RESUMO
For a decade, The Cancer Genome Atlas (TCGA) program collected clinicopathologic annotation data along with multi-platform molecular profiles of more than 11,000 human tumors across 33 different cancer types. TCGA clinical data contain key features representing the democratized nature of the data collection process. To ensure proper use of this large clinical dataset associated with genomic features, we developed a standardized dataset named the TCGA Pan-Cancer Clinical Data Resource (TCGA-CDR), which includes four major clinical outcome endpoints. In addition to detailing major challenges and statistical limitations encountered during the effort of integrating the acquired clinical data, we present a summary that includes endpoint usage recommendations for each cancer type. These TCGA-CDR findings appear to be consistent with cancer genomics studies independent of the TCGA effort and provide opportunities for investigating cancer biology using clinical correlates at an unprecedented scale.
Assuntos
Neoplasias/patologia , Bases de Dados Genéticas , Genômica , Humanos , Estimativa de Kaplan-Meier , Neoplasias/genética , Neoplasias/mortalidade , Modelos de Riscos ProporcionaisRESUMO
BACKGROUND: Delays in hospital presentation limit access to acute stroke treatments. While prior research has focused on patient-level factors, broader ecological and social determinants have not been well studied. We aimed to create a geospatial map of prehospital delay and examine the role of community-level social vulnerability. METHODS: We studied patients with ischemic stroke who arrived by emergency medical services in 2015 to 2017 from the American Heart Association Get With The Guidelines-Stroke registry. The primary outcome was time to hospital arrival after stroke (in minutes), beginning at last known well in most cases. Using Geographic Information System mapping, we displayed the geography of delay. We then used Cox proportional hazard models to study the relationship between community-level factors and arrival time (adjusted hazard ratios [aHR] <1.0 indicate delay). The primary exposure was the social vulnerability index (SVI), a metric of social vulnerability for every ZIP Code Tabulation Area ranging from 0.0 to 1.0. RESULTS: Of 750â 336 patients, 149â 145 met inclusion criteria. The mean age was 73 years, and 51% were female. The median time to hospital arrival was 140 minutes (Q1: 60 minutes, Q3: 458 minutes). The geospatial map revealed that many zones of delay overlapped with socially vulnerable areas (https://harvard-cga.maps.arcgis.com/apps/webappviewer/index.html?id=08f6e885c71b457f83cefc71013bcaa7). Cox models (aHR, 95% CI) confirmed that higher SVI, including quartiles 3 (aHR, 0.96 [95% CI, 0.93-0.98]) and 4 (aHR, 0.93 [95% CI, 0.91-0.95]), was associated with delay. Patients from SVI quartile 4 neighborhoods arrived 15.6 minutes [15-16.2] slower than patients from SVI quartile 1. Specific SVI themes associated with delay were a community's socioeconomic status (aHR, 0.80 [95% CI, 0.74-0.85]) and housing type and transportation (aHR, 0.89 [95% CI, 0.84-0.94]). CONCLUSIONS: This map of acute stroke presentation times shows areas with a high incidence of delay. Increased social vulnerability characterizes these areas. Such places should be systematically targeted to improve population-level stroke presentation times.
Assuntos
Serviços Médicos de Emergência , Sistema de Registros , Tempo para o Tratamento , Humanos , Feminino , Masculino , Idoso , Idoso de 80 Anos ou mais , Pessoa de Meia-Idade , Acidente Vascular Cerebral/terapia , Acidente Vascular Cerebral/epidemiologia , AVC Isquêmico/terapia , AVC Isquêmico/epidemiologia , Estados Unidos/epidemiologiaRESUMO
Epidemiologic studies frequently use risk ratios to quantify associations between exposures and binary outcomes. When the data are physically stored at multiple data partners, it can be challenging to perform individual-level analysis if data cannot be pooled centrally due to privacy constraints. Existing methods either require multiple file transfers between each data partner and an analysis center (e.g., distributed regression) or only provide approximate estimation of the risk ratio (e.g., meta-analysis). Here we develop a practical method that requires a single transfer of eight summary-level quantities from each data partner. Our approach leverages an existing risk-set method and software originally developed for Cox regression. Sharing only summary-level information, the proposed method provides risk ratio estimates and confidence intervals identical to those that would be provided - if individual-level data were pooled - by the modified Poisson regression. We justify the method theoretically, confirm its performance using simulated data, and implement it in a distributed analysis of COVID-19 data from the U.S. Food and Drug Administration's Sentinel System.
