Prediction of Overall Survival of Patients with Completely Resected Non-Small Cell Lung Cancer: Analyses of Preoperative Spirometry, Preoperative Blood Tests, and Other Clinicopathological Data

Analysis of prognostic factors and establishment of prediction model of lung adenocarcinoma based on SEER database

Background

Few models have been developed to predict survival outcomes for lung adenocarcinoma (LUAD). In this study, we aimed to establish a nomogram for the prediction of cancer-specific survival (CSS) in LUAD patients which can be further developed as a convenient web-based calculator.

Methods

We performed a retrospective analysis of 50,007 LUAD patients selected from the Surveillance, Epidemiology, and End Result (SEER) 18 registry database. To enhance the reliability of the analysis, the patients' data were further randomly divided into the training cohort (70%) and validation cohort (30%). The optimal age cut-off points were determined using X-tile software, and patients were divided into three age groups: 10-72, 73-79, and 80-99 years. We selected independent prognostic factors from 17 variables by Cox regression, and plotted a visual nomogram to predict the 1-, 3-, and 5-year CSS. The predictive performance of the nomogram was evaluated through the concordance index (C-index), calibration curve and receiver operating characteristic (ROC) curve. To facilitate CSS forecast, a web-based calculator has subsequently been developed.

Results

We selected sex, age, race, marital status, N stage, tumor size, surgery, radiotherapy, chemotherapy, and metastasis (bone, brain, liver, and lung) as independent prognostic factors. The C-index was 0.779 [95% confidence interval (CI): 0.775-0.783] in the training set prediction model, and 0.782 (95% CI: 0.778-0.786) in the validation set. ROC analysis showed that area under the curve (AUC) values were 0.700, 0.733 and 0.669 for the 1-, 3- and 5-year CSS in the training set and 0.700, 0.744 and 0.669 in the validation set, respectively. In the nomogram calibration curve, there was strong correlation between the observed and predictive values. A web-based calculator can be accessed at: https://hjhlovelfb.shinyapps.io/DynNomapp/.

Conclusions

This nomogram model has good predictive power and can help clinicians identify LUAD patients at high risk of cancer-related death. This nomogram is expected to be a precise and personalized tool for predicting the prognosis of patients with LUAD.

A Nomogram Based on CT Deep Learning Signature: A Potential Tool for the Prediction of Overall Survival in Resected Non-Small Cell Lung Cancer Patients

PURPOSE

To develop and further validate a deep learning signature-based nomogram from computed tomography (CT) images for prediction of the overall survival (OS) in resected non-small cell lung cancer (NSCLC) patients.

PATIENTS AND METHODS

A total of 1792 deep learning features were extracted from non-enhanced and venous-phase CT images for each NSCLC patient in training cohort (n=231). Then, a deep learning signature was built with the least absolute shrinkage and selection operator (LASSO) Cox regression model for OS estimation. At last, a nomogram was constructed with the signature and other independent clinical risk factors. The performance of nomogram was assessed by discrimination, calibration and clinical usefulness. In addition, in order to quantify the improvement in performance added by deep learning signature, the net reclassification improvement (NRI) was calculated. The results were validated in external validation cohort (n=77).

RESULTS

A deep learning signature with 9 selected features was significantly associated with OS in both training cohort (hazard ratio [HR]=5.455, 95% CI: 3.393-8.769, P<0.001) and external validation cohort (HR=3.029, 95% CI: 1.673-5.485, P=0.004). The nomogram combining deep learning signature with clinical risk factors of TNM stage, lymphatic vessel invasion and differentiation grade showed favorable discriminative ability with C-index of 0.800 as well as a good calibration, which was validated in external validation cohort (C-index=0.723). Additional value of deep learning signature to the nomogram was statistically significant (NRI=0.093, P=0.027 for training cohort; NRI=0.106, P=0.040 for validation cohort). Decision curve analysis confirmed the clinical usefulness of this nomogram in predicting OS.

CONCLUSION

The deep learning signature-based nomogram is a robust tool for prognostic prediction in resected NSCLC patients.

The correlation between novel peripheral blood cell ratios and 90-day mortality in patients with acute ischemic stroke

BACKGROUND

We aimed to investigate the correlation between the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), platelet-to-neutrophil ratio (PNR), platelet-to-white blood cell ratio (PWR) and 90-day mortality in patients with acute ischemic stroke (AIS).

METHODS

We retrospectively included 633 patients with AIS from January 2017 to May 2018. The correlation between each indicator and the degree of neurologic deficit was assessed. Kaplan-Meier survival curves based on blood cell ratios were used to analyze the 90-day survival rate of patients with AIS.

RESULTS

A total of 663 patients with AIS were enrolled, of which 24 (3.6%) experienced recurrence and 13 (2.0%) died. NLR>3.23 (odds ratio; OR = 2.236; 95% confidence interval [CI], 1.472-3.397; P<0.001), PNR<31.14 (OR = 0.471; 95% CI, 0.297-0.749; P = 0.001), and PWR<20.62 (OR = 0.498; 95% CI, 0.309-0.800; P = 0.004) were associated with an unfavorable 90-day prognosis. NLR>3.23, PWR<20.62, and PNR<31.14 were associated with an increased risk of 90-day mortality.

CONCLUSION

PNR, PWR, and NLR were associated with the 90-day mortality of patients with AIS. Patients with high NLRs or low PWRs and PNRs may have a greater risk of mortality than other patients. These clinical indicators may help clinicians judge unfavorable prognosis early and implement the appropriate interventions.