Harrell's C-index Research Articles

Association between high-density-lipoprotein cholesterol and risk of abdominal aortic aneurysm among males and females aged 60 years and over.

Whether high-density-lipoprotein cholesterol (HDL-c) has a protective role against abdominal aortic aneurysm (AAA) development in both older males and females remains uncertain. This study aims to assess the sex-specific association between HDL-c and incident AAA in older adults from the UK Biobank. This cohort study included 86,184 males and 95,682 females aged ≥60 years from the UK biobank. Baseline HDL-c was modelled either as a continuous or categorical variable. The primary outcome was incident AAA. Cox proportional-hazard models were used for sex-stratified analysis, adjusting for baseline confounders. Restricted cubic splines were plotted to visualise any non-linear relationship. Harrell's C-index was calculated to assess the added value of HDL-c to the discrimination of model including age and smoking. Over a mean follow-up of 14.4 years, 1,549 and 328 incident AAA were observed in males and females, respectively. Adjusted HR (95%CI) for AAA with a 1 mmol/L HDL-c increase was 0.26 (0.21- 0.32) and 0.31 (0.21-0.46) in males and females, respectively, both p<0.001. Consistent with the results from Cox model modelling HDL-c as a categorical variable showing an inverse dose-dependent relationship between HDL-c and incident AAA in both sexes, restricted cubic splines confirmed the monotonic, inverse associations. Adding HDL-c to a model including age and smoking significantly improve the model discrimination for AAA in both sexes (C-index +2.1% in males and +1.5% in females, both p<0.05). This study revealed a significant association between low HDL-c levels and high risk of incident AAA in both older males and females, suggesting the potential clinical utility of HDL-c for AAA risk stratification. Our study was limited by its observational design and presence of possible residual confounding. Studies using real-world data are warranted to evaluate the practical implications of incorporating HDL-c into AAA screening guidelines and its impact on patient outcomes.

Dual-Phase Enhanced CT-Derived Radiomics Nomogram for Progression-Free Survival Prediction in Stage IV Lung Adenocarcinoma.

The objective is to establish a radiomics nomogram (Rad-nomogram) using dual-phase enhanced computed tomography (DPE-CT) for the prediction of progression-free survival (PFS) in patients diagnosed with stage IV lung adenocarcinoma (ADC). From DPE-CT scans, radiomic characteristics were retrieved from 133 patients diagnosed with stage IV lung ADC. Clinical data were analyzed using univariate and multivariate Cox regression analyses. The radiomics signature was combined with clinical features employing multivariate Cox analysis in order to develop a Rad-nomogram. The predictive efficiency of the nomogram was evaluated using survival studies, such as Kaplan-Meier curves and Harrell's C-index. The benefits and clinical utility of various models were compared using the net reclassification index (NRI), decision curve analysis (DCA), and integrated discrimination improvement (IDI). In the test cohort, the C-indexes for the clinical, artery, and vein phase CT models were 0.675, 0.691, and 0.678, respectively. The dual-phase achieved a C-index of 0.731, exceeding the CT model, while the developed nomogram reached a C-index of 0.783. The Kaplan-Meier survival study classified patients into low-risk and high-risk groups related to PFS using the Rad-nomogram (p < 0.05). The Rad-nomogram demonstrated a greater net advantage when compared with clinical and Rad models, as indicated by positive values of the NRI and IDI (ranging from 11.6% to 52.6%, p < 0.05). The Rad-nomogram, employing DPE-CT scans, offers a promising approach to predict PFS in individuals diagnosed with stage IV lung ADC.

Prediction of Hepatocellular Carcinoma After Hepatitis C Virus Sustained Virologic Response Using a Random Survival Forest Model.

Postsustained virologic response (SVR) screening following clinical guidelines does not address individual risk of hepatocellular carcinoma (HCC). Our aim is to provide tailored screening for patients using machine learning to predict HCC incidence after SVR. Using clinical data from 1,028 SVR patients, we developed an HCC prediction model using a random survival forest (RSF). Model performance was assessed using Harrel's c-index and validated in an independent cohort of 737 SVR patients. Shapley additive explanation (SHAP) facilitated feature quantification, whereas optimal cutoffs were determined using maximally selected rank statistics. We used Kaplan-Meier analysis to compare cumulative HCC incidence between risk groups. We achieved c-index scores and 95% CIs of 0.90 (0.85 to 0.94) and 0.80 (0.74 to 0.85) in the derivation and validation cohorts, respectively, in a model using platelet count, gamma-glutamyl transpeptidase, sex, age, and ALT. Stratification resulted in four risk groups: low, intermediate, high, and very high. The 5-year cumulative HCC incidence rates and 95% CIs for these groups were as follows: derivation: 0% (0 to 0), 3.8% (0.6 to 6.8), 26.2% (17.2 to 34.3), and 54.2% (20.2 to 73.7), respectively, and validation: 0.7% (0 to 1.6), 7.1% (2.7 to 11.3), 5.2% (0 to 10.8), and 28.6% (0 to 55.3), respectively. The integration of RSF and SHAP enabled accurate HCC risk classification after SVR, which may facilitate individualized HCC screening strategies and more cost-effective care.

