1
|
Beyene KM, Chen DG, Kifle YG. A novel nonparametric time-dependent precision-recall curve estimator for right-censored survival data. Biom J 2024; 66:e2300135. [PMID: 38637327 DOI: 10.1002/bimj.202300135] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/20/2023] [Revised: 10/04/2023] [Accepted: 12/27/2023] [Indexed: 04/20/2024]
Abstract
In order to assess prognostic risk for individuals in precision health research, risk prediction models are increasingly used, in which statistical models are used to estimate the risk of future outcomes based on clinical and nonclinical characteristics. The predictive accuracy of a risk score must be assessed before it can be used in routine clinical decision making, where the receiver operator characteristic curves, precision-recall curves, and their corresponding area under the curves are commonly used metrics to evaluate the discriminatory ability of a continuous risk score. Among these the precision-recall curves have been shown to be more informative when dealing with unbalanced biomarker distribution between classes, which is common in rare event, even though except one, all existing methods are proposed for classic uncensored data. This paper is therefore to propose a novel nonparametric estimation approach for the time-dependent precision-recall curve and its associated area under the curve for right-censored data. A simulation is conducted to show the better finite sample property of the proposed estimator over the existing method and a real-world data from primary biliary cirrhosis trial is used to demonstrate the practical applicability of the proposed estimator.
Collapse
Affiliation(s)
- Kassu Mehari Beyene
- College of Health Solutions, Arizona State University, Phoenix, Arizona, USA
| | - Ding-Geng Chen
- College of Health Solutions, Arizona State University, Phoenix, Arizona, USA
- Department of Statistics, University of Pretoria, Pretoria, South Africa
| | - Yehenew Getachew Kifle
- Department of Mathematics and Statistics, University of Maryland Baltimore County, Baltimore, Maryland, USA
| |
Collapse
|
2
|
Beyene KM, Chen DG. Time-dependent receiver operating characteristic curve estimator for correlated right-censored time-to-event data. Stat Methods Med Res 2024; 33:162-181. [PMID: 38130110 DOI: 10.1177/09622802231220496] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/23/2023]
Abstract
In clinical trials, evaluating the accuracy of risk scores (markers) derived from prognostic models for prediction of survival outcomes is of major concern. The time-dependent receiver operating characteristic curve and the corresponding area under the receiver operating characteristic curve are appealing measures to evaluate the predictive accuracy. Several estimation methods have been proposed in the context of classical right-censored data which assumes the event time of individuals are independent. In many applications, however, this may not hold true if, for example, individuals belong to clusters or experience recurrent events. Estimates may be biased if this correlated nature is not taken into account. This paper is then aimed to fill this knowledge gap to introduce a time-dependent receiver operating characteristic curve and the corresponding area under the receiver operating characteristic curve estimation method for right-censored data that take the correlated nature into account. In the proposed method, the unknown status of censored subjects is imputed using conditional survival functions given the marker and frailty of the subjects. An extensive simulation study is conducted to evaluate and demonstrate the finite sample performance of the proposed method. Finally, the proposed method is illustrated using two real-world examples of lung cancer and kidney disease.
Collapse
Affiliation(s)
| | - Ding-Geng Chen
- Arizona State University, College of Health Solutions, AZ, USA
- Department of Statistics, University of Pretoria, Pretoria, South Africa
| |
Collapse
|
3
|
Somasundaram E, Anderson PM, Smile TD, Halima A, Broughman JB, Reddy CA, Parsai S, Scott JG, Chan T, Campbell S, Angelov L, Zahler S, Trucco M, Thomas SM, Johnson S, Qi P, Magnelli A, Murphy ES. Neutrophil to lymphocyte ratio (NTLR) predicts local control and overall survival after stereotactic body radiotherapy (SBRT) in metastatic sarcoma. Sci Rep 2023; 13:19256. [PMID: 37935813 PMCID: PMC10630331 DOI: 10.1038/s41598-023-46476-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/10/2023] [Accepted: 11/01/2023] [Indexed: 11/09/2023] Open
Abstract
The neutrophil to lymphocyte ratio (NTLR) and absolute lymphocyte count (ALC) recovery are prognostic across many cancers. We investigated whether NLTR predicts SBRT success or survival in a metastatic sarcoma cohort treated with SBRT from 2014 and 2020 (N = 42). Wilcox Signed Rank Test and Friedman Test compare NTLR changes with local failure vs. local control (N = 138 lesions). Cox analyses identified factors associated with overall survival. If local control was successful, NLTR change was not significant (p = 0.30). However, NLTR significantly changed in patients with local failure (p = 0.027). The multivariable Cox model demonstrated higher NLTR before SBRT was associated with worse overall survival (p = 0.002). The optimal NTLR cut point was 5 (Youden index: 0.418). One-year overall survival in SBRT metastatic sarcoma cohort was 47.6% (CI 34.3%-66.1%). Patients with an NTLR above 5 had a one-year overall survival of 37.7% (21.4%-66.3%); patients with an NTLR below 5 had a significantly improved overall survival of 63% (43.3%-91.6%, p = 0.014). Since NTLR at the time of SBRT was significantly associated with local control success and overall survival in metastatic sarcoma treated with SBRT, future efforts to reduce tumor inhibitory microenvironment factors and improve lymphocyte recovery should be investigated.
