death were identified using Cox survival analysis and random survival forest-based machine-learning methods separately. The consistency index, area under the curve (AUC), and integrated Brier scores were used to compare the predictive performance of the different prognostic models. The results of Cox-based multivariate regression analysis suggested that pulmonary arterial hypertension, digital ulcer, and Modified Rodnan Skin Score (mRSS) were independent risk factors for poor prognosis in patients with SSc and significant risk factors in random survival forest (RSF) surveys. A nomogram was plotted to evaluate the prognostic risk to facilitate clinical assessment; the RSF model had better predictive performance than the Cox model, with 3- and 5-year AUCs of 0.74 and 0.78, respectively. Machine-learning models can help us better understand the prognosis of patients with SSc and comprehensively evaluate the clinical characteristics of each individual. The early identification of the characteristics of high-risk patients can improve the prognosis of those with SSc. Key Points • Regarding predictive performance, the random survival forest model was more effective than the Cox model and had unique advantages in analyzing nonlinear effects and variable importance. • Machine learning using the simple clinical features of patients with systemic sclerosis (SSc) to predict mortality can guide attending physicians, and the early identification of high-risk patients with SSc and referral to experts will assist rheumatologists in monitoring and management planning.

Magnetic resonance imaging-based prediction models for differentiating intraspinal schwannomas from meningiomas: classification and regression tree and random forest analysis.

Due to the variations in surgical approaches and prognosis between intraspinal schwannomas and meningiomas, it is crucial to accurately differentiate between the two prior to surgery. Currently, there is limited research exploring the implementation of machine learning (ML) methods for distinguishing between these two types of tumors. This study aimed to establish a classification and regression tree (CART) model and a random forest (RF) model for distinguishing schwannomas from meningiomas. We retrospectively collected 88 schwannomas (52 males and 36 females) and 51 meningiomas (10 males and 41 females) who underwent magnetic resonance imaging (MRI) examinations prior to the surgery. Simple clinical data and MRI imaging features, including age, sex, tumor location and size, T1-weighted images (T1WI) and T2-weighted images (T2WI) signal characteristics, degree and pattern of enhancement, dural tail sign, ginkgo leaf sign, and intervertebral foramen widening (IFW), were reviewed. Finally, a CART model and RF model were established based on the aforementioned features to evaluate their effectiveness in differentiating between the two types of tumors. Meanwhile, we also compared the performance of the ML models to the radiologists. The receiver operating characteristic (ROC) curve, accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were used to evaluate the models and clinicians' discrimination performance. Our investigation reveals significant variations in ten out of 11 variables in the training group and five out of 11 variables in the test group when comparing schwannomas and meningiomas (P<0.05). Ultimately, the CART model incorporated five variables: enhancement pattern, the presence of IFW, tumor location, maximum diameter, and T2WI signal intensity (SI). The RF model combined all 11 variables. The CART model, RF model, radiologist 1, and radiologist 2 achieved an area under the curve (AUC) of 0.890, 0.956, 0.681, and 0.723 in the training group, and 0.838, 0.922, 0.580, and 0.659 in the test group, respectively. The RF prediction model exhibits more exceptional performance than an experienced radiologist in discriminating intraspinal schwannomas from meningiomas. The RF model seems to be better in discriminating the two tumors than the CART model.

A case of severe pneumonia complicated by sepsis, endocarditis, and prolonged comatose state, with successful rehabilitation

Sepsis with acute organ dysfunction is an urgent problem of modern healthcare. A clinical case of a 38-year-old young woman with communityacquired severe viral-bacterial pneumonia complicated by sepsis, endocarditis, and multiple organ failure is presented. On the 68th day of her stay in the intensive care unit (ICU), the patient was brought out of comatose state and transferred to the pulmonology department for rehabilitation.The purpose of the work was to demonstrate the significance of this clinical case because of the urgent need to increase the effectiveness of treatment and long-term rehabilitation of patients with this severe comorbid pathology.Conclusion. Early diagnosis is extremely important to select effective treatment. The Quick SOFA (Sepsis-related sequential Organ Failure Assessment) score should be used to identify patients with suspected sepsis outside the ICU. This score is based on simple and accessible clinical characteristics that do not require laboratory analysis of homeostatic parameters. The patient was discharged in satisfactory condition to continue rehabilitation in outpatient settings.