RESUMO
BACKGROUND: The objective of this study was to investigate the role of clinical factors together with FOXO1 fusion status in patients with nonmetastatic rhabdomyosarcoma (RMS) to develop a predictive model for event-free survival and provide a rationale for risk stratification in future trials. METHODS: The authors used data from patients enrolled in the European Pediatric Soft Tissue Sarcoma Study Group (EpSSG) RMS 2005 study (EpSSG RMS 2005; EudraCT number 2005-000217-35). The following baseline variables were considered for the multivariable model: age at diagnosis, sex, histology, primary tumor site, Intergroup Rhabdomyosarcoma Studies group, tumor size, nodal status, and FOXO1 fusion status. Main effects and significant second-order interactions of candidate predictors were included in a multiple Cox proportional hazards regression model. A nomogram was generated for predicting 5-year event-free survival (EFS) probabilities. RESULTS: The EFS and overall survival rates at 5 years were 70.9% (95% confidence interval, 68.6%-73.1%) and 81.0% (95% confidence interval, 78.9%-82.8%), respectively. The multivariable model retained five prognostic factors, including age at diagnosis interacting with tumor size, tumor primary site, Intergroup Rhabdomyosarcoma Studies clinical group, and FOXO1 fusion status. Based on each patient's total score in the nomogram, patients were stratified into four groups. The 5-year EFS rates were 94.1%, 78.4%, 65.2%, and 52.1% in the low-risk, intermediate-risk, high-risk, and very-high-risk groups, respectively, and the corresponding 5-year overall survival rates were 97.2%, 91.5%, 74.3%, and 60.8%, respectively. CONCLUSIONS: The results presented here provide the rationale to modify the EpSSG stratification, with the most significant change represented by the replacement of histology with fusion status. This classification was adopted in the new international trial launched by the EpSSG.
Assuntos
Nomogramas , Rabdomiossarcoma , Humanos , Rabdomiossarcoma/mortalidade , Rabdomiossarcoma/patologia , Rabdomiossarcoma/terapia , Masculino , Feminino , Pré-Escolar , Criança , Prognóstico , Lactente , Medição de Risco , Adolescente , Europa (Continente)/epidemiologia , Proteína Forkhead Box O1/genética , Proteína Forkhead Box O1/metabolismo , Proteínas de Fusão Oncogênica/genéticaRESUMO
BACKGROUND: Severe heart failure (HF) has a higher mortality during vulnerable period while targeted predictive tools, especially based on drug exposures, to accurately assess its prognoses remain largely unexplored. Therefore, this study aimed to utilize drug information as the main predictor to develop and validate survival models for severe HF patients during this period. METHODS: We extracted severe HF patients from the MIMIC-IV database (as training and internal validation cohorts) as well as from the MIMIC-III database and local hospital (as external validation cohorts). Three algorithms, including Cox proportional hazards model (CoxPH), random survival forest (RSF), and deep learning survival prediction (DeepSurv), were applied to incorporate the parameters (partial hospitalization information and exposure durations of drugs) for constructing survival prediction models. The model performance was assessed mainly using area under the receiver operator characteristic curve (AUC), brier score (BS), and decision curve analysis (DCA). The model interpretability was determined by the permutation importance and Shapley additive explanations values. RESULTS: A total of 11,590 patients were included in this study. Among the 3 models, the CoxPH model ultimately included 10 variables, while RSF and DeepSurv models incorporated 24 variables, respectively. All of the 3 models achieved respectable performance metrics while the DeepSurv model exhibited the highest AUC values and relatively lower BS among these models. The DCA also verified that the DeepSurv model had the best clinical practicality. CONCLUSIONS: The survival prediction tools established in this study can be applied to severe HF patients during vulnerable period by mainly inputting drug treatment duration, thus contributing to optimal clinical decisions prospectively.
