Key words: Diffuse large B-cell lymphoma, Natural killer cell count, prognosis.
Introduction
Natural killer (NK) cell counts at diagnosis have been reported to be associated with poor outcomes, but the dynamic change and prognostic value of NK cell counts after treatment in diffuse large B-cell lymphoma (DLBCL)remain unknown.
NK cell count was test by flow cytometry at diagnosis, interim and the end of treatment. Receiver operating characteristic (ROC) curve was used to determine the best cutoff value for survival analysis. Overall survival (OS) and progression-free survival (PFS) were analyzed by the method of Kaplan-Meier and Cox proportional hazard regression model.
A total of 175 DLBCL patients were analyzed and 45 of them were divided into the low NK cell counts with the cut-off value was 94.5/ul. Patients with low NK cell counts were associated with B symptoms (p < 0.001), high LDH (p=0.002), advanced stage (p < 0.001) and high IPI scores (p=0.001). Survival analysis showed patients with low NK cell counts at diagnosis showed more inferior OS and PFS than those with high NK cell counts (p < 0.001 and p < 0.001). There were 60 patients showed low NK cell counts at the interim, consist of 31patients with low and 29 with high NK cell counts at diagnosis and 65 patients had low NK cell counts at the end of the treatment, consist of 31patients with low and 34 with high NK cell counts at diagnosis. But NK cell count at the interim or end of treatment had no impact on the PFS (p = 0.415 and p = 0.011) and OS (p = 0.537 and p = 0.836) of DLBCL patients. Multivariable analysis revealed that NK cell counts was an independent prognostic factor OS [relative risk (RR);95% confidence interval (CI),0.157-0.984, p = 0.046] and PFS (RR,0.335;95%CI,0.191-0.588, p < 0.001） in all DLBCL patients independent of IPI.
In conclusion, low NK cell counts at diagnosis is a simple and effective adverse prognostic factor in DLBCL patients treated with R-CHOP. However, the NK cell counts at the interim and the end of the treatment showed no impact on outcome.
Disclosures
No relevant conflicts of interest to declare.