Collapse
Affiliation(s)
| | - Peter M Anderson
- Department of Pediatric Hematology Oncology and Blood and Marrow Transplantation, Cleveland Clinic, Cleveland, OH, USA
| | - Timothy D Smile
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Ahmed Halima
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - James B Broughman
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Chandana A Reddy
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Shireen Parsai
- Department of Radiation Oncology, Ohio Health Riverside Methodist Hospital, Columbus, OH, USA
| | - Jacob G Scott
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Timothy Chan
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Shauna Campbell
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Lilyana Angelov
- Rose Ella Burkhardt Brain Tumor and Neuro-Oncology Center, Cleveland Clinic, Cleveland, OH, USA
| | - Stacey Zahler
- Department of Pediatric Hematology Oncology and Blood and Marrow Transplantation, Cleveland Clinic, Cleveland, OH, USA
| | - Matteo Trucco
- Department of Pediatric Hematology Oncology and Blood and Marrow Transplantation, Cleveland Clinic, Cleveland, OH, USA
| | - Stefanie M Thomas
- Department of Pediatric Hematology Oncology and Blood and Marrow Transplantation, Cleveland Clinic, Cleveland, OH, USA
| | - Shavaughn Johnson
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Peng Qi
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Anthony Magnelli
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA
| | - Erin S Murphy
- Department of Radiation Oncology, Taussig Cancer Center, Cleveland Clinic, R3 9500 Euclid Ave, Cleveland, 44195, OH, USA.
| |
Collapse
|
4
|
Cheng W, Li X. A semi-parametric approach for time-dependent ROC curves with nonignorable missing biomarker. J Biopharm Stat 2023; 33:555-574. [PMID: 36852969 DOI: 10.1080/10543406.2023.2170394] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/15/2021] [Accepted: 12/30/2022] [Indexed: 03/01/2023]
Abstract
The main purpose of this paper is to survey the statistical inference for covariate-specific time-dependent receiver operating characteristic (ROC) curves with nonignorable missing continuous biomarker values. To construct time-dependent ROC curves, we consider a joint model which assumes that the failure time depends on the continuous biomarker and the covariates through a Cox proportional hazards model and that the continuous biomarker depends on the covariates through a semiparametric location model. Assuming a purely parametric model on the propensity score, we utilize instrumental variables to deal with the identifiable issue and estimate the unknown parameters of the propensity score by a simple and efficient method. In addition, when the propensity score is estimated, we develop HT and AIPW approaches to estimate our interested quantities. In the presence of nonignorable missing biomarker, our AIPW estimators of the interested quantities are still doubly robust when the true propensity score is a special parametric logistic model. At last, simulation studies are conducted to assess the performance of our proposed approaches, and a real data analysis is also carried out to illustrate its application.
