This study evaluates the efficacy, optimal duration, and target population for maintenance immunotherapy (MI) in patients with de novo metastatic nasopharyngeal carcinoma (dmNPC) who have achieved disease control following systemic treatment. A multicenter retrospective cohort study included 258 patients whose disease was controlled after chemoimmunotherapy and locoregional radiotherapy. The primary outcome was progression-free survival (PFS), with conditional survival and restricted mean survival time (RMST) analyses used to determine the optimal MI duration. Immune infiltration was assessed via multiplex immunohistochemistry. The results showed that the 2-year PFS was significantly higher in the MI group compared to the non-MI group (69.7% vs. 53.5%, p = .02). Multivariable analysis showed MI was an independent predictor of improved PFS (HR: 0.581, p = .006). Conditional survival and RMST analyses confirmed a significant improvement in PFS with MI continuation within 15 months. Patients with high densities of CD3+ T cells (HR: 0.546, p = .023), CD20+ B cells (HR: 370, p <
.001), and a high percentage of PD-L1+ tumor cells (HR: 0.440, p = .006) had significantly better PFS compared to those with lower levels. Furthermore, MI was particularly beneficial for patients with lower densities of CD3+ T cells (p = .018), CD20+ B cells (p <
.001), and lower PD-L1+ tumor cell percentages (p <
.001), while this benefit was not observed in patients with higher immune infiltration levels. In conclusion, a 15-month duration of MI significantly improves PFS in patients with dmNPC after systemic treatment. Patients with lower levels of immune infiltration tend to have poorer PFS but appear to gain greater benefit from MI.