STANLEy score: a new predictive model of 3-year major adverse cardiac events following "Full Metal Jacket" using new-generation drug-eluting stents

Abstract Introduction Percutaneous Coronary Intervention (PCI) using a "Full Metal Jacket" (FMJ) procedure (≥60 mm of continuous stent length) is often required to treat very long lesions, but its clinical efficacy and safety using contemporary drug-eluting stents (DES) remains to be determined. Objectives To identify predictors of major adverse cardiac events (MACE) associated with FMJ using new-generation DES. We sought to develop a simple model capable of accurately predict 3-year MACE following FMJ PCI. Methods A retrospective single-centre analysis of consecutive FMJ PCI performed between January 2015 and December 2018. Exclusion criteria were use of non-third generation DES, unsuccessful procedure and loss to follow-up. The primary endpoint was a composite of all-cause death, myocardial infarction, and target vessel revascularization. Demographic, clinical, angiographic, and procedural variables were evaluated. Logistic regression analysis was performed to determine independent predictors of outcome. Based on the results, a weighted scoring system was developed. Results The derivation cohort included 162 patients, mean age 65.9 ± 11.1 years, 83% male. The mean stent length was 73.8 ± 12.3 mm (range 60 to 116 mm), and the average number of stents was 2.74 ± 0.74 (range 2 to 5). 30-day, 1-year and 3-year MACE were 5.5%, 12.9% and 29.0%, respectively. Multivariate analysis identified diabEtes mellitus (hazard ratio (HR) 5.92; 95% confidence interval (CI) 1.57-11.54; p=0.011), Smoking habits (HR 8.37; 95%CI 3.05-15.55; p=0.001), Acute coronary syndrome (HR 3.51; 95%CI 1.030-8.98; p=0.045), Non-right coronary artery (HR 4.38; 95%CI 1.09-7.68; p=0.038), osTial lesion (HR 10.01; 95%CI 2.91-19.14; p=0.006) and severe caLcification (HR 8.59; 95%CI 2.65-15.72; p=0.001) as independent predictors of 3-year MACE. A composite score based on these 6 variables (STANLEy score) was created, ranging from 0 to 10 (Fig.1). High acuity was verified by ROC curve analysis (AUC 0.941, p &amp;lt;0.001). A score ≤ 3 was associated with a reduced probability (&amp;lt;15%) and a score ≥ 7 with high probability (&amp;gt;90%) of 3-year MACE. Conclusion The STANLEy score is a good predictive model which relies on 6 simple clinical and angiographic characteristics. This score may assist physicians in selecting high-risk patients for closer monitoring or aggressive antithrombotic strategy.Figure 1

Predicting survival of Iranian COVID-19 patients infected by various variants including omicron from CT Scan images and clinical data using deep neural networks

Purpose: The rapid spread of the COVID-19 omicron variant virus has resulted in an overload of hospitals around the globe. As a result, many patients are deprived of hospital facilities, increasing mortality rates. Therefore, mortality rates can be reduced by efficiently assigning facilities to higher-risk patients. Therefore, it is crucial to estimate patients' survival probability based on their conditions at the time of admission so that the minimum required facilities can be provided, allowing more opportunities to be available for those who need them. Although radiologic findings in chest computerized tomography scans show various patterns, considering the individual risk factors and other underlying diseases, it is difficult to predict patient prognosis through routine clinical or statistical analysis.Method: In this study, a deep neural network model is proposed for predicting survival based on simple clinical features, blood tests, axial computerized tomography scan images of lungs, and the patients' planned treatment. The model's architecture combines a Convolutional Neural Network and a Long Short Term Memory network. The model was trained using 390 survivors and 108 deceased patients from the Rasoul Akram Hospital and evaluated 109 surviving and 36 deceased patients infected by the omicron variant.Results: The proposed model reached an accuracy of 87.5% on the test data, indicating survival prediction possibility. The accuracy was significantly higher than the accuracy achieved by classical machine learning methods without considering computerized tomography scan images (p-value <= 4E-5). The images were also replaced with hand-crafted features related to the ratio of infected lung lobes used in classical machine-learning models. The highest-performing model reached an accuracy of 84.5%, which was considerably higher than the models trained on mere clinical information (p-value <= 0.006). However, the performance was still significantly less than the deep model (p-value <= 0.016).Conclusion: The proposed deep model achieved a higher accuracy than classical machine learning methods trained on features other than computerized tomography scan images. This proves the images contain extra information. Meanwhile, Artificial Intelligence methods with multimodal inputs can be more reliable and accurate than computerized tomography severity scores.

