5-year AUC Research Articles

Development and Validation of Dynamic 5-Year Breast Cancer Risk Model Using Repeated Mammograms.

Current image-based long-term risk prediction models do not fully use previous screening mammogram images. Dynamic prediction models have not been investigated for use in routine care. We analyzed a prospective WashU clinic-based cohort of 10,099 cancer-free women at entry (between November 3, 2008 and February 2012). Follow-up through 2020 identified 478 pathology-confirmed breast cancers (BCs). The cohort included 27% Black women. An external validation cohort (Emory) included 18,360 women screened from 2013, followed through 2020. This included 42% Black women and 332 pathology-confirmed BC excluding those diagnosed within 6 months of screening. We trained a dynamic model using repeated screening mammograms at WashU to predict 5-year risk. This opportunistic screening service presented a range of mammogram images for each woman. We applied the model to the external validation data to evaluate discrimination performance (AUC) and calibrated to US SEER. Using 3 years of previous mammogram images available at the current screening visit, we obtained a 5-year AUC of 0.80 (95% CI, 0.78 to 0.83) in the external validation. This represents a significant improvement over the current visit mammogram AUC 0.74 (95% CI, 0.71 to 0.77; P < .01) in the same women. When calibrated, a risk ratio of 21.1 was observed comparing high (>4%) to very low (<0.3%) 5-year risk. The dynamic model classified 16% of the cohort as high risk among whom 61% of all BCs were diagnosed. The dynamic model performed comparably in Black and White women. Adding previous screening mammogram images improves 5-year BC risk prediction beyond static models. It can identify women at high risk who might benefit from supplemental screening or risk-reduction strategies.

Constructing and identifying an eighteen-gene tumor microenvironment prognostic model for non-small cell lung cancer

BackgroundThe tumor microenvironment (TME) plays a crucial role in tumorigenesis and tumor progression. This study aimed to identify novel TME-related biomarkers and develop a prognostic model for patients with non-small-cell lung cancer (NSCLC).MethodsAfter downloading and preprocessing data from The Cancer Genome Atlas (TCGA) data portal and Gene Expression Omnibus (GEO) datasets, we classified the molecular subtypes using the “NMF” R package. We performed survival analysis and quantified immune scores between clusters. A Cox proportional hazards model was then constructed, and its formula was produced. We assessed model performance and clinical utility. A prediction nomogram was also constructed and validated. Additionally, we explored the potential regulatory mechanisms of our TME gene signature using Gene Set Enrichment Analysis (GSEA).ResultsFrom data processing and univariate Cox regression analysis, 57 TME-related prognostic genes were identified, and two significantly distinct clusters were established. Using Cox regression and Lasso regression, an 18-gene TME-related prognostic model was developed. Patients were stratified into high- and low-risk groups based on the risk score, with survival analysis showing that the low-risk group had significantly better outcomes than the high-risk group (P < 0.01). ROC curve analysis demonstrated strong predictive performance, with 1-year, 3-year, and 5-year AUC values ranging from 0.654 to 0.702 across different cohorts. The model accurately predicted survival outcomes across subgroups with varying clinical features, and its predictive accuracy was validated through a nomogram.ConclusionsWe developed a prognostic model based on TME-related genes in NSCLC. Our 18-gene TME signature can effectively predict the prognosis of NSCLC with high accuracy.

Predicting Immune Flares in Untreated Chronic Hepatitis B Patients Using Novel Risk Factors and the FLARE-B Score.

Risk factors of chronic hepatitis B (CHB) immune flares are poorly understood. The primary aim of this study was to discover predictors of the CHB flare in non-cirrhotic, untreated CHB patients and develop a simple risk-stratifying score to predict the CHB flare. The secondary aim was to compare different machine learning methods for prediction. A retrospective cohort of untreated, non-cirrhotic CHB patients with normal baseline ALT was followed up over time until an immune flare as defined by ALT twice the upper limit of normal. Statistical learning and machine learning algorithms were used to develop predictive models using baseline variables. Bootstrap validation was used to internally validate the models. Of 405 patients (median age 44y; 41% male, 10% HBeAg positive), 67 (17%) experienced an immune flare by 5 years (annual incidence 4.0%). Predictors of flare included raised serum globulin, younger age, HBeAg positive status, higher viral load and raised liver stiffness. A simple predictive model "FLARE-B" had optimism-adjusted 1, 3 and 5-year AUCs of 0.813, 0.728 and 0.702, respectively. The random survival forest algorithm had the highest optimism-adjusted AUCs of 0.861, 0.766 and 0.725, respectively. New, novel predictors of the CHB flare include a raised serum globulin and possibly raised liver stiffness and the absence of liver steatosis. FLARE-B can be used to risk-stratify individuals and potentially guide personalized management strategies such as monitoring schedules and proactive antiviral treatment in high-risk patients.

