Pre-treatment peripheral absolute monocyte count predicts metastatic progression and survival outcomes in treatment-naive non-metastatic nasopharyngeal carcinoma

Background The tumor node metastasis (TNM) staging system does not fully capture tumor heterogeneity, underscoring the needs for reliable biomarkers for prognostication in non-metastatic nasopharyngeal carcinoma (NPC). While inflammatory markers have shown potential, the prognostic value of pre-treatment peripheral absolute monocyte count (AMC) in treatment-naive, non-metastatic NPC remains underexplored. Methods This retrospective cohort study analyzed 2,046 patients with newly diagnosed treatment-naive, non-metastatic NPC from 2009 to 2022. The optimal AMC cut-off value (0.63×10 9 /L) for distant metastasis-free survival (DMFS) was determined using maximally selected rank statistics method. Patients were stratified into low AMC (<0.63×10 9 /L, n=1370) and high AMC (≥0.63×10 9 /L, n=676) groups. The primary endpoints were DMFS, bone metastasis-free survival (BMFS), and overall survival (OS). Associations were assessed using Kaplan–Meier analysis, log-rank tests, Cox proportional hazards models, restricted cubic splines (RCS), subgroup analyses and sensitivity analyses. Diagnostic performance was evaluated using time-dependent receiver operating characteristic (tROC) analysis at 1, 3, and 5 years. Results Over a median follow-up of 77.4 months, distant metastasis, bone metastasis, and death occurred in 13.5%, 6.9%, and 24.2% of patients, respectively. High pre-treatment AMC was independently associated with significantly worse DMFS (hazard ratio [HR]=1.33, 95% confidence interval [CI]: 1.05–1.70, P = 0.020), BMFS (HR = 1.86, 95%CI: 1.33–2.60, P<0.001), and OS (HR = 1.34, 95%CI: 1.11–1.61, P = 0.002) in fully adjusted models. RCS revealed a linear association between AMC and metastasis risk (P for non-linearity >0.05), but a non-linear threshold effect for OS (P for non-linearity=0.011). The risk of all-cause mortality increased with AMC >0.54×10 9 /L, eventually reaching a plateau. Subgroup analyses confirmed the feasibility of AMC’s prognostic value across all patient strata (P for interaction>0.05). Time-dependent ROC analysis demonstrated the highest discriminatory accuracy for predicting 1-year bone metastasis (AUC = 0.720), while maintaining moderate prognostic utility for distant metastasis and overall survival across 1 to 5 years. Conclusion Pre-treatment peripheral AMC ≥0.63×10 9 /L could serve as an independent predictor of metastatic progression, particularly bone metastasis, and reduced survival in treatment-naive, non-metastatic NPC. Being easily obtainable via routine complete blood count, AMC provides significant prognostic value to enhance clinical risk stratification and guide individualized treatment planning.