THU414 Asthma And Cardiometabolic Diseases As Major Risk Factors For Fractures In Women With Osteopenia: Nationwide Cohort Study

Abstract Disclosure: S. Kong: None. A. Jo: None. C. Park: None. K. Park: None. J. Yun: None. J. Kim: None. S. Kim: None. C. Shin: None. Background: The study aimed to demonstrate the risk factors for fractures and to develop prediction models for major osteoporotic and hip fractures in osteopenic patients using the nationwide cohort study in South Korea. Methods: The study was a retrospective nationwide study using the national screening program for transitional ages from the National Health Insurance Services database in Korea from 2008 to 2019. Primary outcomes were incident fracture events of major osteoporotic and hip fractures. Major osteoporotic and hip fracture events were defined as diagnostic and procedural codes. Patients were followed until the fragility fractures, death, or 2019, whichever came first. Results: All participants were 66-year-old females, with a mean body mass index was 25.0±3.1 kg/m2. During a median follow-up of 10.5 years, 26.9% and 6.7% of participants experienced major osteoporotic and hip fractures. In multivariate analysis, a history of fracture, asthma/chronic obstructive pulmonary disease (COPD), falls, diabetes mellitus and cerebrovascular diseases were significant risk factors for major osteoporotic (hazard ratio [HR] 2.35 for a history of fracture; 1.17 for asthma/COPD; 1.10 for falls; 1.12 for diabetes mellitus; 1.11 for cerebrovascular disease) and hip fractures (HR 1.75 for a history of fracture; 1.54 for diabetes mellitus; 1.27 for cerebrovascular disease; 1.17 for fall; 1.15 for asthma/COPD). The performances of the prediction models were area under the receiver operating curve of 0.73 and 0.75 for major osteoporotic and hip fractures. Conclusion: The study presented prediction models of major osteoporotic and hip fractures for osteopenia patients using simple clinical features, and showed that asthma and cardiometabolic diseases are major risk factors for fractures in women with osteopenia. Presentation: Thursday, June 15, 2023

A radiomics model fusing clinical features to predict microsatellite status preoperatively in colorectal cancer liver metastasis

PurposeTo study the combined model of radiomic features and clinical features based on enhanced CT images for noninvasive evaluation of microsatellite instability (MSI) status in colorectal liver metastasis (CRLM) before surgery.MethodsThe study included 104 patients retrospectively and collected CT images of patients. We adjusted the region of interest to increase the number of MSI-H images. Radiomic features were extracted from these CT images. The logistic models of simple clinical features, simple radiomic features, and radiomic features with clinical features were constructed from the original image data and the expanded data, respectively. The six models were evaluated in the validation set. A nomogram was made to conveniently show the probability of the patient having a high MSI (MSI-H).ResultsThe model including radiomic features and clinical features in the expanded data worked best in the validation group.ConclusionA logistic regression prediction model based on enhanced CT images combining clinical features and radiomic features after increasing the number of MSI-H images can effectively identify patients with CRLM with MSI-H and low-frequency microsatellite instability (MSI-L), and provide effective guidance for clinical immunotherapy of CRLM patients with unknown MSI status.

Chronic airway disease as a major risk factor for fractures in osteopenic women: Nationwide cohort study.

The study aimed to demonstrate the risk factors for fractures and to develop prediction models for major osteoporotic and hip fractures in osteopenic patients using the nationwide cohort study in South Korea. The study was a retrospective nationwide study using the national screening program for transitional ages from the National Health Insurance Services database in Korea from 2008 to 2019. Primary outcomes were incident fracture events of major osteoporotic and hip fractures. Major osteoporotic and hip fracture events were defined as diagnostic and procedural codes. Patients were followed until the fragility fractures, death, or 2019, whichever came first. All participants were 66-year-old females, with a mean body mass index was 25.0 ± 3.1 kg/m2. During a median follow-up of 10.5 years, 26.9% and 6.7% of participants experienced major osteoporotic and hip fractures. In multivariate analysis, a history of fracture, chronic airway disease, falls, diabetes mellitus and cerebrovascular diseases were significant risk factors for major osteoporotic (hazard ratio [HR] 2.35 for a history of fracture; 1.17 for chronic airway disease; 1.10 for falls; 1.12 for diabetes mellitus; 1.11 for cerebrovascular disease) and hip fractures (HR 1.75 for a history of fracture; 1.54 for diabetes mellitus; 1.27 for cerebrovascular disease; 1.17 for fall; 1.15 for chronic airway disease). The performances of the prediction models were area under the receiver operating curve of 0.73 and 0.75 for major osteoporotic and hip fractures. The study presented prediction models of major osteoporotic and hip fractures for osteopenia patients using simple clinical features.

