The pretreatment platelet count predicts survival outcomes of diffuse large B-cell lymphoma: An analysis of 1007 patients in the rituximab era.

BACKGROUND The prognostic value of platelet count in diffuse large B-cell lymphoma (DLBCL) has not been extensively investigated. We aimed to examine the association of pretreatment platelet count with disease features, and further examine the prognostic significance of platelet count in DLBCL treated with the R-CHOP regimen. METHODS Patients with DLBCL diagnosed between Jan 1 st, 2005 and Dec 31 st, 2018 at Cancer Hospital, Chinese Academy of Medical Sciences were retrospectively analyzed. Propensity score matching (PSM) was used to balance confounding factors. RESULTS A total of 1007 eligible patients who received frontline R-CHOP or R-CHOP-like regimens were included in this study. The optimal cutoff value of platelet count was 157 × 109/L, as determined by the Maximally Selected Rank Statistics method. Patients with the platelet count ≤157 × 109/L had significantly inferior overall survival (OS) (5-year OS, 44.4 % vs. 74.9 %, P  157×109/L. Multivariate analyses showed that pretreatment platelet count ≤ 157 × 109/L was an adversely independent prognostic factor for OS (hazard ratio [HR] 1.960, 95 % confidence interval [CI] 1.418-2.709, P<0.001) and PFS (HR 1.443, 95 %CI 1.080-1.927, P = 0.013). The PSM analysis and subgroup analyses further confirmed the significantly negative impact of low platelet count on OS and PFS. CONCLUSION The pretreatment platelet count may be a simple, cost-effective and useful prognostic factor in DLBCL patients treated with frontline R-CHOP regimens. Further investigation is warranted to elucidate the biologic mechanism underlying the prognostic significance of platelet count in DLBCL.