Assuntos
Insuficiência Cardíaca , Modelos de Riscos Proporcionais , Humanos , Insuficiência Cardíaca/mortalidade , Insuficiência Cardíaca/tratamento farmacológico , Feminino , Masculino , Idoso , Reprodutibilidade dos Testes , Prognóstico , Análise de Sobrevida , Pessoa de Meia-Idade , Curva ROC , Algoritmos , Área Sob a Curva , Bases de Dados Factuais , Aprendizado Profundo , Índice de Gravidade de DoençaRESUMO
BACKGROUND: Non-Hodgkin lymphoma (NHL) has been identified as a significant contributor to the cancer burden. This study investigates the incidence, mortality, and survival trends of NHL cancer in Brunei Darussalam from 2011 to 2020. METHODS: This is a registry-based retrospective study using de-identified data from the Brunei Darussalam Cancer Registry on patients diagnosed with NHL from 2011 to 2020 based on the ICD-10 codes C82-86. Statistical methods include descriptive statistics, age-specific and age-standardised incidence (ASIR) and mortality rates (ASMR), and joinpoint regression for trend analysis. Survival analysis was conducted using Kaplan-Meier plots, log-rank test, and Cox Proportional Hazards regression. RESULTS: From 2011 to 2020, 330 patients were diagnosed with NHL. The majority of patients were males (51.8%) and of Malay descent (82.7%). The age group most diagnosed was 55-74 years (42.3%), with a mean age at diagnosis being 55.1 years. The ASIRs were 12.12 for males and 10.39 per 100,000 for females; ASMRs were 6.11 for males and 4.76 per 100,000 for females. Diffuse large B-cell lymphoma was the most prevalent subtype, accounting for 39.1% of cases. The overall 5-year survival rate was 61.2%, with lower rates observed in older patients and those diagnosed at distant metastasis stage. Furthermore, older age and advanced stage diagnosis significantly increased mortality risk. NHL incidence and mortality rates in Brunei Darussalam remain stable over the period of 10 years, but highlights significant disparities in gender and age. CONCLUSIONS: The findings emphasize the importance of early detection and tailored treatments, especially for high-risk groups, in managing NHL's burden. These insights underline the need for focused healthcare strategies and continued research to address NHL's challenges.
Assuntos
Linfoma não Hodgkin , Humanos , Masculino , Feminino , Brunei/epidemiologia , Linfoma não Hodgkin/epidemiologia , Linfoma não Hodgkin/mortalidade , Pessoa de Meia-Idade , Incidência , Idoso , Estudos Retrospectivos , Adulto , Adulto Jovem , Sistema de Registros , Idoso de 80 Anos ou mais , Adolescente , Criança , Pré-Escolar , Lactente , Taxa de SobrevidaRESUMO
Recently, it has become common for applied works to combine commonly used survival analysis modeling methods, such as the multivariable Cox model and propensity score weighting, with the intention of forming a doubly robust estimator of an exposure effect hazard ratio that is unbiased in large samples when either the Cox model or the propensity score model is correctly specified. This combination does not, in general, produce a doubly robust estimator, even after regression standardization, when there is truly a causal effect. We demonstrate via simulation this lack of double robustness for the semiparametric Cox model, the Weibull proportional hazards model, and a simple proportional hazards flexible parametric model, with both the latter models fit via maximum likelihood. We provide a novel proof that the combination of propensity score weighting and a proportional hazards survival model, fit either via full or partial likelihood, is consistent under the null of no causal effect of the exposure on the outcome under particular censoring mechanisms if either the propensity score or the outcome model is correctly specified and contains all confounders. Given our results suggesting that double robustness only exists under the null, we outline 2 simple alternative estimators that are doubly robust for the survival difference at a given time point (in the above sense), provided the censoring mechanism can be correctly modeled, and one doubly robust method of estimation for the full survival curve. We provide R code to use these estimators for estimation and inference in the supporting information.