Real world evidence of clinical predictors of glycaemic response to glucose-lowering drugs among Chinese with type 2 diabetes.

To examine whether simple clinical features can predict the 1-year glycaemic response to glucose-lowering drugs (GLDs) among Chinese with type 2 diabetes. We used data from a diabetes risk assessment and complication screening programme and electronic medical records. We used linear regression models to examine the association between clinical features and 1-year glycaemic response to GLDs. Use of metformin (n=15,433), sulphonylureas (SU) (n=15,190), dipeptidyl peptidase-4 inhibitor (DPP-4i) (n=7947), thiazolidinedione (TZD) (n=4107), and sodium-glucose cotransporter 2 inhibitors (SGLT-2i) (n=1883) were associated with a mean reduction of HbA1c ranging from 0.7% to 1.3% at one year. Men had a greater response to SU but a poorer response to metformin and TZD. Older age predicted a better response to all GLDs but not SGLT-2i, whereas increasing diabetes duration was associated with a poorer response to all GLDs except for DPP-4i. Obese patients responded greater to TZD and SGLT-2i but poorer to SU than those with normal weight. Patients with a higher level of triglycerides to high-density lipoprotein cholesterol ratio had a greater glycaemic response to TZD but a smaller response to SU and DPP-4i. Glycaemic response to GLDs differed considerably by clinical features among Chinese patients with type 2 diabetes.

Early prediction of the severe course, survival, and ICU requirements in acute pancreatitis by artificial intelligence

ObjectiveTo evaluate the success of artificial intelligence for early prediction of severe course, survival, and intensive care unit(ICU) requirement in patients with acute pancreatitis(AP). MethodsRetrospectively, 1334 patients were included the study. Severity is determined according to the Revised Atlanta Classification(RAC). The success of machine learning(ML) method was evaluated by 13 simple demographic, clinical, etiologic, and laboratory features obtained on ER admission. Additionally, it was evaluated whether Balthazar-computerized tomography severity index(CTSI) at 48-h contributed to success. The dataset was split into two parts, 90% for ML(of which 70% for learning and 30% for testing) and 10% for validation and 5-fold stratified sampling has been utilized. Variable Importance was used in the selection of features during training phase of machine. The Gradient Boost Algorithm trained the machine by KNIME analytics platform. SMOTE has been applied to increase the minority classes for training. The combined effects of the measured features were examined by multivariate logistic regression analysis and reciever operating curve curves of the prediction and confidence of the target variables were obtained. ResultsAccuracy values for the early estimation of Atlanta severity score, ICU requirement, and survival were found as 88.20%, 98.25%, and 92.77% respectively. When Balthazar-CTSI score is used, results were found as 91.02%, 92.25%, and 98% respectively. ConclusionsThe ML method we used successfully predicted the severe course, ICU requirement and survival, with promising accuracy values of over 88%. If 48-h Balthazar-CTSI is included in the calculation, the severity score and survival rates increase even more.

Abstract 10181: Impact of Modified H2FPEF Score on Chronic Limb Threatening Ischemia in Patients With Lower Extremity Artery Disease Who Underwent Endovascular Therapy