Different ways of diagnosing selective glomerular hypofiltration syndromes such as shrunken pore syndrome and the associated increase in mortality.

In 2015, a selective decrease in the glomerular filtration of middle-sized molecules such as cystatin C compared to small molecules such as creatinine was first described and tentatively termed "Shrunken pore syndrome." Numerous studies have thereafter found an association between this syndrome (defined by a low eGFRcystatin C to eGFRcreatinine ratio) and mortality and morbidity. In 2023, the syndrome was renamed selective glomerular hypofiltration syndromes (SGHS) as shrunken pores are not the only pathophysiological mechanism. Recently, some studies have used the difference between eGFRcystatin C and eGFRcreatinine to describe a similar disorder, and this investigation compares the two measures. Using a cohort of 2781 adults with a median follow-up of 5.6 years, referred for determination of glomerular filtration rate (GFR), estimated GFR (eGFR) was determined using four equations. SGHS was defined using the eGFRdifference and the eGFRratio and association to mortality investigated through adjusted Cox proportional hazard models. From each adjusted regression model, Harrell's C-index and 95% confidence intervals were calculated. Both measures were associated with mortality. No significant differences concerning hazard ratios or Harrell's C-index were found between the two measures to estimate mortality, and both identified SGHS and increased mortality in a subpopulation of 567 "healthy" individuals with no prior diagnosis and with no kidney disorder according to the kidney disease improving global outcomes-criteria. The eGFRdifference is not superior to the eGFRratio in diagnosing SGHS or estimating mortality. However, as the two measures do not identify the same subpopulation, using them simultaneously might improve risk stratification.

Do Time-Dependent Repeated Measures of Anthropometric and Body Composition Indices Improve the Prediction of Incident Diabetes in the Cohort Study? Findings from a Community-Based Korean Genome and Epidemiology Study.

Cumulative evidence consistently shows that anthropometric and body composition measurements are strongly linked to the risk of incident diabetes, typically based on baseline measurements. This study aims to assess whether repeated measurements enhance the prediction of diabetes risk beyond baseline assessments alone. We utilized data from a 16-year population-based follow-up cohort within the Korean Genome and Epidemiology Study, comprising 6,030 individuals aged 40 to 69 years at baseline. We included eight indices: a body shape index (ABSI), body adiposity index (BAI), waist circumference (WC), body mass index (BMI), waist-to-hip ratio (WHR), weight-adjusted skeletal muscle index (SMI), percent body fat, and fat-to-muscle ratio. The effect of these measurements for incident diabetes was estimated using Harrell's C-indexes and hazard ratios with 95% confidence intervals, employing time-dependent Cox proportional hazard models. Over the 16-year follow-up, 939 new diabetes cases were identified (cumulative incidence, 15.6%). The median number of indicator measurements per participant was eight. The basic model, including 10 features (sex, age, education levels, physical activity, alcohol intake, current smoking, total energy intake, dietary diversity score, and log-transformed C-reactive protein levels, and quartiles of unweighted genetic risk score at baseline), yielded a Harrell's C-index of 0.610. The highest C-index in repeated measurements was for WC (0.668) across the general population, weight-adjusted SMI in men, and WHR in women. However, except for ABSI and BAI, the diabetes predictive power of the other indicators was comparable. Additionally, repeated measurements of WC, BMI, and WHR in women were found to contribute to improved discrimination compared to baseline measurements, but not in men. Utilizing repeated measurements of general and central adiposity to predict diabetes may be helpful in predicting hidden risks, especially in women.

Prognosis analysis and nomogram for predicting lateral lymph node metastasis in Medullary Thyroid Microcarcinoma.