Assuntos
Simulação por Computador , Pontuação de Propensão , Modelos de Riscos Proporcionais , Humanos , Análise de Sobrevida , Funções Verossimilhança , Biometria/métodosRESUMO
We estimate relative hazards and absolute risks (or cumulative incidence or crude risk) under cause-specific proportional hazards models for competing risks from double nested case-control (DNCC) data. In the DNCC design, controls are time-matched not only to cases from the cause of primary interest, but also to cases from competing risks (the phase-two sample). Complete covariate data are available in the phase-two sample, but other cohort members only have information on survival outcomes and some covariates. Design-weighted estimators use inverse sampling probabilities computed from Samuelsen-type calculations for DNCC. To take advantage of additional information available on all cohort members, we augment the estimating equations with a term that is unbiased for zero but improves the efficiency of estimates from the cause-specific proportional hazards model. We establish the asymptotic properties of the proposed estimators, including the estimator of absolute risk, and derive consistent variance estimators. We show that augmented design-weighted estimators are more efficient than design-weighted estimators. Through simulations, we show that the proposed asymptotic methods yield nominal operating characteristics in practical sample sizes. We illustrate the methods using prostate cancer mortality data from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Study of the National Cancer Institute.
Assuntos
Modelos de Riscos Proporcionais , Neoplasias da Próstata , Estudos de Casos e Controles , Humanos , Masculino , Medição de Risco/estatística & dados numéricos , Medição de Risco/métodos , Neoplasias da Próstata/mortalidade , Simulação por Computador , Interpretação Estatística de Dados , Biometria/métodos , Fatores de RiscoRESUMO
OBJECTIVES: Prognostic and diagnostic models must work in their intended clinical setting, proven via "external evaluation", preferably by authors uninvolved with model development. By systematic review, we determined the proportion of models published in high-impact radiological journals that are evaluated subsequently. METHODS: We hand-searched three radiological journals for multivariable diagnostic/prognostic models 2013-2015 inclusive, developed using regression. We assessed completeness of data presentation to allow subsequent external evaluation. We then searched literature to August 2022 to identify external evaluations of these index models. RESULTS: We identified 98 index studies (73 prognostic; 25 diagnostic) describing 145 models. Only 15 (15%) index studies presented an evaluation (two external). No model was updated. Only 20 (20%) studies presented a model equation. Just 7 (15%) studies developing Cox models presented a risk table, and just 4 (9%) presented the baseline hazard. Two (4%) studies developing non-Cox models presented the intercept. Just 20 (20%) articles presented a Kaplan-Meier curve of the final model. The 98 index studies attracted 4224 citations (including 559 self-citations), median 28 per study. We identified just six (6%) subsequent external evaluations of an index model, five of which were external evaluations by researchers uninvolved with model development, and from a different institution. CONCLUSIONS: Very few prognostic or diagnostic models published in radiological literature are evaluated externally, suggesting wasted research effort and resources. Authors' published models should present data sufficient to allow external evaluation by others. To achieve clinical utility, researchers should concentrate on model evaluation and updating rather than continual redevelopment. CLINICAL RELEVANCE STATEMENT: The large majority of prognostic and diagnostic models published in high-impact radiological journals are never evaluated. It would be more efficient for researchers to evaluate existing models rather than practice continual redevelopment. KEY POINTS: ⢠Systematic review of highly cited radiological literature identified few diagnostic or prognostic models that were evaluated subsequently by researchers uninvolved with the original model. ⢠Published radiological models frequently omit important information necessary for others to perform an external evaluation: Only 20% of studies presented a model equation or nomogram. ⢠A large proportion of research citing published models focuses on redevelopment and ignores evaluation and updating, which would be a more efficient use of research resources.
Assuntos
Publicações Periódicas como Assunto , Humanos , Prognóstico , Modelos de Riscos Proporcionais , Radiografia , NomogramasRESUMO
The Cox proportional hazards model is commonly used to analyze time-to-event data in clinical trials. Standard inference procedures for the Cox model are based on asymptotic approximations and may perform poorly when there are few events in one or both treatment groups, as may be the case when the event of interest is rare or when the experimental treatment is highly efficacious. In this article, we propose an exact test of equivalence and efficacy under a proportional hazard model with treatment effect as the only fixed effect, together with an exact confidence interval that is obtained by inverting the exact test. The proposed test is based on a conditional error method originally proposed for sample size reestimation problems. In the present context, the conditional error method is used to combine information from a sequence of hypergeometric distributions, one at each observed event time. The proposed procedures are evaluated in simulation studies and illustrated using real data from an HIV prevention trial. A companion R package "ExactCox" is available for download on CRAN.
