Metastatic sites of baseline as predictors in recurrent or metastatic nasopharyngeal carinoma treated with PD-L1 inhibitor: a secondary analysis of multicenter, single-arm, phase II study (KL-A167).

Background: Immune checkpoint inhibitors (ICIs) show optimal treatment effects on recurrent or metastatic nasopharyngeal carcinoma(R/M NPC). Nonetheless, whether metastatic sites impact ICIs efficacy remains unclear.

Methods: We performed a secondary analysis of R/M NPC patients treated with KL-A167, a programmed cell death-ligand 1(PD-L1) inhibitor, based on a multicenter, single-arm, phase II study from China between 2019 and 2021 years, which represents the first and most comprehensive analysis of the effectiveness of a PD-L1 inhibitor in patients who have been previously treated. The Cox proportional hazard model was utilized to evaluate the association between sites and PFS and OS. Sensitivity analysis and subgroup analysis were carried out to confirm the reliability of our findings.

Results: A total of 153 R/M NPC patients were included. The mean age was 47 years and 81% of patients were males. All patients in our study had distant metastasis, with a majority (n = 69) presenting with more than 2 sites of distant metastasis upon admission. The collected sites of metastasis included liver, lung, lymph and bone. Among the 153 patients, 37.9% (58 patients) received anti-PD-L1 treatment for a minimum of 6 months, and 17.6% (27 patients) were treated for at least 12 months. By conducting multivariate analysis, R/M NPC patients with non-liver metastases presented significantly longer progress-free survival (PFS, HR:1.67, CI:1.09-0.2.55, p = 0.018) and overall survival (OS, HR:2.52, CI:1.49-4.28, p < 0.001) compared with those with liver metastasis. The median PFS (72 vs. 144 days, p < 0.0001) and OS (730 vs. 305 days, p < 0.0001) were significantly longer for patients with non-liver metastases. However, lung, bone and lymph node metastasis had no statistical significance on PFS and OS (p > 0.005). Our sensitive analysis showed liver metastases patients with less other site metastases (0 or 1) had shorter OS compared to non-liver metastases patients with more other metastases(≥ 2). Furthermore, subgroup analysis indicated the robustness evidence liver metastasis indeed a valuable prognostic factor for survival.

Conclusions: Compared to patients with other metastatic sites, R/M NPC patients with liver metastasis have poor survival patterns when receiving anti-PD-L1 therapy. Our study provides rational evidence for the urgent need to explore more efficacy treatment modalities for NPC patients with liver metastasis.