Currently, the incidence rate of Medullary Thyroid Microcarcinoma (micro-MTC) has an increasing trend, but the incidence of LNM and prognosis were still ambiguous. We analyzed the status of neck LNM of micro-MTC patients and created a prognostic nomogram to predict the probability of lateral lymph node metastasis (LLNM) for clinical practice. This is a retrospective study included patients with micro-MTC from SEER database for the period from 2004 to 2017 and patients from our medical center for the period from 2011 to 2019. A nomogram was constructed and the accuracy and clinical practicability were separately tested by Harrell's C-indexes, calibration plots, Receiver operating characteristic curve (ROC) and decision curve analyses (DCA). A total of 413 patients with micro-MTC from SEER database and 64 patients with micro-MTC from our department enrolled in the study. There were 16.0% and 9.4% cases in SEER database and 39.1% and 25.0% cases in our department appeared LNM and LLNM, respectively. Besides, a nomogram was constructed to assess the incidence of LLNM with good C-index, which was 0.850 in training cohort and 0.856 in validation cohort. The results of the area under the curve (AUC) were 0.830 in training cohort, 0.801 in validation cohort and 0.832 in external testing cohort, respectively. A relatively high rate of LLNM than expected was found, which should be emphasized. The prediction model could facilitate clinicians to assess the probability of LLNM and make a personalized treatment strategy.

International Metabolic Prognostic Index Is Superior to Other Metabolic Tumor Volume-Based Prognostication Methods in a Real-Life Cohort of Diffuse Large B-Cell Lymphoma.

Baseline metabolic tumor volume (MTV) is a promising prognostic marker in diffuse large B-cell lymphoma (DLBCL). We assessed the prognostic value of 4 novel metabolic risk scores in a real-life DLBCL cohort and compared them with the revised international prognostic index (IPI). Methods: We included a consecutive series of untreated DLBCL, not otherwise specified cases that were diagnosed in our hospital from 2008 to 2021 with available baseline [18F]FDG PET/CT. Clinical data were collected retrospectively, including the individual components of the revised IPI. MTV and other radiomic features, including lesion dissemination and tumor volume surface ratio, were calculated. Four novel metabolic risk scores including the international metabolic prognostic index (IMPI), the MTV/World Health Organization performance status, the MTV/standardized maximum distance, and clinical PET models were used to calculate the risk of progression using predefined cutoffs. Survival outcomes considered were 3-y progression free survival (PFS), 3-y time to progression (TTP), and 3-y overall survival (OS). The Harrell C-index was used to assess the discriminative performance of the risk scores. A multivariable model was built. Results: We included 355 DLBCL, not otherwise specified cases with a median MTV of 219 cm3 (range, 0-5,656 cm3). The IMPI had the highest C-index for 3-y PFS, 3-y TTP, and 3-y OS among the 4 metabolic risk scores (0.674, 0.696, and 0.677, respectively). For the 3-y TTP, the IMPI outperformed the strongest clinical risk score, the IPI, although the difference in the Harrell C-indices was small (0.696 vs. 0.693). Regarding the 3-y PFS and 3-y OS, the IPI has the highest C-index of all risk scores (0.696 and 0.693). The IMPI, the MTV/World Health Organization performance status, and the IPI score can recognize a poor risk group with a 3-y OS below 50% (43%, 32%, and 39%, respectively). In multivariable analysis, the IMPI remains an independent prognostic factor (P = 0.0089; hazard ratio, 1.207; 95% CI, 1.048-1.389). MTV and standardized maximum distance have the strongest prognostic values when used as a continuous variable. The tumor volume surface ratio has no significant prognostic value in our analysis. Conclusion: The IMPI has the strongest prognostic performance compared with the other 3 novel metabolic risk scores. However, in our real-world dataset, the IMPI could not replace the IPI, and further prospective trials are needed to compare their performance.

Which Scoring System Best Predicts Long-term Survival in Patients with Spinal Metastasis in the Era of Targeted Systemic Treatment? A Comparative Study of Eight Prognostic Models.

Retrospective observational study. To evaluate the accuracy of eight scoring systems, including the Tomita, modified Tokuhashi, modified Bauer, Rades, Oswestry Spinal Risk index (OSRI), Lei, New England Spinal Metastasis Score, and Skeletal Oncology Research Group (SORG) nomogram, for predicting long-term survival of patients with spinal metastasis. Predicting the prognosis of spinal metastasis is vital for surgical decisions, yet the effectiveness of existing scoring systems in identifying long-term survival remains unclear. 456 cases were finally included. Prognostic scores were compared with survival outcomes. Receiver operating characteristic (ROC) curves were analyzed for the entire cohort and across three distinct time periods to evaluate the area under the curve (AUC) for 1-year and 2-year survival, alongside Harrell's C-statistic. The mean patient age was 58.9 years, and the median survival time was 8.6 months. For the entire cohort, the SORG nomogram, OSRI, and modified Tokuhashi scores yielded Harrell's C-index values of 0.64, 0.63, and 0.62, respectively. For 1-year survival prediction, the SORG nomogram, OSRI, and modified Tokuhashi score demonstrated moderate discriminative power, with AUC values of 0.72, 0.71, and 0.70, respectively. Similarly, for 2-year survival prediction, the modified Tokuhashi score, SORG nomogram, and OSRI also revealed moderate discriminative power, with AUC values of 0.73, 0.72, and 0.70, respectively. For patients who underwent surgery in the most recent period, OSRI demonstrated the highest predictive accuracy for 1-year survival, with a Harrell's C-index of 0.63 and an AUC of 0.68, and 2-year survival, with a Harrell's C-index of 0.63 and an AUC of 0.64. Most scoring systems exhibited low discriminative power, with only the SORG nomogram, OSRI, and modified Tokuhashi scores demonstrating moderate power for predicting long-term survival. In the most recent period, the OSRI demonstrated the highest predictive accuracy for both 1-year and 2-year survival.