RESUMO
We propose a new simultaneous variable selection and estimation procedure with the Gaussian seamless- L 0 $$ {L}_0 $$ (GSELO) penalty for Cox proportional hazard model and additive hazards model. The GSELO procedure shows good potential to improve the existing variable selection methods by taking strength from both best subset selection (BSS) and regularization. In addition, we develop an iterative algorithm to implement the proposed procedure in a computationally efficient way. Theoretically, we establish the convergence properties of the algorithm and asymptotic theoretical properties of the proposed procedure. Since parameter tuning is crucial to the performance of the GSELO procedure, we also propose an extended Bayesian information criteria (EBIC) parameter selector for the GSELO procedure. Simulated and real data studies have demonstrated the prediction performance and effectiveness of the proposed method over several state-of-the-art methods.
Assuntos
Algoritmos , Humanos , Teorema de Bayes , Modelos de Riscos ProporcionaisRESUMO
Many longitudinal studies are designed to monitor participants for major events related to the progression of diseases. Data arising from such longitudinal studies are usually subject to interval censoring since the events are only known to occur between two monitoring visits. In this work, we propose a new method to handle interval-censored multistate data within a proportional hazards model framework where the hazard rate of events is modeled by a nonparametric function of time and the covariates affect the hazard rate proportionally. The main idea of this method is to simplify the likelihood functions of a discrete-time multistate model through an approximation and the application of data augmentation techniques, where the assumed presence of censored information facilitates a simpler parameterization. Then the expectation-maximization algorithm is used to estimate the parameters in the model. The performance of the proposed method is evaluated by numerical studies. Finally, the method is employed to analyze a dataset on tracking the advancement of coronary allograft vasculopathy following heart transplantation.
Assuntos
Algoritmos , Transplante de Coração , Modelos de Riscos Proporcionais , Humanos , Funções Verossimilhança , Transplante de Coração/estatística & dados numéricos , Estudos Longitudinais , Simulação por Computador , Modelos Estatísticos , Interpretação Estatística de DadosRESUMO
Left truncated right censored (LTRC) data arise quite commonly from survival studies. In this article, a model based on piecewise linear approximation is proposed for the analysis of LTRC data with covariates. Specifically, the model involves a piecewise linear approximation for the cumulative baseline hazard function of the proportional hazards model. The principal advantage of the proposed model is that it does not depend on restrictive parametric assumptions while being flexible and data-driven. Likelihood inference for the model is developed. Through detailed simulation studies, the robustness property of the model is studied by fitting it to LTRC data generated from different processes covering a wide range of lifetime distributions. A sensitivity analysis is also carried out by fitting the model to LTRC data generated from a process with a piecewise constant baseline hazard. It is observed that the performance of the model is quite satisfactory in all those cases. Analyses of two real LTRC datasets by using the model are provided as illustrative examples. Applications of the model in some practical prediction issues are discussed. In summary, the proposed model provides a comprehensive and flexible approach to model a general structure for LTRC lifetime data.
Assuntos
Modelos Estatísticos , Humanos , Análise de Sobrevida , Modelos de Riscos Proporcionais , Simulação por Computador , Funções VerossimilhançaRESUMO
Wide heterogeneity exists in cancer patients' survival, ranging from a few months to several decades. To accurately predict clinical outcomes, it is vital to build an accurate predictive model that relates the patients' molecular profiles with the patients' survival. With complex relationships between survival and high-dimensional molecular predictors, it is challenging to conduct nonparametric modeling and irrelevant predictors removing simultaneously. In this article, we build a kernel Cox proportional hazards semi-parametric model and propose a novel regularized garrotized kernel machine (RegGKM) method to fit the model. We use the kernel machine method to describe the complex relationship between survival and predictors, while automatically removing irrelevant parametric and nonparametric predictors through a LASSO penalty. An efficient high-dimensional algorithm is developed for the proposed method. Comparison with other competing methods in simulation shows that the proposed method always has better predictive accuracy. We apply this method to analyze a multiple myeloma dataset and predict the patients' death burden based on their gene expressions. Our results can help classify patients into groups with different death risks, facilitating treatment for better clinical outcomes.