Artificial intelligence-enhanced electrocardiography predicts 10-year risk of atherosclerotic cardiovascular disease

Abstract Background The ACC/AHA pooled cohort equations (PCE) calculates the 10-year primary risk of atherosclerotic cardiovascular disease (ASCVD), an indication to initiate primary prevention statin therapy in international guidelines. The validity of PCE is limited by its overestimation of risk in high-risk cohorts and an ethnic heterogeneity. Electrocardiography (ECG) has not been incorporated in current ASCVD risk estimation tools. Artificial Intelligence (AI)-based models are capable to capture patterns that are informative for cardiovascular prognosis. Purpose We explored the predictive value of a discrete-time survival model in 10-year ASCVD risk in comparison to PCE. Methods We developed an AI-ECG model to predict the 10-year ASCVD risk in an unselected secondary care population from the USA comprised of 1,163,401 ECGs from 189,539 patients. Data was split into training/validation and hold-out test sets by patient ID. Our AI-ECG model uses deep learning with a residual convolutional neural network with a discrete-time survival loss function to output subject-specific survival curves, accounting for both time to event and censorship (i.e., loss to follow up). We compared the performance of out AI-ECG model using a subset of outpatient data (n=4590) of the BIDMC dataset to PCE and ASCVD risk factors including systolic blood pressure, total cholesterol, HDL cholesterol, hypertension, smoking history, diabetes mellitus, and ethnicity. Performances were evaluated using cox survival analysis. Our AI-ECG model was externally validated in the UKB Biobank (UKB), where we compared the performance of our AI-ECG model to predictions of existing AI-ECG models such as the Stanford Estimator of Electrocardiogram Risk (SEER) model. Results Our AI-ECG successfully predicted future ASCVD events in individuals (C-index 0.721 (0.719-0.723), 5-year AUC 0.761 (0.758-0.763), and differentiated between high and low risk group with an adjusted HR of 2.33 (2.26-2.41). Upon internal validation our AI-ECG model showed a superior performance (C-index 0.679 (0.651-0.708)) than PCE (0.605 (0.577-0.634)) and all risk factors combined (C index 0.642 (0.613-0.672)). When combining AI-ECG predictions with all risk factors, the model showed additional predictive value (0.686 (0.657-0.715)) (Figure 1). Our AI-ECG model showed modest predictive value within the UKB dataset (C-index 0.655 (0.637-0.673)), but significantly outperformed the SEER model (C-index 0.547 (0.527-0.567), p &amp;lt;0.0001 for comparison with AIRE). Conclusion We have shown the potential of AI-ECG models in predicting future ASCVD risks. AI-ECG models demonstrated accurate discrete time survival prediction and good classification performance superior to that of the SEER model, as well as PCE and ASCVD risk factors. . Accurate risk assessment of ASCVD may help prevent adverse events in high risk subjects and save unnecessary pharmacological interventios in low risk subjects.Figure 1

The INFLUENCE 3.0 model: Updated predictions of locoregional recurrence and contralateral breast cancer, now also suitable for patients treated with neoadjuvant systemic therapy

BackgroundIndividual risk prediction of 5-year locoregional recurrence (LRR) and contralateral breast cancer (CBC) supports decisions regarding personalised surveillance. The previously developed INFLUENCE tool was rebuild, including a recent population and patients who received neoadjuvant systemic therapy (NST). MethodsWomen, surgically treated for nonmetastatic breast cancer, diagnosed between 2012 and 2016, were selected from the Netherlands Cancer Registry. Cox regression with restricted cubic splines was compared to Random Survival Forest (RSF) to predict five-year LRR and CBC risks. Separate models were developed for NST patients. Discrimination and calibration were assessed by 100x bootstrap resampling. ResultsIn the non-NST and NST group, 49,631 and 10,154 patients were included, respectively. Age, mode of detection, histology, sublocalisation, grade, pT, pN, hormonal receptor status ± endocrine treatment, HER2 status ± targeted treatment, surgery ± immediate reconstruction ± radiation therapy, and chemotherapy were significant predictors for LRR and/or CBC in non-NST patients. For NST patients this was similar, but excluding (y)pT and (y)pN status, and including presence of ductal carcinoma in situ, axillary lymph node dissection and pathologic complete response.For non-NST patients, the Cox and RSF models were integrated in the online tool with 5-year AUCs of 0.77 (95%CI:0.77–0.77) and 0.68 (95%CI:0.67–0.68)] for LRR and CBC prediction, respectively. For NST patients, the RSF model performed best (AUCs 0.77 (95%CI:0.76–0.78) and 0.73 (95%CI:0.69–0.76) for LRR and CBC, respectively). Regarding calibration, observed-predicted differences were all <1 %. ConclusionThis INFLUENCE 3.0 models showed moderate performance in LRR and CBC prediction. The models have been made available as online tool to enable clinical decision support regarding personalised follow-up.