The added value of deep learning to submaximal exercise electrocardiogram for the 10-year prediction of major adverse cardiovascular and cerebrovascular events

Abstract Background The added value of exercise ECG data to traditional risk factors in predicting cardiovascular events remains unclear. Deep learning of ECG signals has demonstrated state-of-the-art performance in detecting subtle abnormalities indicative of cardiovascular disease. We investigated whether deep learning analysis of exercise ECGs could improve the prediction of major cardiovascular and cerebrovascular events (MACCE) with respect to established risk models in a large population-based cohort. Purpose To assess the added value of submaximal exercise ECG measurements to established cardiovascular risk prediction models. Methods We obtained ECG recordings from 41,076 UK Biobank participants without cardiovascular disease (CVD) who underwent a submaximal exercise ECG test. We obtained ECG recordings from 41,076 UK Biobank participants without cardiovascular disease (CVD) who underwent a submaximal exercise ECG test. We created two deep neural network ECG risk scores: one derived from conventional ECG parameters, measured at rest, peak exercise and late-stage recovery, and another based on the complete ECG (Q-ECG). We assessed the added value of conventional ECG parameters and Q-ECG risk scores to the UK's current prediction algorithm (QRISK3). We estimate the association between ECG parameters and MACCE, using Cox Proportional hazard models, following adjustment for traditional risk factors. All models were internally validated via 5-fold cross-validation and 1000 bootstrap iterations. Predictive performance was evaluated using Harrel's C-index, Net Reclassification Index (NRI) and net benefit. Findings: Incident MACCE was reported in 4,082 (9.9%) individuals in the study population and 3,463(9.7%) individuals with valid ECG parameters over a median follow-up period of 12.5 years. We found combined conventional ECG and Q-ECG scores were independently associated with MACCE, following adjustment for multiple testing: adjusted hazard ratio [HZ] = 1.76 (95% CI:1.63-1.91); HZ = 1.14 (95% CI:1.10-1.18), respectively, per standard deviation increase. Both conventional ECG parameters and Q-ECG were predictive of MACCE, independent of clinical risk factors, C-index = 0.64 (95%CI 0.63-0.65); net benefit = 0.09(95% CI 0.07 - 0.11) and C-index = 0.56 (95% CI 0.55-0.57); net benefit = 0.07(95% CI 0.05 - 0.09). ECG measurement's modestly improved model discrimination over the bassline QRISK3 risk score when combined with QRISK3 risk factors for conventional markers and Q-ECG score, respectively; ΔC-index 0.03 (95% CI: 0.02 – 0.04) and ΔC-index 0.03 (95% CI: 0.02 – 0.03); However, we observed no significant improvements in classification at the current recommended threshold of 10%. Conclusion In individuals without a history of prior cardiovascular disease, ECG measures independently predict the risk of MACCE. When combined with QRISK3, neural-network-derived ECG risk scores marginally improve cardiovascular risk prediction over QRISK3.

ABLE-SCORE, a simplified risk score for major adverse cardiovascular outcomes in left ventricular hypertrabeculation: a multicenter longitudinal cohort study.

