Dataset Information

An interactive Bayesian model for prediction of lymph node ratio and survival in pancreatic cancer patients.

ABSTRACT:

Background

Regional lymph node status has long been used as a dichotomous predictor of clinical outcomes in cancer patients. More recently, interest has turned to the prognostic utility of lymph node ratio (LNR), quantified as the proportion of positive nodes examined. However, statistical tools for the joint modeling of LNR and its effect on cancer survival are lacking.

Methods

Data were obtained from the NCI SEER cancer registry on 6400 patients diagnosed with pancreatic ductal adenocarcinoma from 2004 to 2010 and who underwent radical oncologic resection. A novel Bayesian statistical approach was developed and applied to model simultaneously patients' true, but unobservable, LNR statuses and overall survival. New web development tools were then employed to create an interactive web application for individualized patient prediction.

Results

Histologic grade and T and M stages were important predictors of LNR status. Significant predictors of survival included age, gender, marital status, grade, histology, T and M stages, tumor size, and radiation therapy. LNR was found to have a highly significant, non-linear effect on survival. Furthermore, predictive performance of the survival model compared favorably to those from studies with more homogeneous patients and individualized predictors.

Conclusions

We provide a new approach and tool set for the prediction of LNR and survival that are generally applicable to a host of cancer types, including breast, colon, melanoma, and stomach. Our methods are illustrated with the development of a validated model and web applications for the prediction of survival in a large set of pancreatic cancer patients.

SUBMITTER: Smith BJ

PROVIDER: S-EPMC4173165 | biostudies-literature | 2014 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

An interactive Bayesian model for prediction of lymph node ratio and survival in pancreatic cancer patients.

Smith Brian J BJ Mezhir James J JJ

Journal of the American Medical Informatics Association : JAMIA 20140120 e2

<h4>Background</h4>Regional lymph node status has long been used as a dichotomous predictor of clinical outcomes in cancer patients. More recently, interest has turned to the prognostic utility of lymph node ratio (LNR), quantified as the proportion of positive nodes examined. However, statistical tools for the joint modeling of LNR and its effect on cancer survival are lacking.<h4>Methods</h4>Data were obtained from the NCI SEER cancer registry on 6400 patients diagnosed with pancreatic ductal ...[more]