Assuntos
Algoritmos , Neoplasias , Humanos , Modelos Lineares , Modelos de Riscos Proporcionais , Simulação por Computador , Neoplasias/genéticaRESUMO
Lung cancer is a leading cause of cancer deaths and imposes an enormous economic burden on patients. It is important to develop an accurate risk assessment model to determine the appropriate treatment for patients after an initial lung cancer diagnosis. The Cox proportional hazards model is mainly employed in survival analysis. However, real-world medical data are usually incomplete, posing a great challenge to the application of this model. Commonly used imputation methods cannot achieve sufficient accuracy when data are missing, so we investigated novel methods for the development of clinical prediction models. In this article, we present a novel model for survival prediction in missing scenarios. We collected data from 5,240 patients diagnosed with lung cancer at the Weihai Municipal Hospital, China. Then, we applied a joint model that combined a BN and a Cox model to predict mortality risk in individual patients with lung cancer. The established prognostic model achieved good predictive performance in discrimination and calibration. We showed that combining the BN with the Cox proportional hazards model is highly beneficial and provides a more efficient tool for risk prediction.
Assuntos
Neoplasias Pulmonares , Humanos , Neoplasias Pulmonares/diagnóstico , Teorema de Bayes , Prognóstico , Calibragem , China/epidemiologiaRESUMO
BACKGROUND: Semiparametric survival analysis such as the Cox proportional hazards (CPH) regression model is commonly employed in endometrial cancer (EC) study. Although this method does not need to know the baseline hazard function, it cannot estimate event time ratio (ETR) which measures relative increase or decrease in survival time. To estimate ETR, the Weibull parametric model needs to be applied. The objective of this study is to develop and evaluate the Weibull parametric model for EC patients' survival analysis. METHODS: Training (n = 411) and testing (n = 80) datasets from EC patients were retrospectively collected to investigate this problem. To determine the optimal CPH model from the training dataset, a bi-level model selection with minimax concave penalty was applied to select clinical and radiomic features which were obtained from T2-weighted MRI images. After the CPH model was built, model diagnostic was carried out to evaluate the proportional hazard assumption with Schoenfeld test. Survival data were fitted into a Weibull model and hazard ratio (HR) and ETR were calculated from the model. Brier score and time-dependent area under the receiver operating characteristic curve (AUC) were compared between CPH and Weibull models. Goodness of the fit was measured with Kolmogorov-Smirnov (KS) statistic. RESULTS: Although the proportional hazard assumption holds for fitting EC survival data, the linearity of the model assumption is suspicious as there are trends in the age and cancer grade predictors. The result also showed that there was a significant relation between the EC survival data and the Weibull distribution. Finally, it showed that Weibull model has a larger AUC value than CPH model in general, and it also has smaller Brier score value for EC survival prediction using both training and testing datasets, suggesting that it is more accurate to use the Weibull model for EC survival analysis. CONCLUSIONS: The Weibull parametric model for EC survival analysis allows simultaneous characterization of the treatment effect in terms of the hazard ratio and the event time ratio (ETR), which is likely to be better understood. This method can be extended to study progression free survival and disease specific survival. TRIAL REGISTRATION: ClinicalTrials.gov NCT03543215, https://clinicaltrials.gov/ , date of registration: 30th June 2017.
