Plasma Epstein-Barr virus (EBV) DNA has been identified as a significant prognostic marker for nasopharyngeal carcinoma (NPC), yet there is limited research on the prognosis of NPC patients with negative EBV DNA. We explore the prognostic value of comprehensive immune-inflammatory and nutritional indicators to offer personalized treatment recommendations and prognosis predictions for non-metastatic NPC patients with negative EBV DNA. This was a retrospective study. This study retrospectively analyzed 257 non-metastatic NPC patients with negative EBV DNA between January 2015 and December 2019. The Kaplan-Meier survival curves evaluated survival endpoints, and group discrepancies were assessed with log-rank tests. Principal component analysis (PCA) reduced data dimensionality. Univariate and multivariate Cox regression analyses identified significant prognostic variables. Risk stratification was performed based on recursive partitioning analysis (RPA). A robust prognostic model was constructed by nomogram and evaluated by calibration curves, decision curves, and the time-dependent area under the curve analysis. PCA was employed to compute the immune-inflammation index (III) and nutrition index (NI). Multivariate Cox regression analysis revealed lactate dehydrogenase, III, and NI as significant prognostic variables for overall survival (OS). Utilizing RPA, we stratified the risk into three categories: low-risk group (low III + high NI), middle-risk group (low III + low NI), and high-risk group (high III). Both the middle- (p = 0.025) and high-risk groups (p < 0.001) exhibited poorer OS compared with the low-risk group. The nomogram model exhibited superior predictive accuracy compared to tumor lymph node metastasis stage alone (C-index: 0.774 vs 0.679). Our study validated the prognostic significance of III and NI in non-metastatic NPC patients with negative EBV DNA. Additionally, a clinical risk stratification was constructed to offer valuable insights into the individualized treatment of these patients.