Integrative prognostic modeling for stage III lung adenosquamous carcinoma post-tumor resection: machine learning insights and web-based implementation.

The prognostic landscape of stage III Lung Adenosquamous Carcinoma (ASC) following primary tumor resection remains underexplored. A thoughtfully developed prognostic model has the potential to guide clinicians in patient counseling and the formulation of effective therapeutic strategies. Utilizing data from the Surveillance, Epidemiology, and End Results database spanning 2000 to 2018, this study identified independent prognostic factors influencing Overall Survival (OS) in ASC using Boruta analysis. Employing Gradient Boosting, Random Forest, and Neural Network algorithms, predictive models were constructed. Model performance was assessed through key metrics, including Area Under the Receiver Operating Characteristic Curve (AUC), calibration plot, Brier score, and Decision Curve Analysis (DCA). Among 241 eligible patients, seven clinical parameters-age, sex, primary tumor size, N stage, primary tumor site, chemotherapy, and systemic therapy-were identified as significant predictors of OS. Advanced age, male gender, larger tumor size, absence of chemotherapy, and lack of systemic therapy were associated with poorer survival. The Random Forest model outperformed others, achieving 3- and 5-year AUCs of 0.80/0.79 (training) and 0.74/0.65 (validation). It also demonstrated better calibration, lower Brier scores (training: 0.189/0.171; validation: 0.207/0.199), and more favorable DCA. SHAP values enhanced model interpretability by highlighting the impact of each parameter on survival predictions. To facilitate clinical application, the Random Forest model was deployed on a web-based server for accessible prognostic assessments. This study presents a robust machine learning model and a web-based tool that assist healthcare practitioners in personalized clinical decision-making and treatment optimization for ASC patients following primary tumor resection.

Long-term outcomes in rapamycin on renal allograft function: a 30-year follow-up from a single-center experience

ObjectivesTo evaluate long-term renal graft prognosis and the role of rapamycin from a single-center in China over a 30-year follow-up.MethodsThis study enrolled a total of 654 patients who underwent kidney transplantation between 1989 and 2020. The basic characteristics of the included patients were collected. Graft survival was described and compared using Kaplan-Meier curves (K-M curves). Both continuous and categorical variables were included in a multivariate Cox proportional-hazards model. Patients were divided into rapamycin-based quadruple immunosuppression regimen group (rapa group, n = 41) and conventional tacrolimus-based triple immunosuppression regimen group (control group, n = 218). The indication biopsy results of the two groups were further reviewed to compare the incidence of rejection, acute rejection, and banff score.ResultsThe overall 5, 10, 15, 20-year graft survival rate of our center is 87.5%, 62.4%, 46.4% and 20.9%, respectively. The median survival time after surgery is 14 years. Multiple Cox regression analysis identified BMI (p = 0.035), dialysis type (p < 0.001), immunosuppressants (p < 0.01), urine albumen (p < 0.001), globulin (p = 0.041), and blood glucose (p = 0.002) as risk factors. The 20-year, 10-year and 5-year AUC is 0.78, 0.75 and 0.75. The combination of FK506 and rapamycin was further suggested by the model to effectively improve the graft prognosis (p < 0.01, HR = 0.763). The K-M curve showed that the long-term survival rate of renal grafts in the rapa group was significantly better than that in the conventional group (p < 0.001). In addition, indication biopsy records revealed a lower possibility of immune rejection in the rapa group than that in the conventional group (p < 0.001). Banff score indicated that rapa group had less vascular inflammation in the transplanted kidney.ConclusionsIn this study, a 30-year follow-up was performed in a single center, and a total graft 20-year survival rate of 20.9% was reported. The prognostic model and subgroup analysis suggested that FK506 combined with rapamycin could effectively improve the prognosis of renal transplantation, which could be explained by reduced acute rejection and less vascular inflammation.

Machine Learning-Based Predictive Model for Mortality in Female Breast Cancer Patients Considering Lifestyle Factors.