Left ventricular hypertrabeculation (LVHT) is a heterogeneous entity with life-threatening complications and variable prognosis. However, there are limited prediction models available to identify individuals at high risk of adverse outcomes, and the current risk score in LVHT is comparatively complex for clinical practice. This study aimed to develop and validate a simplified risk score to predict major adverse cardiovascular events (MACE) in LVHT. This multicenter longitudinal cohort study consecutively enrolled morphologically diagnosed LVHT patients between January 2009 and December 2020 at Fuwai Hospital (derivation cohort, n = 300; internal validation cohort, n = 129), and between January 2014 and December 2022 at two national-level medical centers (external validation cohort, n = 95). The derivation/internal validation cohorts and the external validation cohort were followed annually until December 2022 and December 2023, respectively. MACE was defined as a composite of all-cause mortality, heart transplantation/left ventricular assist device implantation, cardiac resynchronization therapy, malignant ventricular arrhythmia, and thromboembolism. A simplified risk score, the ABLE-SCORE, was developed based on independent risk factors in the multivariable Cox regression predictive model for MACE, and underwent both internal and external validations to confirm its discrimination, calibration, and clinical applicability. A total of 524 LVHT patients (43.5 ± 16.6years, 65.8% male) were included in the study. The ABLE-SCORE was established using four easily accessible clinical variables: age at diagnosis, N-terminal pro-brain natriuretic peptide levels, left atrium enlargement, and left ventricular ejection fraction ≤ 40% measured by echocardiography. The risk score showed excellent performance in discrimination, with Harrell's C-index of 0.821 [95% confidence interval (CI), 0.772-0.869], 0.786 (95%CI, 0.703-0.869), and 0.750 (95%CI, 0.644-0.856) in the derivation, internal validation, and external validation cohort, respectively. Calibration plots of the three datasets suggested accurate agreement between the predicted and observed 5-year risk of MACE in LVHT. According to decision curve analysis, the ABLE-SCORE displayed greater net benefits than the existing risk score for LVHT, indicating its strength in clinical applicability. A simplified and efficient risk score for MACE was developed and validated using a large LVHT cohort, making it a reliable and convenient tool for the risk stratification and clinical management of patients with LVHT.

Exploring the Potential of Enhanced Prognostic Performance of NCCN-IPI in Diffuse Large B-Cell Lymphoma by Integrating Tumor Microenvironment Markers: Stromal FOXC1 and Tumor pERK1/2 Expression.

FOXC1 and ERK1-2 are proteins implicated in aggressive biological behavior of various malignancies including lymphomas. We investigate the additive prognostic value of stromal FOXC1 expression and tumor phosphorylated ERK1-2 (pERK1-2) expression to the established National Comprehensive Cancer Network International Prognostic Index (NCCN-IPI), in 92 diffuse large B-cell lymphoma (DLBCL) cases. Multidimensional analysis using statistics and machine learning (ML) models assessed prognostic value of established clinicopathologic variables with stromal FOXC1 and tumor pERK1-2 expressions. Both high FOXC1 stroma group and high pERK1-2 tumor group were significantly associated with shorter progression-free survival (PFS) and overall survival (OS) compared with low group (p = 0.015, 0.034 and p = 0.025, 0.025 each respectively). In multivariable analysis, high FOXC1 stromal expression was an independent prognostic factor of OS (p = 0.037). The addition of stromal FOXC1 and tumor pERK1-2 to the NCCN-IPI score significantly improved prediction of time to death compared with NCCN-IPI score alone (Harrell's C-index = 0.801 vs. 0.764; p = 0.030). ML models reconfirmed the addition of stromal FOXC1 expression and tumor pERK1-2 to NCCN-IPI score had the highest C-index (0.952) among combinations. Stromal FOXC1 and tumor pERK1-2 were determinants of DLBCL prognosis, whose addition significantly improved prognostic performance of the NCCN-IPI.

Deep-learning-driven optical coherence tomography analysis for cardiovascular outcome prediction in patients with acute coronary syndrome.

Optical coherence tomography (OCT) can identify high-risk plaques indicative of worsening prognosis in patients with acute coronary syndrome (ACS). However, manual OCT analysis has several limitations. In this study, we aim to construct a deep-learning model capable of automatically predicting ACS prognosis from patient OCT images following percutaneous coronary intervention (PCI). Post-PCI OCT images from 418 patients with ACS were input into a deep-learning model comprising a convolutional neural network (CNN) and transformer. The primary endpoint was target vessel failure (TVF). Model performances were evaluated using Harrell's C-index and compared against conventional models based on human observation of quantitative (minimum lumen area, minimum stent area, average reference lumen area, stent expansion ratio, and lesion length) and qualitative (irregular protrusion, stent thrombus, malapposition, major stent edge dissection, and thin-cap fibroatheroma) factors. GradCAM activation maps were created after extracting attention layers by using the transformer architecture. A total of 60 patients experienced TVF during follow-up (median 961 days). The C-index for predicting TVF was 0.796 in the deep-learning model, which was significantly higher than that of the conventional model comprising only quantitative factors (C-index: 0.640) and comparable to that of the conventional model, including both quantitative and qualitative factors (C-index: 0.789). GradCAM heat maps revealed high activation corresponding to well-known high-risk OCT features. The CNN and transformer-based deep-learning model enabled fully automatic prognostic prediction in patients with ACS, with a predictive ability comparable to a conventional survival model using manual human analysis. The study was registered in the University Hospital Medical Information Network Clinical Trial Registry (UMIN000049237).

A Machine Learning Approach for Predicting Biochemical Outcome After PSMA-PET-Guided Salvage Radiotherapy in Recurrent Prostate Cancer After Radical Prostatectomy: Retrospective Study.

