Establishment of a Prognostic Nomogram for Patients With Locoregionally Advanced Nasopharyngeal Carcinoma Incorporating TNM Stage, Post-Induction Chemotherapy Tumor Volume and Epstein-Barr Virus DNA Load.

OBJECTIVES

To establish and validate an effective nomogram to predict clinical outcomes for patients with locoregionally advanced nasopharyngeal carcinoma (LA-NPC).

MATERIALS AND METHODS

The clinicopathological parameters and follow-up information of 402 locoregionally advanced NPC patients (training cohort, n = 302; validation cohort, n = 100) were retrospectively enrolled. The nomogram was built with the important prognostic variables identified by Cox regression analysis. Overall survival (OS) and progression-free survival (PFS) were the primary and secondary endpoints, respectively. The predictive power and clinical utility of the nomogram were assessed using the Harrell concordance index (C-index), calibration curve, and decision curve analysis. We compared the eighth staging system model with the nomogram to analyze whether the model could improve the accuracy of prognosis.

RESULTS

Epstein-Barr virus (EBV) DNA load, the gross tumor volume (GTVnx), and cervical lymph node tumor volume (GTVnd) after induction chemotherapy were the independent predictors of OS and PFS. The calibration curves indicated superb agreement between the nomogram-predicted probabilities and observed actual probabilities of survival. The C-index and area under the receiver operator characteristic curve (AUC) of the nomogram integrating these significant factors and N stage, and TNM stage were higher than those of the eighth TNM system alone. In addition, the decision curve analyses demonstrated the clinical value and higher overall net benefit of the nomogram. High-risk groups identified by the nomogram had significantly poorer OS and PFS than the low-risk group (p < 0.05).

CONCLUSIONS

The multidimensional nomogram incorporating TNM stage, EBV DNA load, and tumor volume after induction chemotherapy led to a more precise prognostic prediction and could be helpful for stratifying risk and guiding treatment decisions in locoregionally advanced NPC patients who have undergone induction chemotherapy and concurrent chemoradiation.