To construct a free and accurate breast cancer mortality prediction tool by incorporating lifestyle factors, aiming to assist healthcare professionals in making informed decisions. In this retrospective study, we utilized a ten-year follow-up dataset of female breast cancer patients from a major Chinese hospital and included 1,390 female breast cancer patients with a 7% (96) mortality rate. We employed six machine learning algorithms (ridge regression, k-nearest neighbors, neural network, random forest, support vector machine, and extreme gradient boosting) to construct a mortality prediction model for breast cancer. This model incorporated significant lifestyle factors, such as postsurgery sexual activity, use of totally implantable venous access ports, and prosthetic breast wear, which were identified as independent protective factors. Meanwhile, ten-fold cross-validation demonstrated the superiority of the random forest model (average AUC = 0.918; 1-year AUC = 0.914, 2-year AUC = 0.867, 3-year AUC = 0.883). External validation further supported the model's robustness (average AUC = 0.782; 1-year AUC = 0.809, 2-year AUC = 0.785, 3-year AUC = 0.893). Additionally, a free and user-friendly web tool was developed using the Shiny framework to facilitate easy access to the model. Our breast cancer mortality prediction model is free and accurate, providing healthcare professionals with valuable information to support their clinical decisions and potentially promoting healthier lifestyles for breast cancer patients.

Identification of novel M2 macrophage-related molecule ATP6V1E1 and its biological role in hepatocellular carcinoma based on machine learning algorithms.

Hepatocellular carcinoma (HCC) remains the most prevalent form of primary liver cancer, characterized by late detection and suboptimal response to current therapies. The tumour microenvironment, especially the role of M2 macrophages, is pivotal in the progression and prognosis of HCC. We applied the machine learning algorithm-CIBERSORT, to quantify cellular compositions within the HCC TME, focusing on M2 macrophages. Gene expression profiles were analysed to identify key molecules, with ATP6V1E1 as a primary focus. We employed Gene Set Enrichment Analysis (GSEA) and Kaplan-Meier survival analysis to investigate the molecular pathways and prognostic significance of ATP6V1E1. A prognostic model was developed using multivariate Cox regression analysis based on ATP6V1E1-related molecules, and functional impacts were assessed through cell proliferation assays. M2 macrophages were the dominant cell type in the HCC TME, significantly correlating with adverse survival outcomes. ATP6V1E1 was robustly associated with advanced disease stages and poor prognostic features such as vascular invasion and elevated alpha-fetoprotein levels. GSEA linked high ATP6V1E1 expression to critical oncogenic pathways, including immunosuppression and angiogenesis, and reduced activity in metabolic processes like bile acid and fatty acid metabolism. The prognostic model stratified HCC patients into distinct risk categories, showing high predictive accuracy (1-year AUC = 0.775, 3-year AUC = 0.709 and 5-year AUC = 0.791). Invitro assays demonstrated that ATP6V1E1 knockdown markedly inhibited the proliferation of HCC cells. The study underscores the significance of M2 macrophages and ATP6V1E1 in HCC, highlighting their potential as therapeutic and prognostic targets.

A novel nomogram based on inflammatory-nutritional biomarkers for gallbladder cancer after surgical resection

PurposeSystemic inflammation and nutrition are vital for tumor progression. This study aimed to identify prognostic inflammation nutrition markers and develop a predictive nomogram for gallbladder cancer (GBC).MethodsA total of 123 patients with GBC who underwent surgical resection at the First Affiliated Hospital of Soochow University and Suzhou Kowloon Hospital were included in our study. The final prognostic variables were identified using univariate and multivariate analyses. A nomogram model was then established, and the consistency index (C-index), calibration curves, and Kaplan–Meier analysis were performed to evaluate the accuracy and discrimination of the nomogram. The area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA) suggested that our nomogram had better predictive ability and clinical feasibility than a published model.ResultsThe cox regression analysis showed that carcinoembryonic antigen (CEA) > 4.580, albumin-bilirubin (ALBI) > –2.091, geriatric nutritional risk index (GNRI) < 90.83, T3–T4, and N2 are independent prognostic factors. A predictive nomogram was constructed with a C-index of 0.793. In the calibration curves, the nomogram-predicted 1-, 3-, and 5-year survival matched well with the actual survival. Kaplan–Meier analysis showed that the high-risk group had worse survival than the low-risk group (P < 0.001). Finally, our nomogram achieved better 1-, 3- and 5-year AUCs than an established model (0.871, 0.844, and 0.781 vs. 0.753, 0.750, and 0.693). DCA also confirmed that our model outperformed the established model.ConclusionsIn conclusion, our study revealed that CEA > 4.580, GNRI < 90.83, ALBI > –2.091, T3–T4 stage, and N2 were related to clinical outcomes of patients with GBC after surgical resection. The constructed nomogram has superior predictive ability and clinical practicality.

Recurrence of Hepatocellular Carcinoma in Patients with Low Albumin-Bilirubin Grade in TACE Combined with Ablation: A Random Forest Cox Predictive Model.