Introduction: Lower extremity artery disease (LEAD) is an arterial occlusive disease characterized by an insufficient blood supply to the lower limb arteries. The H2FPEF score, calculated using simple clinical characteristics and echocardiographic findings, has been developed to identify patients at high risk for heart failure (HF) with preserved ejection fraction. Purpose: This study aimed to assess the impact of modified H2FPEF scores on chronic limb-threatening ischemia (CLTI) in patients with LEAD. Methods: Since the definition of obesity differs by race, we calculated the modified H2FPEF score using a body mass index &gt; 25 kg/m2 in 293 patients with LEAD who underwent their first endovascular therapy. Patients were retrospectively followed for a median follow-up period of 718 days. The primary and secondary endpoints were newly developed CLTI and composite events, including mortality due to worsening HF or rehospitalization in addition to CLTI, respectively. Results: The modified H2FPEF score significantly increased with advancing Fontaine classes. Kaplan-Meier analysis demonstrated that the highly modified H2FPEF score group (≥ 3) had a higher incidence of newly developed CLTI and composite events than the low H2FPEF score group (&lt;3). Multivariate Cox proportional hazard analysis revealed that the modified H2FPEF score was an independent predictor of newly developed CLTI and composite events after adjustment for confounding risk factors. The net reclassification index and integrated discrimination improvement were significantly improved by adding the modified H2FPEF score to the basic predictors. Conclusions: The modified H2FPEF score was associated with LEAD severity and future CLTI development, suggesting that it could be a feasible marker for patients with LEAD.

Clinical and Molecular Features of KRAS-Mutated Lung Cancer Patients Treated with Immune Checkpoint Inhibitors.

Simple SummaryThe molecular and clinical features of KRAS-mutated lung cancer patients treated with immunotherapy have yet to be well characterized, and little information is known about resistance in these patients. The goal of this study is to better understand the survival results of KRAS-mutated patients who undergo immunotherapy treatment. For this effort, we have included 87 patients with NSCLC who received immunotherapy at the City of Hope, and we found that, among 87 patients, 32 had a KRAS G12C mutation (36.8%), 19 had G12V (21.9%), 18 had G12D (20.7%), 6 had G12A (6.9%), 3 had G12R (3.45%), and 10 had amplification (11.49%) and other uncommon mutations. G12D patients were found to respond differently compared to other KRAS-mutated patients. The OS with other KRAS comutations was not statistically significant, including STK11 and KEAP1. KRAS mutation subtypes such as G12D and comutations such as CDKN2/A and MET may modulate the immunotherapy responses and outcome in lung cancer.Background: The molecular and clinical features of KRAS-mutated lung cancer patients treated with immunotherapy have yet to be characterized, which could guide the development of therapeutics targeting KRAS with potential immuno-oncology treatment combinations. Research Question: Do KRAS-mutated patients with different subtypes and comutations have different clinical responses and overall survival (OS) to checkpoint inhibitors? Study Design and Methods: 87 patients with NSCLC at the City of Hope who received immune checkpoint inhibitors were identified and analyzed retrospectively. Tumor genomic alterations were extracted from the clinical data with next-generation sequencing using various platforms. Demographic, clinical, molecular, and pathological information was collected with the approval of the institutional review board of the City of Hope. OS was calculated if it was available at the study time point, and responses were determined according to the RECIST v1.1. Results: Among 87 patients, 32 had a KRAS G12C mutation (36.8%), 19 had G12V (21.9%), 18 had G12D (20.7%), 6 had G12A (6.9%), 3 had G12R (3.45%), and 10 had amplification (11.49%) and other uncommon mutations. G12D had a statistically significant Odds Ratio (OR) between patients who had responses and progression of the disease (OR (95% CI) = 0.31 (0.09–0.95), p < 0.05), with 5 G12D-mutated patients having responses and 11 G12D-mutated patients having progression of the disease. In the univariate analysis with OS, there was a trend of better OS in the G12D-mutated patients, with no statistically significant difference in terms of OS between the patients who had G12D mutation and the patients who had other KRAS mutations (HR (95% CI) = 0.53 (0.21–1.36), p = 0.185). The median OS was significantly worse with KRAS comutation CDKN2A/B loss (4.2 vs. 16.9 months, HR = 3.07 (1.09–8.69), p < 0.05) and MET (3.4 vs. 17 months, HR = 3.80 (1.44–10.05), p < 0.01), which were included for the multivariate analysis. The OS with other KRAS comutations was not statistically significant, including STK11 and KEAP1. Conclusion: KRAS mutation subtypes such as G12D and comutations such as CDKN2/A and MET may modulate the immunotherapy responses and outcomes in lung cancer.

Cerebral Blood Flow and Other Predictors of Responsiveness to Erenumab and Fremanezumab in Migraine-A Real-Life Study.