Collapse
Affiliation(s)
- Weili Cheng
- School of Mathematics and Statistics, North China University of Water Resources and Electric Power, Zhengzhou, China
| | - Xiaorui Li
- School of Mathematics and Statistics, North China University of Water Resources and Electric Power, Zhengzhou, China
| |
Collapse
|
5
|
Somasundaram E, Smile TD, Halima A, Broughman JB, Reddy CA, Parsai S, Scott JG, Chan T, Campbell S, Angelov L, Zahler S, Trucco M, Thomas SM, Johnson S, Qi P, Magnelli A, Anderson PM, Murphy ES. Neutrophil to Lymphocyte Ratio (NTLR) Predicts Local Control Failure and Overall Survival after Stereotactic Body Radiotherapy (SBRT) In Metastatic Sarcoma. RESEARCH SQUARE 2023:rs.3.rs-2570832. [PMID: 37333401 PMCID: PMC10275040 DOI: 10.21203/rs.3.rs-2570832/v1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/20/2023]
Abstract
The neutrophil to lymphocyte ratio (NTLR) and absolute lymphocyte count (ALC) recovery are prognostic across many cancers. We investigated whether NLTR predicts SBRT success or survival in a metastatic sarcoma cohort treated with SBRT from 2014 and 2020 (N = 42). Wilcox Signed Rank Test and Friedman Test compare NTLR changes with local failure vs. local control (N = 138 lesions). Cox analyses identified factors associated with overall survival. If local control was successful, NLTR change was not significant (p = 0.30). However, NLTR significantly changed in patients local failure (p = 0.027). The multivariable Cox model demonstrated higher NLTR before SBRT was associated with worse overall survival (p = 0.002). The optimal NTLR cut point was 5 (Youden index: 0.418). One-year overall survival in SBRT metastatic sarcoma cohort was 47.6% (CI 34.3%-66.1%). Patients with an NTLR above 5 had a one-year overall survival of 37.7% (21.4%-66.3%); patients with an NTLR below 5 had a significantly improved overall survival of 63% (43.3%-91.6%, p = 0.014). Since NTLR at the time of SBRT was significantly associated with local control success and overall survival in metastatic sarcoma treated with SBRT, future efforts to reduce tumor inhibitory microenvironment factors and improved lymphocyte recovery should be investigated.
Collapse
|
6
|
Liu H, Qian SC, Shao YF, Li HY, Zhang HJ. Prognostic Impact of Systemic Coagulation-Inflammation Index in Acute Type A Aortic Dissection Surgery. JACC. ASIA 2022; 2:763-776. [PMID: 36444319 PMCID: PMC9700012 DOI: 10.1016/j.jacasi.2022.06.007] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 11/29/2021] [Revised: 06/15/2022] [Accepted: 06/16/2022] [Indexed: 06/16/2023]
Abstract
BACKGROUND A novel hematologic parameter, systemic coagulation-inflammation (SCI) index reflecting inflammation and coagulation pathways could be easily obtained from clinically routine laboratory findings. We hypothesize that the SCI index has prognostic implication in predicting operative mortality for patients with acute type A aortic dissection (ATAAD). OBJECTIVES This study aims to investigate the prognostic value of the SCI index and to establish an SCI-adding nomogram for mortality prediction in ATAAD patients. METHODS A total of 1,967 ATAAD patients surgically repaired were collected from 12 Chinese cardiovascular centers by the 5A (Additive Anti-inflammatory Action for Aortopathy & Arteriopathy [Multicenter Retrospective Study]) study III (2016-2020). SCI index was calculated as platelet count × fibrinogen/white blood cell count on admission. By adding SCI index, a nomogram was developed and evaluated for 90-day mortality prediction with conventional predictors via the Cox model with 10-fold cross-validation. RESULTS Patients were stratified with low SCI (<40), middle SCI (40-100), or high SCI (>100). The 90-day survival rates increased with SCI index (low 86.9%; [95% CI: 84.9%-89.0%], middle 92.7% [95% CI: 90.9%-94.9%], and high 96.4% [95% CI: 94.2%-98.6%]; log-rank P < 0.001). SCI index is independently associated with 90-day mortality (adjusted hazard ratio: 0.549; 95% CI: 0.424-0.710; P < 0.001). The addition of SCI index provided significantly incremental prognostic value to base model including age, serum creatinine, DeBakey class, and location of intimal entry (area under the curve: 0.677; 95% CI: 0.641-0.716 vs 0.724; 95% CI: 0.685-0.760; P = 0.002), which was confirmed by net reclassification improvement index (0.158; 95% CI: 0.065-0.235; P < 0.001) and integrated discrimination improvement index (0.070; 95% CI: 0.007-0.036; P < 0.001). CONCLUSIONS SCI index is easily obtainable, performs moderately well as a predictor of short-term mortality in ATAAD patients, and may be useful for risk stratification in emergency settings. (Additive Anti-inflammatory Action for Aortopathy & Arteriopathy [Multicenter Retrospective Study] III NCT04918108).
