A novel nomogram based on DLL3 and PD-L1 for predicting the prognosis of patients with small cell lung cancer
Xiaoling Zhang1, Weiling Li2, Yunyi Du1
1Department of Oncology, Changzhi People's Hospital Affiliated to Changzhi Medical College, Changzhi, 046000, Shanxi, China.
Background:
Small cell lung cancer (SCLC) is known for its highly invasive nature and low 5-year survival rate. Although immunotherapy has shown enhancements in survival among SCLC patients, there is a call for improved precision in targeting beneficiary populations. The development of the Delta-like ligand 3 (DLL3)-targeted drug Rovalpituzumab Tesirine (Rova-T) was halted due to insufficient efficacy, prompting a closer examination of factors influencing treatment effectiveness. This study aims to create a prognostic model for SCLC patients based on protein expression levels to enhance prognosis determination and guide clinical decision-making.
Methods:
Immunohistochemistry assessed DLL3, Programmed death-ligand 1 (PD-L1), and c-kit expression in SCLC patients. Univariate and multivariate Cox proportional hazards regression identified relevant variables such as clinical characteristics, DLL3, and PD-L1 expression to formulate a prognostic model represented as a nomogram. Model performance was evaluated using the consistency index (C-index), bootstrap resampling, and decision curve analysis (DCA).
Results:
A total of 141 patients with SCLC were included in this study. Expression rates for DLL3, PD-L1, and c-kit were 62.4%, 10.6%, and 66.0% respectively. Differences in overall survival (OS) were found between SCLC patients by gender, smoking, stage, treatment, DLL3 and PD-L1. Neuron-Specific Enolase, stage, treatment, DLL3 and PD-L1 were independent risk factors for OS in SCLC patients (P < 0.05). Using these factors, a prognostic model predicting the 12-month survival probability for SCLC patients was developed with a distinguishable C-index of 0.739. The model's calibration curve exhibited accurate alignment between predicted and actual 12-month OS probabilities. The DCA curve showcased valuable clinical applicability. Patients were categorized into low-risk and high-risk groups using a cutoff value of 49.11 for the nomogram, underscoring the model's clinical discernibility (p < 0.0001).
Conclusion:
The prognostic model developed in this study offers predictive value in estimating 12-month survival probability for SCLC patients, aiding clinicians in making more informed treatment decisions.