IntroductionMonoclonal antibodies (mAbs) showed efficacy in migraine prevention. The aim of this study was to check if baseline clinical parameters and cerebral blood flow (CBF) measured by transcranial Doppler (TCD) may help predict mAbs efficacy.MethodsElectronic charts of migraineurs treated with erenumab or fremanezumab, with baseline TCD evaluations were collected, including data on migraine type, pain localization, monthly migraine days (MMD), medication overuse headache (MOH), mean blood flow velocity (Vm), and pulsatility index (PI) in cerebral arteries.ResultsA total of 123 patients were enrolled, mean age 38, 75 years, 87 with chronic migraine, 61 with MOH, 72 were good responders (GR), and reported ≥50% reduction in MMD, 43 ≥75% reduction in MMD. Baseline Vm values in MCAs were significantly lower in GR as compared with non-responders. MAbs responsiveness ≥50% was positively associated with unilateral pain localization (OR: 6.53, 95% CI: 2.01–23.93; p = 0.003) and HIT-6 score (OR: 1.14, 95% CI: 1.01–1.30; p = 0.036) whereas negatively associated with Vm in right MCA (OR: 0.96, 95% CI: 0.92–0.99; p = 0.012), and having no relatives with migraine (OR: 0.40, 95% CI: 0.16–0.95; p = 0.040).ConclusionsBaseline Vm in MCA is lower in mAbs GR as compared with non-responders which may reflect increased secretion of CGRP with further vasodilation in GR. Simple clinical features and baseline CBF in anterior circulation might help to predict the patient's responsiveness.

A Nomogram for Predicting In-Hospital Major Adverse Cardio- and Cerebro-Vascular Events in Patients Undergoing Major Noncardiac Surgery: A Large-Scale Nested Case-Control Study.

PurposeFew evidence-based predictive tools are available to evaluate major adverse cardio- and cerebro-vascular events (MACCEs) before major noncardiac surgery. We sought to develop a new simple but effective tool for estimating surgical risk.Patients and MethodsUsing a nested case-control study design, we recruited 105 patients who experienced MACCEs and 481 patients without MACCEs during hospitalization from 10,507 patients undergoing major noncardiac surgery in Beijing Chaoyang hospital. Least absolute shrinkage and selection operator (LASSO) regression and likelihood ratio were applied to screen 401 potential features for logistic regression. A nomogram was constructed using the selected variables.ResultsChronic heart failure, valvular heart disease, preoperative serum creatinine >2.0 mg/dL, ASA class, neutrophil count and age were most associated with in-hospital MACCEs among all the factors. A new prediction model established based on these showed a good discriminatory ability (AUC, 0.758 [95% confidence interval (CI), 0.708–0.808] and a well-performed calibration curve (Hosmer–Lemeshow χ2 = 7.549, p = 0.479), which upheld in the 10-fold cross-validation (AUC, 0.742 [95% CI, 0.718–0.767]. This model also demonstrated an improved performance in comparison to the modified Revised Cardiac Risk Index (RCRI) score (increase in AUC by 0.119 [95% CI, 0.056–0.180]; NRI, 0.445 [95% CI, 0.237–0.653]; IDI, 0.133 [95% CI, 0.087–0.178]. The decision curve analysis showed a positive net benefit of our new model.ConclusionOur nomogram, which relies upon simple clinical characteristics and laboratory tests, is able to predict MACCEs in patients undergoing major noncardiac surgery. This prediction shows better discrimination than the standardized modified RCRI score, laying a promising foundation for further large-scale validation.

Precision Medicine in Diabetes.

Tailoring treatment or management to groups of individuals based on specific clinical, molecular, and genomic features is the concept of precision medicine. Diabetes is highly heterogenous with respect to clinical manifestations, disease progression, development of complications, and drug response. The current practice for drug treatment is largely based on evidence from clinical trials that report average effects. However, around half of patients with type 2 diabetes do not achieve glycaemic targets despite having a high level of adherence and there are substantial differences in the incidence of adverse outcomes. Therefore, there is a need to identify predictive markers that can inform differential drug responses at the point of prescribing. Recent advances in molecular genetics and increased availability of real-world and randomised trial data have started to increase our understanding of disease heterogeneity and its impact on potential treatments for specific groups. Leveraging information from simple clinical features (age, sex, BMI, ethnicity, and co-prescribed medications) and genomic markers has a potential to identify sub-groups who are likely to benefit from a given drug with minimal adverse effects. In this chapter, we will discuss the state of current evidence in the discovery of clinical and genetic markers that have the potential to optimise drug treatment in type 2 diabetes.