Assuntos
Neoplasias do Endométrio , Imageamento por Ressonância Magnética , Modelos de Riscos Proporcionais , Humanos , Feminino , Neoplasias do Endométrio/mortalidade , Neoplasias do Endométrio/diagnóstico por imagem , Pessoa de Meia-Idade , Imageamento por Ressonância Magnética/métodos , Estudos Retrospectivos , Análise de Sobrevida , Idoso , Curva ROC , Adulto , Modelos Estatísticos , RadiômicaRESUMO
BACKGROUND: Pocock-Simon's minimisation method has been widely used to balance treatment assignments across prognostic factors in randomised controlled trials (RCTs). Previous studies focusing on the survival outcomes have demonstrated that the conservativeness of asymptotic tests without adjusting for stratification factors, as well as the inflated type I error rate of adjusted asymptotic tests conducted in a small sample of patients, can be relaxed using re-randomisation tests. Although several RCTs using minimisation have suggested the presence of non-proportional hazards (non-PH) effects, the application of re-randomisation tests has been limited to the log-rank test and Cox PH models, which may result in diminished statistical power when confronted with non-PH scenarios. To address this issue, we proposed two re-randomisation tests based on a maximum combination of weighted log-rank tests (MaxCombo test) and the difference in restricted mean survival time (dRMST) up to a fixed time point τ , both of which can be extended to adjust for randomisation stratification factors. METHODS: We compared the performance of asymptotic and re-randomisation tests using the MaxCombo test, dRMST, log-rank test, and Cox PH models, assuming various non-PH situations for RCTs using minimisation, with total sample sizes of 50, 100, and 500 at a 1:1 allocation ratio. We mainly considered null, and alternative scenarios featuring delayed, crossing, and diminishing treatment effects. RESULTS: Across all examined null scenarios, re-randomisation tests maintained the type I error rates at the nominal level. Conversely, unadjusted asymptotic tests indicated excessive conservatism, while adjusted asymptotic tests in both the Cox PH models and dRMST indicated inflated type I error rates for total sample sizes of 50. The stratified MaxCombo-based re-randomisation test consistently exhibited robust power across all examined scenarios. CONCLUSIONS: The re-randomisation test is a useful alternative in non-PH situations for RCTs with minimisation using the stratified MaxCombo test, suggesting its robust power in various scenarios.
Assuntos
Modelos de Riscos Proporcionais , Ensaios Clínicos Controlados Aleatórios como Assunto , Humanos , Ensaios Clínicos Controlados Aleatórios como Assunto/métodos , Ensaios Clínicos Controlados Aleatórios como Assunto/estatística & dados numéricos , Projetos de Pesquisa/estatística & dados numéricos , Análise de Sobrevida , Modelos Estatísticos , Interpretação Estatística de DadosRESUMO
BACKGROUND: Predicting an individual's risk of death from COVID-19 is essential for planning and optimising resources. However, since the real-world mortality rate is relatively low, particularly in places like Hong Kong, this makes building an accurate prediction model difficult due to the imbalanced nature of the dataset. This study introduces an innovative application of graph convolutional networks (GCNs) to predict COVID-19 patient survival using a highly imbalanced dataset. Unlike traditional models, GCNs leverage structural relationships within the data, enhancing predictive accuracy and robustness. By integrating demographic and laboratory data into a GCN framework, our approach addresses class imbalance and demonstrates significant improvements in prediction accuracy. METHODS: The cohort included all consecutive positive COVID-19 patients fulfilling study criteria admitted to 42 public hospitals in Hong Kong between January 23 and December 31, 2020 (n = 7,606). We proposed the population-based graph convolutional neural network (GCN) model which took blood test results, age and sex as inputs to predict the survival outcomes. Furthermore, we compared our proposed model to the Cox Proportional Hazard (CPH) model, conventional machine learning models, and oversampling machine learning models. Additionally, a subgroup analysis was performed on the test set in order to acquire a deeper understanding of the relationship between each patient node and its neighbours, revealing possible underlying causes of the inaccurate predictions. RESULTS: The GCN model was the top-performing model, with an AUC of 0.944, considerably outperforming all other models (p < 0.05), including the oversampled CPH model (0.708), linear regression (0.877), Linear Discriminant Analysis (0.860), K-nearest neighbours (0.834), Gaussian predictor (0.745) and support vector machine (0.847). With Kaplan-Meier estimates, the GCN model demonstrated good discriminability between low- and high-risk individuals (p < 0.0001). Based on subanalysis using the weighted-in score, although the GCN model was able to discriminate well between different predicted groups, the separation was inadequate between false negative (FN) and true negative (TN) groups. CONCLUSION: The GCN model considerably outperformed all other machine learning methods and baseline CPH models. Thus, when applied to this imbalanced COVID survival dataset, adopting a population graph representation may be an approach to achieving good prediction.