Salvage radiation therapy (sRT) is often the sole curative option in patients with biochemical recurrence after radical prostatectomy. After sRT, we developed and validated a nomogram to predict freedom from biochemical failure. This study aims to evaluate prostate-specific membrane antigen-positron emission tomography (PSMA-PET)-based sRT efficacy for postprostatectomy prostate-specific antigen (PSA) persistence or recurrence. Objectives include developing a random survival forest (RSF) model for predicting biochemical failure, comparing it with a Cox model, and assessing predictive accuracy over time. Multinational cohort data will validate the model's performance, aiming to improve clinical management of recurrent prostate cancer. This multicenter retrospective study collected data from 13 medical facilities across 5 countries: Germany, Cyprus, Australia, Italy, and Switzerland. A total of 1029 patients who underwent sRT following PSMA-PET-based assessment for PSA persistence or recurrence were included. Patients were treated between July 2013 and June 2020, with clinical decisions guided by PSMA-PET results and contemporary standards. The primary end point was freedom from biochemical failure, defined as 2 consecutive PSA rises >0.2 ng/mL after treatment. Data were divided into training (708 patients), testing (271 patients), and external validation (50 patients) sets for machine learning algorithm development and validation. RSF models were used, with 1000 trees per model, optimizing predictive performance using the Harrell concordance index and Brier score. Statistical analysis used R Statistical Software (R Foundation for Statistical Computing), and ethical approval was obtained from participating institutions. Baseline characteristics of 1029 patients undergoing sRT PSMA-PET-based assessment were analyzed. The median age at sRT was 70 (IQR 64-74) years. PSMA-PET scans revealed local recurrences in 43.9% (430/979) and nodal recurrences in 27.2% (266/979) of patients. Treatment included dose-escalated sRT to pelvic lymphatics in 35.6% (349/979) of cases. The external outlier validation set showed distinct features, including higher rates of positive lymph nodes (47/50, 94% vs 266/979, 27.2% in the learning cohort) and lower delivered sRT doses (<66 Gy in 57/979, 5.8% vs 46/50, 92% of patients; P<.001). The RSF model, validated internally and externally, demonstrated robust predictive performance (Harrell C-index range: 0.54-0.91) across training and validation datasets, outperforming a previously published nomogram. The developed RSF model demonstrates enhanced predictive accuracy, potentially improving patient outcomes and assisting clinicians in making treatment decisions.

Stroke Measures Analysis of pRognostic Testing-Mortality nomogram predicts long-term mortality after ischemic stroke.

Predicting long-term mortality is essential for understanding prognosis and guiding treatment decisions in patients with ischemic stroke. Therefore, this study aimed to develop and validate the method for predicting 1- and 5-year mortality after ischemic stroke. We used data from the linked dataset comprising the administrative claims database of the Health Insurance Review and Assessment Service and the Clinical Research Center for Stroke registry data for patients with acute stroke within 7 days of onset. The outcome was all-cause mortality following ischemic stroke. Clinical variables linked to long-term mortality following ischemic stroke were determined. A nomogram was constructed based on the Cox's regression analysis. The performance of the risk prediction model was evaluated using the Harrell's C-index. This study included 42,207 ischemic stroke patients, with a mean age of 66.6 years and 59.2% being male. The patients were randomly divided into training (n = 29,916) and validation (n = 12,291) groups. Variables correlated with long-term mortality in patients with ischemic stroke, including age, sex, body mass index, stroke severity, stroke mechanisms, onset-to-door time, pre-stroke dependency, history of stroke, diabetes mellitus, hypertension, coronary artery disease, chronic kidney disease, cancer, smoking, fasting glucose level, previous statin therapy, thrombolytic therapy, such as intravenous thrombolysis and endovascular recanalization therapy, medications, and discharge modified Rankin Scale were identified as predictors. We developed a predictive system named Stroke Measures Analysis of pRognostic Testing-Mortality (SMART-M) by constructing a nomogram using the identified features. The C-statistics of the nomogram in the developing and validation groups were 0.806 (95% confidence interval (CI), 0.802-0.812) and 0.803 (95% CI, 0.795-0.811), respectively. The SMART-M method demonstrated good performance in predicting long-term mortality in ischemic stroke patients. This method may help physicians and family members understand the long-term outcomes and guide the appropriate decision-making process.

Spectral Ultrasound Combined With Clinical Pathological Parameters in Prediction of Axillary Lymph Node Metastasis in Breast Cancer.