The aim of our study was to investigate the relationship between albumin-bilirubin (ALBI) grade and recurrence in patients who underwent TACE sequential ablation. We developed and validated a nomogram to predict low levels of ALBI patients' recurrence. A total of 880 patients undergoing TACE combined ablation at Beijing Youan Hospital from January 2014 to December 2021 were retrospectively enrolled, including 415 patients with L-ALBI (≤-2.6) and 465 patients with high levels (>-2.6) of ALBI (H-ALBI). L-ALBI patients were randomized in a 7:3 ratio into the training cohort (N=289) and validation cohort (N=126). Multivariate Cox regression followed by random survival forest was carried out to identify independent risk factors for prediction nomogram construction. An examination of nomogram accuracy was performed using the C-index, receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA) curves. According to the nomogram, the patients were divided into low-risk, intermediate-risk, and high-risk groups. Kaplan-Meier (KM) curves were applied to compare the difference in recurrence-free survival (RFS) among the three groups. The median RFS in L-ALBI patients was significantly longer than the H-ALBI patients (40.8m vs 20.1m, HR:1.71, 95% CI:1.44-2.04, P<0.0001). The nomogram was composed of five variables, such as age, Barcelona Clinic Liver Cancer (BCLC) stage, globulin, gamma-glutamyl transferase to lymphocyte ratio (GLR), and international normalized ratio (INR). The C-index (0.722 and 0.731) and 1-, 3-, and 5-year AUCs (0.725, 0.803, 0.870, and 0.764, 0.816, 0.798) of the training and validation cohorts proved the good predictive performance of the nomogram. Calibration curves and DCA curves demonstrated good consistency and good clinical utility. There were significant differences in RFS between the low-risk, intermediate-risk, and high-risk groups (P<0.0001). L-ALBI Patients who underwent TACE combined ablation had better recurrence-free survival than patients with H-ALBI. The nomogram developed and validated in our study had good predictive ability in recurrence for L-ALBI patients.

Integrative nomogram model based on anoikis-related genes enhances prognostic evaluation in colorectal cancer

BackgroundRevealing the role of anoikis resistance plays in CRC is significant for CRC diagnosis and treatment. This study integrated the CRC anoikis-related key genes (CRC-AKGs) and established a novel model for improving the efficiency and accuracy of the prognostic evaluation of CRC. MethodsCRC-ARGs were screened out by performing differential expression and univariate Cox analysis. CRC-AKGs were obtained through the LASSO machine learning algorithm and the LASSO Risk-Score was constructed to build a nomogram clinical prediction model combined with the clinical predictors. In parallel, this work developed a web-based dynamic nomogram to facilitate the generalization and practical application of our model. ResultsWe identified 10 CRC-AKGs and a risk-related prognostic Risk-Score was calculated. Multivariate COX regression analysis indicated that the Risk-Score, TNM stage, and age were independent risk factors that significantly associated with the CRC prognosis(p < 0.05). A prognostic model was built to predict the outcome with satisfied accuracy (3-year AUC = 0.815) for CRC individuals. The web interactive nomogram (https://yuexiaozhang.shinyapps.io/anoikisCRC/) showed strong generalizability of our model. In parallel, a substantial correlation between tumor microenvironment and Risk-Score was discovered in the present work. ConclusionThis study reveals the potential role of anoikis in CRC and sets new insights into clinical decision-making in colorectal cancer based on both clinical and sequencing data. Also, the interactive tool provides researchers with a user-friendly interface to input relevant clinical variables and obtain personalized risk predictions or prognostic assessments based on our established model.

Serum lipid ratios as novel prognostic biomarkers for patients with locally advanced breast cancer treated with neoadjuvant therapy

ABSTRACT Objectives To investigate the association between lipid ratios and survival outcomes in patients with locally advanced breast cancer (LABC) undergoing neoadjuvant chemotherapy. Method This retrospective study included patients with LABC receiving neoadjuvant chemotherapy. Serum lipid levels were prospectively measured at baseline. Associations of triglyceride to total cholesterol (TG/TC), triglyceride to high-density lipoprotein (TG/HDL) and triglyceride to low-density lipoprotein (TG/LDL) ratios with prognosis were evaluated. Results Patients with high TG/TC (adjusted hazard ratio [aHR] = 2.47, 95% CI: 1.10, 5.56, p = 0.029), TG/HDL (aHR = 2.73, 95% CI: 1.16, 6.41, p = 0.021) and TG/LDL (aHR = 2.50, 95% CI: 1.11, 5.65, p = 0.027) ratios were more likely to experience disease-free survival (DFS) events. Subgroup analysis suggested that the prognostic impact of lipid ratios was more pronounced in patients with negative HER2 status or those at a high risk of recurrence (e.g. clinical stage III, Ki67 > 30%). Additionally, higher lipid ratios tended to indicate early DFS events (0 ~ 2 years) (TG/TC p = 0.021, TG/HDL p = 0.046, TG/LDL p < 0.001), and the TG/LDL ratio demonstrated the best predictive efficacy (TG/TC vs. TG/HDL vs. TG/LDL, 1-year AUC: 0.724 vs. 0.676 vs. 0.846, 2-year AUC: 0.653 vs. 0.638 vs. 0.708). Conclusion Baseline serum TG/TC, TG/HDL and TG/LDL ratios were independent prognostic factors in patients with LABC undergoing neoadjuvant therapy. However, their utility in predicting the early DFS events warrants further investigation. Clinical trial registration NCT05621564.

