Comparing the effectiveness of various cumulative doses of cisplatin in treating locally advanced nasopharyngeal cancer during the period of intensity-modulated radiotherapy: a meta-analysis

Objective In the age of intensity-modulated radiotherapy (IMRT), the best cumulative cisplatin dose (CCD) for concurrent chemoradiotherapy for locally advanced nasopharyngeal cancer (LA-NPC) is still up for debate. In order to guide clinical dose decisions, this meta-analysis sought to examine the effects of various CCD on survival outcomes. Methods A thorough search was conducted in PubMed, Embase, Web of Science, and Cochrane Library through December 2025. Randomized controlled trials or observational studies comparing CCD thresholds (≥200 mg/m 2 vs. <200 mg/m 2 ) in IMRT-treated patients were included. Overall survival (OS) and progression-free survival (PFS) were the main results. Treatment-related toxicities and distant metastasis-free survival (DMFS) were secondary outcomes. Random-effects models were used to pool hazard ratios (HRs) with 95% confidence intervals (CIs) for survival outcomes. Relative risks (RRs) with 95% CIs were computed for toxicities (≥2 trials). To investigate the robustness of the results, subgroup and sensitivity analyses were carried out. Results Included were 10 retrospective cohort studies with 11,220 individuals. OS displayed a favorable trend approaching statistical significance (HR = 0.65, 95% CI: 0.51–0.82). Moderate heterogeneity was observed (I 2 = 50.4%, P for heterogeneity = 0.060). PFS showed no significant difference between dose groups (HR = 0.81, 95% CI: 0.54–1.22), with substantial heterogeneity (I 2 = 81.0%, P for heterogeneity <0.001). Higher CCD was associated with significantly improved DMFS (HR = 0.54, 95% CI: 0.40–0.74), with moderate heterogeneity (I 2 = 61.4%, P for heterogeneity = 0.011). Subgroup analysis revealed that the OS and DMFS benefit was consistent across treatment modalities, while the PFS effect differed significantly by treatment modality ( p < 0.001). For toxicity, the lower-dose group demonstrated significantly reduced risks of any-grade leukopenia (RR = 0.91, 95% CI: 0.86–0.97, p = 0.002), any-grade ALT elevation (RR = 0.69, 95% CI: 0.58–0.82, p < 0.001), any-grade creatinine elevation (RR = 0.73, 95% CI: 0.60–0.88, p = 0.001), any-grade hearing impairment (RR = 0.81, 95% CI: 0.70–0.94, p = 0.005), and any-grade skin fibrosis (RR = 0.77, 95% CI: 0.64–0.92, p = 0.004), suggesting a more favorable safety profile. Conclusion Higher CCD improves DMFS in LA-NPC patients with IMRT, with a near-significant OS trend. However, this survival benefit must be balanced against increased risks of certain acute and late toxicities associated with higher doses. These findings support risk-adapted dose intensification strategies while emphasizing individualized decision-making that integrates efficacy, toxicity, and patient characteristics.