Assuntos
COVID-19 , Redes Neurais de Computação , SARS-CoV-2 , Humanos , COVID-19/mortalidade , COVID-19/diagnóstico , Masculino , Feminino , Pessoa de Meia-Idade , Hong Kong/epidemiologia , Idoso , Adulto , Testes Hematológicos/métodos , Aprendizado de Máquina , Modelos de Riscos Proporcionais , Estudos de CoortesRESUMO
OBJECTIVE: Survival analysis is widely utilized in healthcare to predict the timing of disease onset. Traditional methods of survival analysis are usually based on Cox Proportional Hazards model and assume proportional risk for all subjects. However, this assumption is rarely true for most diseases, as the underlying factors have complex, non-linear, and time-varying relationships. This concern is especially relevant for pregnancy, where the risk for pregnancy-related complications, such as preeclampsia, varies across gestation. Recently, deep learning survival models have shown promise in addressing the limitations of classical models, as the novel models allow for non-proportional risk handling, capturing nonlinear relationships, and navigating complex temporal dynamics. METHODS: We present a methodology to model the temporal risk of preeclampsia during pregnancy and investigate the associated clinical risk factors. We utilized a retrospective dataset including 66,425 pregnant individuals who delivered in two tertiary care centers from 2015 to 2023. We modeled the preeclampsia risk by modifying DeepHit, a deep survival model, which leverages neural network architecture to capture time-varying relationships between covariates in pregnancy. We applied time series k-means clustering to DeepHit's normalized output and investigated interpretability using Shapley values. RESULTS: We demonstrate that DeepHit can effectively handle high-dimensional data and evolving risk hazards over time with performance similar to the Cox Proportional Hazards model, achieving an area under the curve (AUC) of 0.78 for both models. The deep survival model outperformed traditional methodology by identifying time-varied risk trajectories for preeclampsia, providing insights for early and individualized intervention. K-means clustering resulted in patients delineating into low-risk, early-onset, and late-onset preeclampsia groups-notably, each of those has distinct risk factors. CONCLUSION: This work demonstrates a novel application of deep survival analysis in time-varying prediction of preeclampsia risk. Our results highlight the advantage of deep survival models compared to Cox Proportional Hazards models in providing personalized risk trajectory and demonstrating the potential of deep survival models to generate interpretable and meaningful clinical applications in medicine.
Assuntos
Pré-Eclâmpsia , Humanos , Pré-Eclâmpsia/mortalidade , Gravidez , Feminino , Análise de Sobrevida , Fatores de Risco , Aprendizado Profundo , Adulto , Estudos Retrospectivos , Modelos de Riscos Proporcionais , Redes Neurais de Computação , Medição de Risco/métodosRESUMO
OBJECTIVE: To develop a lossless distributed algorithm for regularized Cox proportional hazards model with variable selection to support federated learning for vertically distributed data. METHODS: We propose a novel distributed algorithm for fitting regularized Cox proportional hazards model when data sharing among different data providers is restricted. Based on cyclical coordinate descent, the proposed algorithm computes intermediary statistics by each site and then exchanges them to update the model parameters in other sites without accessing individual patient-level data. We evaluate the performance of the proposed algorithm with (1) a simulation study and (2) a real-world data analysis predicting the risk of Alzheimer's dementia from the Religious Orders Study and Rush Memory and Aging Project (ROSMAP). Moreover, we compared the performance of our method with existing privacy-preserving models. RESULTS: Our algorithm achieves privacy-preserving variable selection for time-to-event data in the vertically distributed setting, without degradation of accuracy compared with a centralized approach. Simulation demonstrates that our algorithm is highly efficient in analyzing high-dimensional datasets. Real-world data analysis reveals that our distributed Cox model yields higher accuracy in predicting the risk of Alzheimer's dementia than the conventional Cox model built by each data provider without data sharing. Moreover, our algorithm is computationally more efficient compared with existing privacy-preserving Cox models with or without regularization term. CONCLUSION: The proposed algorithm is lossless, privacy-preserving and highly efficient to fit regularized Cox model for vertically distributed data. It provides a suitable and convenient approach for modeling time-to-event data in a distributed manner.