Integrative analyses of multi-omics data constructing tumor microenvironment and immune-related molecular prognosis model in human colorectal cancer

The increasing prevalence and incidence of colorectal cancer (CRC), particularly in young adults, underscore the imperative to comprehend its fundamental mechanisms, discover novel diagnostic and prognostic markers, and enhance therapeutic strategies. Here, we integrated multi-omics data, including gene expression, somatic mutation data and DNA methylation data, to unravel the intricacies of tumor microenvironment (TME) in CRC and search for novel prognostic markers. By calculating the immune score for each patient from the expression profile, we delineated the differential immune cell fraction, constructed an immune-related multi-omics atlas, and identified molecular characteristics. The entire colorectal dataset (n = 343) was randomly divided into training (n = 249) and testing datasets (n = 94). We screened 144 immune-related genes, 6 mutant genes, and 38 methylation probes associated with overall survival (OS). These makers were then incorporated into a 10-gene prognostic model using Lasso and Cox regression in the training dataset, and the model's performance was evaluated in an independent validation dataset. The model exhibited satisfactory results (average concordance index [C-index] = 0.77), with the average 1-year, 3-year, and 5-year AUCs being 0.79, 0.76, and 0.76 in the training dataset and 0.74, 0.80, and 0.90 in the testing dataset. Furthermore, the prognostic model demonstrated applicability in guiding chemotherapy for CRC patients and exhibited a degree of pan-cancer utility in risk stratification. In conclusion, our integrated analysis of multi-omics data revealed immune-related genetic and epigenetic characteristics of the TME. We propose an integrative prognostic model that can stratify risk and guide chemotherapy for CRC patients. The generalizability of the model in risk stratification across different cancer types was validated in Pan-Cancer cohort.

Evaluating generalizability of landmark randomized controlled trials in common metastatic cancers using machine learning-based emulated trials.

e13618 Background: Randomized controlled trial (RCT) results often lack generalizability to real-world oncology patients (pts). We present TrialTranslator, a machine learning (ML) framework to evaluate the generalizability of Phase III oncology RCTs by emulating trials across prognostic risk groups. Methods: TrialTranslator was applied to 11 landmark RCTs encompassing the most prevalent advanced cancers – non-small cell lung (aNSCLC), breast (mBC), prostate (mPC), and colorectal (mCRC). Flatiron Health’s EHR-derived de-identified database was used for model development and trial emulation. In Stage I, ML models (gradient boosted [GBM], random forest, support vector machine and penalized Cox) were trained to predict survival from time of metastatic diagnosis. Models used 130 longitudinal demographic, cancer-related, lab, and biomarker features. In Stage II, trial emulation, pts meeting RCT cancer and biomarker eligibility criteria were stratified into low, medium, and high-risk groups based on the top-performing ML model. We used inverse probability of treatment weighted survival analysis for each risk group to compare the median overall survival (mOS) or median progression free survival (mPFS) between experimental and control arms. Results: GBM models achieved the highest time-dependent area under the receiver operating curve (AUC) in the test set in all cancer types. For example, the 1-year AUC for the aNSCLC GBM vs the Lung Cancer Prognostic Index (validated Cox model) was 0.783 vs 0.689. ML-predicted low-risk pts typically had similar mOS or mPFS as RCT populations (Table). In emulated trials for low-risk pts, real-world survival benefit – the absolute difference in mOS or mPFS between experimental and control arms – was similar to RCT results. In contrast, for high-risk pts, survival benefits were not statistically or clinically significant. For example, in the FLAURA trial, RCT mPFS for experimental arm (osimertinib) vs control (gefitinib or erlotinib) was 18.9 mos vs 10.2 months; mPFS was 17.1 vs 10.3 months for low-risk pts and 5.1 vs 4.2 months for high-risk pts. Conclusions: ML frameworks like TrialTranslator may facilitate individual patient-level estimation of real-world treatment benefits of novel oncology agents. TrialTranslator is accessible at trialtranslator.com. [Table: see text]