Hepatocellular Carcinoma in Non-Alcoholic Fatty Liver Disease: From Epidemiology to Diagnostic Approach.

Simple SummaryNon-alcoholic fatty liver disease (NAFLD) is expected to become the leading cause of hepatocellular carcinoma (HCC) in the near future. In this article, we review the current knowledge about the epidemiology, risk factors, pathogenesis, clinical presentation and diagnostic approach to HCC in NAFLD. Knowledge of these facts is of great importance to improve the early identification of patients that are at risk, allowing for early detection of HCC and, thus, an improvement in clinical outcomes. This is especially important given that around 30% of NAFLD-related HCCs develop in a non-cirrhotic liver. The presence of diabetes, male sex, older age and Hispanic race, in addition to liver cirrhosis, are the most important risk factors for HCC in this setting. In summarising the current knowledge of genetic susceptibility, metabolic derangements and immunological mechanisms involved in the pathogenesis of NAFLD-related HCC, we illustrate the need for further research on this intriguing topic.Non-alcoholic fatty liver disease (NAFLD) is becoming the leading cause of liver morbidity worldwide and, as such, represents the pathogenic background for the increasing incidence of hepatocellular carcinoma (HCC). The annual incidence of NAFLD-related HCC is expected to increase by 45–130% by 2030. Diabetes mellitus is the most important risk factor for HCC development in NAFLD, with the risk further increased when associated with other metabolic traits, such as obesity, arterial hypertension and dyslipidemia. The highest risk of HCC exists in patients with advanced fibrosis or cirrhosis, although 20–50% of HCC cases arise in NAFLD patients with an absence of cirrhosis. This calls for further investigation of the pathogenic mechanisms that are involved in hepatocarcinogenesis, including genetics, metabolomics, the influence of the gut microbiota and immunological responses. Early identification of patients with or at risk of NAFLD is of utmost importance to improve outcomes. As NAFLD is highly prevalent in the community, the identification of cases should rely upon simple demographic and clinical characteristics. Once identified, these patients should then be evaluated for the presence of advanced fibrosis or cirrhosis and subsequently enter HCC surveillance programs if appropriate. A significant problem is the early recognition of non-cirrhotic NAFLD patients who will develop HCC, where new biomarkers and scores are potential solutions to tackle this issue.

Deep neural survival networks for cardiovascular risk prediction: The Multi-Ethnic Study of Atherosclerosis (MESA)

BackgroundThere is growing interest in utilizing machine learning techniques for routine atherosclerotic cardiovascular disease (ASCVD) risk prediction. We investigated whether novel deep learning survival models can augment ASCVD risk prediction over existing statistical and machine learning approaches. Methods6814 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) were followed over 16 years to assess incidence of all-cause mortality (mortality) or a composite of major adverse events (MAE). Features were evaluated within the categories of traditional risk factors, inflammatory biomarkers, and imaging markers. Data was split into an internal training/testing (four centers) and external validation (two centers). Both machine learning (COXPH, RSF, and lSVM) and deep learning (nMTLR and DeepSurv) models were evaluated. ResultsIn comparison to the COXPH model, DeepSurv significantly improved ASCVD risk prediction for MAE (AUC: 0.82 vs. 0.80, P ≤ 0.001) and mortality (AUC: 0.87 vs. 0.84, P ≤ 0.001) with traditional risk factors alone. Implementing non-categorical NRI, we noted a >40% increase in correct reclassification compared to the COXPH model for both MAE and mortality (P ≤ 0.05). Assessing the relative risk of participants, DeepSurv was the only learning algorithm to develop a significantly improved risk score criteria, which outcompeted COXPH for both MAE (4.22 vs. 3.61, P = 0.043) and mortality (6.81 vs. 5.52, P = 0.044). The addition of inflammatory or imaging biomarkers to traditional risk factors showed minimal/no significant improvement in model prediction. ConclusionDeepSurv can leverage simple office-based clinical features alone to accurately predict ASCVD risk and cardiovascular outcomes, without the need for additional features, such as inflammatory and imaging biomarkers.