Soluble programmed death-ligand 1 as a prognostic biomarker for overall survival in patients with diffuse large B-cell lymphoma: a replication study and combined analysis of 508 patients

Authors: Delphine Rossille, Imane Azzaoui et al. Published in Leukemia 2017; 31(4):988-991

Journal: Leukemia 2017; 31(4):988-991

We recently reported that elevated soluble PD-L1 (sPD-L1) at diagnosis before any treatment was a prognostic factor for overall survival (OS) in a French de novo DLBCL cohort. We aimed to confirm our findings in an independent cohort, and assess sPD-L1 prognostic value with established prognostic factors.
Plasma sPD-L1 was measured on 225 US adults with de novo DLBCL of any stage treated by R-CHOP. A cut-off of twice the median sPD-L1 expression of matched healthy controls was used to define elevated sPD-L1 levels. PD-L1 expressions in the tumor site were measured. After pooling the French and American cohorts, stratified and multivariate survival analyses were performed.
sPD-L1 was increased in DLBCL US patients versus healthy controls (OR=8.8; 95%CI 4.4-19.6), and elevated sPD-L1 patients had an unfavorable outcome (HR=1.72; 95%CI 0.99-2.98), confirming the French results. There was no association of sPD-L1 with PD-L1 expressed in the tumor site. In the pooled analysis of 508 patients, sPD-L1 remained an independent prognostic factor (HR=1.65; 95%CI 1.10-2.45) after adjustment.
Elevated sPD-L1 patients had inferior OS versus not elevated sPD-L1 across IPI groups and we confirmed that sPD-L1 at diagnosis is an independent prognostic factor in DLBCL treated with rituximab plus anthracycline-based therapy.