Collapse
Affiliation(s)
- Hong Liu
- Department of Cardiovascular Surgery, the First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - Si-chong Qian
- Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China
| | - Yong-feng Shao
- Department of Cardiovascular Surgery, the First Affiliated Hospital of Nanjing Medical University, Nanjing, China
| | - Hai-yang Li
- Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China
| | - Hong-jia Zhang
- Department of Cardiovascular Surgery, Beijing Anzhen Hospital, Capital Medical University, Beijing, China
| | | |
Collapse
|
7
|
Geraili Z, Hajian-Tilaki K, Bayani M, Hosseini SR, Khafri S, Ebrahimpour S, Javanian M, Babazadeh A, Shokri M. Prognostic accuracy of inflammatory markers in predicting risk of ICU admission for COVID-19: application of time-dependent receiver operating characteristic curves. J Int Med Res 2022; 50:3000605221102217. [PMID: 35701893 PMCID: PMC9208048 DOI: 10.1177/03000605221102217] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/08/2023] Open
Abstract
Objective Intensive care unit (ICU) admission occurs at different times during hospitalization among patients with COVID-19. We aimed to evaluate the time-dependent receive operating characteristic (ROC) curve and area under the ROC curve, AUC(t), and accuracy of baseline levels of inflammatory markers C-reactive protein (CRP) and neutrophil-to-lymphocyte ratio (NLR) in predicting time to an ICU admission event in patients with severe COVID-19 infection. Methods In this observational study, we evaluated 724 patients with confirmed severe COVID-19 referred to Ayatollah Rohani Hospital, affiliated with Babol University of Medical Sciences, Iran. Results The AUC(t) of CRP and NLR reached 0.741 (95% confidence interval [CI]: 0.661–0.820) and 0.690 (95% CI: 0.607–0.772), respectively, in the first 3 days after hospital admission. The optimal cutoff values of CRP and NLR for stratification of ICU admission outcomes in patients with severe COVID-19 were 78 mg/L and 5.13, respectively. The risk of ICU admission was significantly greater for patients with these cutoff values (CRP hazard ratio = 2.98; 95% CI: 1.58–5.62; NLR hazard ratio = 2.90; 95% CI: 1.45–5.77). Conclusions Using time-dependent ROC curves, CRP and NLR values at hospital admission were important predictors of ICU admission. This approach is more efficient than using standard ROC curves.
Collapse
Affiliation(s)
- Zahra Geraili
- Social Determinants of Health Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| | - Karimollah Hajian-Tilaki
- Social Determinants of Health Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran.,Department of Biostatistics and Epidemiology, School of Public Health, Babol University of Medical Sciences, Babol, Iran
| | - Masomeh Bayani
- Infectious Diseases and Tropical Medicine Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| | - Seyed Reza Hosseini
- Social Determinants of Health Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| | - Soraya Khafri
- Department of Biostatistics and Epidemiology, School of Public Health, Babol University of Medical Sciences, Babol, Iran
| | - Soheil Ebrahimpour
- Infectious Diseases and Tropical Medicine Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| | - Mostafa Javanian
- Infectious Diseases and Tropical Medicine Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| | - Arefeh Babazadeh
- Infectious Diseases and Tropical Medicine Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| | - Mehran Shokri
- Infectious Diseases and Tropical Medicine Research Center, Health Research Institute, Babol University of Medical Sciences, Babol, Iran
| |
Collapse
|
8
|
Beyene KM, El Ghouch A. Time-dependent ROC curve estimation for interval-censored data. Biom J 2022; 64:1056-1074. [PMID: 35523738 DOI: 10.1002/bimj.202000382] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/16/2020] [Revised: 10/20/2021] [Accepted: 11/07/2021] [Indexed: 11/10/2022]
Abstract
The receiver-operating characteristic (ROC) curve is the most popular graphical method for evaluating the classification accuracy of a diagnostic marker. In time-to-event studies, the subject's event status is time-dependent, and hence, time-dependent extensions of ROC curve have been proposed. However, in practice, the calculation of this curve is not straightforward due to the presence of censoring that may be of different types. Existing methods focus on the more standard and simple case of right-censoring and neglect the general case of mixed interval-censored data that may involve left-, right-, and interval-censored observations. In this context, we propose and study a new time-dependent ROC curve estimator. We also consider some summary measures (area under the ROC curve and Youden index) traditionally associated with ROC as well as the Youden-based cutoff estimation method. The proposed method uses available data very efficiently. To this end, the unknown status (positive or negative) of censored subjects are estimated from the data via the estimation of the conditional survival function given the marker. For that, we investigate both model-based and nonparametric approaches. We also provide variance estimates and confidence intervals using Bootstrap. A simulation study is conducted to investigate the finite sample behavior of the proposed methods and to compare their performance with a competitor. Globally, we observed better finite sample performances for the proposed estimators. Finally, we illustrate the methods using two data sets one from a hypobaric decompression sickness study and the other from an oral health study. The proposed methods are implemented in the R package cenROC.
Collapse
Affiliation(s)
- Kassu Mehari Beyene
- ISBA, UCLouvain, Louvain la Neuve, Belgium.,Department of Statistics, College of Natural Sciences, Wollo University, Dessie, Ethiopia
| | | |
Collapse
|