New joint model for the dynamic prediction of heart failure in phospholamban (PLN) p.(Arg14del) positive individuals

Abstract Background PLN p.(Arg14del) positive individuals are at risk for developing arrhythmogenic cardiomyopathy which can cause severe heart failure. Future gene therapy could be lifesaving and is rapidly advancing. It is therefore of importance to identify eligible patients for this evolving treatment, with individuals risk prediction playing an imperative role in this process. Purpose To develop a dynamic prediction model for the individual risk prediction of heart failure in PLN p.(Arg14del) positive individuals. Method Data were collected of 564 PLN p.(Arg14del) positive individuals with no baseline history of heart failure or myocardial infarction. During a median follow-up of 6.3 years (interquartile range 2.9-10.0), 74 individuals experienced heart failure, defined as a composite endpoint of heart failure hospitalization, left ventricle assist device implantation, heart transplantation and heart failure-related death. For the individual prediction of time to heart failure we used a joint model which simultaneously estimates a longitudinal model for repeatedly measured left ventricle ejection fraction (LVEF) during follow-up and a Cox regression model for the hazards of the heart failure endpoint. Using the longitudinal model we predicted the LVEF-values of all patients at future time points using the estimated random-effects and these predicted LVEF values were used as a time-dependent covariate in the Cox model which also contained age at start of follow-up as predictor variable. Results The joint model trained in this study had a very good performance in discriminating the individuals at risk for heart failure, with the 5-year mean AUC of 0.86 and 10-year mean AUC of 0.84. The LVEF had a hazard ratio of 0.89 (95% confidence interval (CI) 0.86-0.93, p&amp;lt;0.001), and age at baseline 1.04 (95% CI 1.03-1.06], p&amp;lt;0.001), meaning a lower LVEF and higher age at presentation were associated with heart failure. Figure 1 shows an example of an individual risk prediction with on the x-axis the follow-up time in years, on the y-axis (left) the LVEF in % and on the y-axis (right) the predicted survival probability. Conclusion Individual risk prediction with a simple model including only LVEF and age is of great importance for identifying PLN p.(Arg14del) positive individuals who are at risk of developing heart failure. The introduction of this novel joint prediction model is the first step in identifying individuals at risk for heart failure and may aid patient selection for future gene therapy interventions.

The Effect of Low HBV-DNA Viral Load on Recurrence in Hepatocellular Carcinoma Patients Who Underwent Primary Locoregional Treatment and the Development of a Nomogram Prediction Model.

(1) Background: HBV-DNA is an essential clinical indicator of primary hepatocellular carcinoma (HCC) prognosis. Our study aimed to investigate the prognostic implication of a low load of HBV-DNA in HCC patients who underwent local treatment. Additionally, we developed and validated a nomogram to predict the recurrence of patients with low (20-100 IU/mL) viral loads (L-VL). (2) Methods: A total of 475 HBV-HCC patients were enrolled, including 403 L-VL patients and 72 patients with very low (<20 IU/mL) viral loads (VL-VL). L-VL HCC patients were randomly divided into a training set (N = 282) and a validation set (N = 121) at a ratio of 7:3. Utilizing the Lasso-Cox regression analysis, we identified independent risk factors for constructing a nomogram. (3) Results: L-VL patients had significantly shorter RFS than VL-VL patients (38.2 m vs. 23.4 m, p = 0.024). The content of the nomogram included gender, BCLC stage, Glob, and MLR. The C-index (0.682 vs. 0.609); 1-, 3-, and 5-year AUCs (0.729, 0.784, and 0.783, vs. 0.631, 0.634, the 0.665); calibration curves; and decision curve analysis (DCA) curves of the training and validation cohorts proved the excellent predictive performance of the nomogram. There was a statistically significant difference in RFS between the low-, immediate-, and high-risk groups both in the training and validation cohorts (p < 0.001); (4) Conclusions: Patients with L-VL had a worse prognosis. The nomogram developed and validated in this study has the advantage of predicting patients with L-VL.

Development of a Prognostic Risk Model Based on Oxidative StressRelated Genes for Platinum-Resistant Ovarian Cancer Patients.