PMID: 24444460

Similar Datasets

Project description:ImportanceOral cavity squamous cell carcinoma (OCSCC) is associated with often-delayed clinical diagnosis, poor prognosis, and expensive therapeutic approaches. Prognostic accuracy is important in improving treatment outcomes of patients with this disease.ObjectivesTo assess lymph node ratio (LNR) and other factors in estimating response to treatment and provide prognostic information helpful for clinical decision making.Design, setting, and participantsA retrospective cohort study was conducted from January 1, 2000, to December 31, 2015, at an academic hospital in Denver, Colorado. Participants included 149 patients with primary OCSCC who received curative-intent surgery and/or postoperative adjuvant therapies. Analysis was performed from December 8, 2017, to August 15, 2018.Main outcomes and measuresOverall survival (OS), disease-free survival (DFS), locoregional disease-free survival (LRDFS), and distant metastasis-free survival (DMDFS) adjusted for known prognostic risk factors, as well as correlation of LNR with other histopathologic prognostic factors.ResultsOf the 149 patients included in analysis, 105 were men (70.5%); the median age at diagnosis was 59 years (range, 28-88 years). Using the Kaplan-Meier method, the 5-year survival estimates for OS rate was 40.4% (95% CI, 31.3%-49.3%); DFS, 48.6% (95% CI, 38.6%-58.0%); LRDFS, 57.7% (95% CI, 46.6%-67.2%); and DMDFS, 74.7% (95% CI, 65.1%-82.0%). The median follow-up was 20 months for all patients and 34.5 months (range, 0-137 months) for surviving patients. Nonwhite race (hazard ratio [HR], 2.15; 95% CI, 1.22-3.81), T3-T4 category (HR, 1.99; 95% CI, 1.18-3.35), and LNR greater than 10% (HR, 2.71; 95% CI, 1.39-5.27) were associated with poorer OS. Nonwhite patients also had higher risk of locoregional failures (HR, 2.47; 95% CI, 1.28-4.79), whereas women were more likely to have distant metastasis (HR, 2.55; 95% CI, 1.14-5.71). Floor-of-mouth subsite had fewer locoregional recurrences than did other subsites (HR, 0.45, 95% CI, 0.21-0.99). An LNR greater than 10% independently was associated with worse OS (HR, 2.71; 95% CI, 1.39-5.27), DFS (HR, 2.48; 95% CI, 1.18-5.22), and DMDFS (HR, 6.05; 95% CI, 1.54-23.71). The LNR was associated with N-stage (Cramer V, 0.69; 95% CI, 0.58-0.78), extracapsular extension (Cramer V, 0.55; 95% CI, 0.44-0.66), lymphovascular invasion (Cramer V, 0.46; 95% CI, 0.27-0.61); number of excised lymph nodes (Cramer V, 0.24; 95% CI, 0.06-0.37), margin (Cramer V, 0.22; 95% CI, 0.05-0.38), and tumor thickness combined with depth of invasion (Cramer V, 0.25; 95% CI, 0.05-0.38).Conclusions and relevanceLocoregional treatment failure remained the predominant pattern of failure. An advanced pathologic stage and nonwhite race were found to be associated with worse outcomes. The findings from this study suggest that LNR is the most robust prognostic factor and appears to have implications for risk stratification in this disease.

Project description:BackgroundNegative regional lymph nodes do not indicate a lack of distant metastasis. A considerable number of patients with negative regional lymph node pancreatic cancer will skip the step of regional lymph node metastasis and directly develop distant metastasis.MethodsWe retrospectively analyzed the clinicopathological characteristics of patients with negative regional lymph node pancreatic cancer and distant metastasis in the Surveillance, Epidemiology, and End Results database from 2010 to 2015. Multivariate logistic analysis and Cox analysis were used to determine the independent risk factors that promoted distant metastasis and the 1-, 2-, and 3-year cancer-specific survival in this subgroup.ResultsSex, age, pathological grade, surgery, radiotherapy, race, tumor location, and tumor size were significantly correlated with distant metastasis (P < 0.05). Among these factors, pathological grade II and above, tumor site other than the pancreatic head, and tumor size >40 mm were independent risk factors for distant metastasis; age ≥60 years, tumor size ≤21 mm, surgery, and radiation were protective factors against distant metastasis. Age, pathological grade, surgery, chemotherapy, and metastasis site were identified as predictors of survival. Among them, age ≥40 years, pathological grade II and above, and multiple distant metastasis were considered independent risk factors for cancer-specific survival. Surgery and chemotherapy were considered protective factors for cancer-specific survival. The prediction performance of the nomogram was significantly better than that of the traditional American Joint Committee on Cancer tumor, node, metastasis staging system. We also established an online dynamic nomogram calculator, which can predict the survival rate of patients at different follow-up time points.ConclusionPathological grade, tumor location, and tumor size were independent risk factors for distant metastasis in pancreatic ductal adenocarcinoma with negative regional lymph nodes. Older age, smaller tumor size, surgery, and radiotherapy were protective factors against distant metastasis. A new nomogram that was constructed could effectively predict cancer-specific survival in pancreatic ductal adenocarcinoma with negative regional lymph nodes and distant metastasis. Furthermore, an online dynamic nomogram calculator was established.

Dataset Information

An interactive Bayesian model for prediction of lymph node ratio and survival in pancreatic cancer patients.

Background

Methods

Results

Conclusions

Publications

An interactive Bayesian model for prediction of lymph node ratio and survival in pancreatic cancer patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets