Development of a novel nomogram for patients with SCLC and comparison with other models

Abstract

Background: Though several nomograms have been established to predict the survival probability of patients with small-cell lung cancer (SCLC), none involved enough variables. This study aimed to construct a novel prognostic nomogram and compare its performance with other models.

Methods: Seven hundred twenty-two patients were pathologically diagnosed with SCLC in Shanxi Province Cancer Hospital, Shanxi Hospital Affiliated to Cancer Hospital, Chinese Academy of Medical Sciences, Cancer Hospital Affiliated to Shanxi Medical University from January 2016 to December 2018. We input Forty-one factors by reviewing the medical records. The nomogram was constructed based on the variables identified by univariate and multivariate analyses in the training set and validated in the validation set. Then we compared the performance of the models in terms of discrimination, calibration, and clinical net benefit.

Results: There were eight variables involved in the nomogram: gender, monocyte (MON), neuron-specific enolase (NSE), cytokeratin 19 fragments (Cyfra211), M stage, radiotherapy (RT), chemotherapy cycles (CT cycles), and prophylactic cranial irradiation (PCI). The calibration curve showed a good correlation between the nomogram prediction and actual observation for overall survival (OS). The area under the curve (AUC) of the nomogram was higher, and the Integrated Brier score (IBS) was lower than other models, indicating a more accurate prediction. Decision curve analysis (DCA) showed a significant improvement in the clinical net benefit compared to the other models.

Conclusions: We constructed a novel nomogram to predict OS for patients with SCLC using more comprehensive and objective variables. It performed better than existing models and would assist clinicians in individually estimating risk and making a therapeutic regimen.

Keywords: (SCLC; Individually treatment); Nomogram; Overall survival; Prediction.

Fig. 1

Flow chart of patient selection in the training and validation set

Fig. 2

Nomogram to predict 1, 2, and 3-year overall survival for patients with SCLC

Fig. 3

The calibration curve of the nomogram for predicting survival probability in the training set (A) and validation set (B)

Fig. 4

The area under the receiver operating characteristic (ROC) curve (AUC) in the training set (A) and validation set (B)

Fig. 5

The Integrated Brier score (IBS) in the training set and validation set

Fig. 6

Decision curve analysis (DCA) in the training set and validation set