Ovarian Cancer (OC) is a heterogeneous malignancy with poor outcomes. Oxidative stress plays a crucial role in developing drug resistance. However, the relationships between Oxidative Stress-related Genes (OSRGs) and the prognosis of platinum-resistant OC remain unclear. This study aimed to develop an OSRGs-based prognostic risk model for platinum-resistant OC patients. Gene Set Enrichment Analysis (GSEA) was performed to determine the expression difference of OSRGs between platinum-resistant and -sensitive OC patients. Cox regression analyses were used to identify the prognostic OSRGs and establish a risk score model. The model was validated by using an external dataset. Machine learning was used to determine the prognostic OSRGs associated with platinum resistance. Finally, the biological functions of selected OSRG were determined via in vitro cellular experiments. Three gene sets associated with oxidative stress-related pathways were enriched (p < 0.05), and 105 OSRGs were found to be differentially expressed between platinum-resistant and - sensitive OC (p < 0.05). Twenty prognosis-associated OSRGs were identified (HR: 0:562-5.437; 95% CI: 0.319-20.148; p < 0.005), and seven independent OSRGs were used to construct a prognostic risk score model, which accurately predicted the survival of OC patients (1-, 3-, and 5-year AUC=0.69, 0.75, and 0.67, respectively). The prognostic potential of this model was confirmed in the validation cohort. Machine learning showed five prognostic OSRGs (SPHK1, PXDNL, C1QA, WRN, and SETX) to be strongly correlated with platinum resistance in OC patients. Cellular experiments showed that WRN significantly promoted the malignancy and platinum resistance of OC cells. The OSRGs-based risk score model can efficiently predict the prognosis and platinum resistance of OC patients. This model may improve the risk stratification of OC patients in the clinic.

Establishing subdivisions of M1 stage nasopharyngeal carcinoma based on decision tree classification: A multicenter retrospective study

ObjectivesTo meet the demand for personalized treatment, effective stratification of patients with metastatic nasopharyngeal carcinoma (mNPC) is essential. Hence, our study aimed to establish an M1 subdivision for prognostic prediction and treatment planning in patients with mNPC. Materials and methodsThis study included 1239 patients with mNPC from three medical centers divided into the synchronous mNPC cohort (smNPC, n = 556) to establish an M1 stage subdivision and the metachronous mNPC cohort (mmNPC, n = 683) to validate this subdivision. The primary endpoint was overall survival. Univariate and multivariate Cox analyses identified covariates for the decision-tree model, proposing an M1 subdivision. Model performance was evaluated using time-dependent receiver operating characteristic curves, Harrell’s concordance index, calibration plots, and decision curve analyses. ResultsThe proposed M1 subdivisions were M1a (≤5 metastatic lesions), M1b (>5 metastatic lesions + absent liver metastases), and M1c (>5 metastatic lesions + existing liver metastases) with median OS of 34, 22, and 13 months, respectively (p < 0.001). This M1 subdivision demonstrated superior discrimination (C-index = 0.698; 3-year AUC = 0.707) and clinical utility over those of existing staging systems. Calibration curves exhibited satisfactory agreement between predictions and actual observations. Internal and mmNPC cohort validation confirmed the robustness. Survival benefits from local metastatic treatment were observed in M1a, while immunotherapy improved survival in patients with M1b and M1c disease. ConclusionThis novel M1 staging strategy provides a refined approach for prognostic prediction and treatment planning in patients with mNPC, emphasizing the potential benefits of local and immunotherapeutic interventions based on individualized risk stratification.

Development and validation of a nomogram for predicting specific mortality risk: A study of competing risk model based on real endometrial cancer patients.

This study aimed to construct a competing risk prediction model for predicting specific mortality risks in endometrial cancer patients from the SEER database based on their demographic characteristics and tumor information. We collected relevant clinical data on patients with histologically confirmed endometrial cancer in the SEER database between 2010 and 2015. Univariate and multivariate competing risk models were used to analyze the risk factors for endometrial cancer-specific death, and a predictive nomogram was constructed. C-index and receiver operating characteristic curve (ROC) at different time points were used to verify the accuracy of the constructed nomogram. There were 26 109 eligible endometrial cancer patients in the training cohort and 11 189 in the validation cohort. Univariate and multivariate analyses revealed that Age, Marriage, Grade, Behav, FIGO, Size, Surgery, SurgOth, Radiation, ParaAortic_Nodes, Peritonea, N positive, DX_liver, and DX_lung were independent prognostic factors for specific mortality in endometrial cancer patients. Based on these factors, a nomogram was constructed. Internal validation showed that the nomogram had a good discriminative ability (C-index = 0.883 [95% confidence interval [CI]: 0.881-0.884]), and the 1-, 3-, and 5-year AUC values were 0.901, 0.886 and 0.874, respectively. External validation indicated similar results (C-index = 0.883 [95%CI: 0.882-0.883]), and the 1-, 3-, and 5- AUC values were 0.908, 0.885 and 0.870, respectively. We constructed a competing risk model to predict the specific mortality risk among endometrial cancer patients. This model has favorable accuracy and reliability and can provide a reference for the development and update of endometrial cancer prognostic risk assessment tools.