To explore the clinical value of the nomogram based on spectral Doppler ultrasound combined with clinical pathological parameter in predicting axillary lymph node metastasis in breast cancer. We prospectively gathered clinicopathologic and ultrasonic data from 240 patients confirmed breast cancer. The risk factors of axillary lymph node metastasis were analyzed by univariate and multivariate logistic regression, and the prediction model was established. The model calibration, predictive ability, and diagnostic efficiency in the training set and the testing set were analyzed by receiver operating characteristic curve and calibration curve analysis, respectively. Univariate analysis showed that lymph node metastasis was related with tumor size, Ki-67, axillary ultrasound, ultrasound spectral quantitative parameter, internal echo, and calcification (P < .05). Multivariate logistic regression analysis showed that the Ki-67, axillary ultrasound, quantitative parameter (the mean of the mid-band fit in tumor and posterior tumor) were independent risk factors of axillary lymph node metastasis (P < .05). The models developed using Ki-67, axillary ultrasound, and quantitative parameters for predicting axillary lymph node metastasis demonstrated an area under the receiver operating characteristic curve of 0.83. Additionally, the prediction model exhibited outstanding predictability for axillary lymph node metastasis, as evidenced by a Harrell C-index of 0.83 (95% confidence interval 0.73-0.93). Axillary ultrasound combined with Ki-67 and spectral ultrasound parameters has the potential to predict axillary lymph node metastasis in breast cancer, which is superior to axillary ultrasound alone.

Nomograms for predicting recurrence of HER2-positive breast cancer with different HR status based on ultrasound and clinicopathological characteristics.

This study aimed to identify ultrasound and clinicopathological characteristics related to recurrence in HER2-positive (HER2+) breast cancer, and to develop nomograms for predicting recurrence. In this dual-center study, we retrospectively enrolled 570 patients with HER2+ breast cancer. The ultrasound and clinicopathological characteristics of hormone receptor (HR)-/HER2+ patients and HR+/HER2+ patients were analyzed separately according to HR status. Eighty percent of the original samples from HR-/HER2+ and HR+/HER2+ patients were extracted by bootstrap sampling as the training cohorts, while the remaining 20% were used as the external validation cohorts. Informative characteristics were screened through univariate and multivariable Cox regression in the training cohorts and used to develop nomograms for predicting recurrence. The predictive accuracy was calculated using Harrell's C-index and calibration curves. Three informative characteristics (axillary nodal status, calcification, and Adler degree) were identified in HR-/HER2+ patients, and another three (histological grade, axillary nodal status, and echogenic halo) in HR+/HER2+ patients. Based on these, two separate nomograms were constructed to assess recurrence risk. In the training cohorts, the C-index was 0.740 (95% CI: 0.667-0.811) for HR-/HER2+ nomogram, and 0.749 (95% CI: 0.679-0.820) for HR+/HER2+ nomogram. In the validation cohorts, the C-index was 0.708 (95% CI: 0.540-0.877) for HR-/HER2+ group, and 0.705 (95% CI: 0.557-0.853) for HR+/HER2+ group. The calibration curves also indicated the excellent accuracy of the nomograms. Ultrasound performance of HER2+ breast cancers with different HR status was significantly different. Nomograms integrating ultrasound and clinicopathological characteristics exhibited favorable performance and have the potential to serve as a reliable method for predicting recurrence in heterogeneous breast cancer.

Predicting oral and esophageal cancers by one model in a Chinese prospective cohort study

ObjectiveOral and esophageal cancers are both upper gastrointestinal cancers that share a number of risk factors. However, most previous risk prediction models only focused on one of these two types of cancer. There is no single model that could predict both cancers simultaneously. Our objective was to develop a model specifically tailored for oral and esophageal cancers. MethodsFrom 1996 to 2007, a total of 431,460 subjects aged 20 and older without a history of cancer at baseline were included and were monitored for an average duration of 7.3 years in Taiwan, China. A total of 704 cases of oral and esophageal cancers were detected. We utilized both univariate and multivariate COX regression for screening predictors and constructing the model. We evaluated the goodness of fit of the model based on discriminatory accuracy, Harrell's C-index, and calibration. ResultsFinally, we developed a Cox regression model using the twelve most significant variables: age, gender, alcohol consumption, betel chewing, smoking status, history of oral ulceration, educational level, marital status, oropharynx status, family history of nasopharyngeal carcinoma, volume ratio of blood cell, and gamma-glutamyl transferase. The AUC (area under the curve) for the complete model was 0.82. Additionally, the C-index was 0.807 (with a 95 % confidence interval ranging from 0.789 to 0.824) and internal calibration results demonstrated that the model performed well. ConclusionsThis study identified the twelve most significant common risk factors for oral and esophageal cancers and developed a single prediction model that performs well for both types of cancer.

Development and validation of a practical score to predict 3-year distant metastatic free survival in nasopharyngeal carcinoma incorporating the number of lymph node regions.

The improvement in diagnosis and treatment for nasopharyngeal carcinoma (NPC) has shifted the pattern of failure toward distant metastasis. This study aimed to develop a simplified prognostic scoring model to predict distant metastatic free survival (DMFS) for NPC patients. Patients with non-metastatic NPC were identified from a retrospective cohort diagnosed between 2010 and 2018. Flexible parametric survival analysis was used to identify potential predictors for DMFS and establish a scoring model. The prognostic accuracy between the 8th AJCC system and the scoring model was compared using Harrell's C-index. Of the total 393 patients, the median follow-up time was 85 months. The 3-year DMFS rate was 83.3%. Gender, T-stage, pre-EBV (cut-off 2300 copies/ml), and the number of metastatic lymph node regions were identified as independent risk factors for distant metastasis and were included in the final scoring model. Our established model achieved a high C-index in predicting DMFS (0.79) and was well-calibrated. The score divided patients into two categories: low-risk (score 0-4) and high-risk (score 5-7), corresponding with the predicted 3-year DMFS of 96% and 64.5%, respectively. A feasible and applicative prognostic score was established and validated to discriminate NPC patients into low- and high-risk groups.

Development of a prognostic model for patients with nodular melanoma of the lower extremities: a study based on the SEER database.

Lower extremity nodular melanoma (NM) is a common malignant tumor with a poor prognosis. We aims to identify the prognostic factors and develop a nomogram model to predict overall survival (OS) in patients with lower extremity NM. A total of 746 patients with lower extremity NM were selected and randomly divided into a training set (522 cases) and a validation set (224 cases) from the Surveillance, Epidemiology, and End Results(SEER) database. The training set underwent univariate and multivariate Cox regression analyses to identify independent prognostic factors associated with patient outcomes, and to develop a nomogram model. The effectiveness of the nomogram was subsequently validated using the validation set. Multivariable Cox regression analysis of the training set indicated that age, ulceration, radiotherapy, chemotherapy, primary site of first malignant tumor, and Breslow thickness were independent variables associated with OS. In the training set, the area under the curve (AUC) of the nomogram for predicting 3-year and 5-year OS was 0.796 and 0.811, respectively. In the validation set, the AUC for predicting 3-year and 5-year OS was 0.694 and 0.702, respectively. The Harrell's C-index for the training set and validation set were 0.754 (95% CI: 0.721-0.787) and 0.670 (95% CI: 0.607-0.733), respectively. Calibration curves for both training and validation sets showed good agreement. In this study, we develop and validate a nomogram to predict OS in patients with lower extremity NM. The nomogram demonstrated reasonable reliability and clinical applicability. Nomograms are important tools assessing prognosis and aiding clinical decision-making.

Mandibular dose-volume predicts time-to-osteoradionecrosis in an actuarial normal-tissue complication probability (NTCP) model: External validation of right-censored clinico-dosimetric and competing risk application across international multi-institutional observational cohorts and online graphical user interface clinical support tool assessment.

Existing studies on osteoradionecrosis of the jaw (ORNJ) have primarily used cross-sectional data, assessing risk factors at a single time point. Determining the time-to-event profile of ORNJ has important implications to monitor oral health in head and neck cancer (HNC) long-term survivors. Demographic, clinical and dosimetric data were retrospectively obtained for a clinical observational cohort of 1129 patients with HNC treated with radiotherapy (RT) at The University of Texas MD Anderson Cancer Center. ORNJ was diagnosed in 198 patients (18%). A multivariable logistic regression analysis with forward stepwise variable selection identified significant predictors for ORNJ. These predictors were then used to train a Weibull Accelerated Failure Time (AFT) model, which was externally validated using an independent cohort of 265 patients (92 ORNJ cases and 173 controls) treated at Guy's and St. Thomas' Hospitals. Our model identified that each unit increase in D25% is significantly associated with a 12% shorter time to ORNJ (Adjusted Time Ratio [ATR] 0·88, p<0·005); pre-RT dental extractions was associated to a 27% faster (ATR 0·73, p=0·13) onset of ORNJ; male patients experienced a 38% shorter time to ORNJ (ATR 0·62, p = 0·11). The model demonstrated strong internal calibration (integrated Brier score of 0·133, D-calibration p-value 0.998) and optimal discrimination at 72 months (Harrell's C-index of 0·72). The model also showed good generalization to the independent cohort, despite a slight drop in performance. This study is the first to demonstrate a direct relationship between radiation dose and the time to ORNJ onset, providing a novel characterization of the impact of delivered dose not only on the probability of a late effect (ORNJ), but the conditional risk during survivorship. This work was supported by various funding sources including NIH, NIDCR, NCI, NAPT, NASA, BCM, Affirmed Pharma, CRUK, KWF Dutch Cancer Society, NWO ZonMw, and the Apache Corporation.