This article is more than 5 years old. Information may no longer be current.
PD-L1 expression predicts survival in DLBCL
Patients with diffuse large B-cell lymphoma whose tumors expressed PD-L1 experienced shorter OS than patients without PD-L1 expression, according to the results of a retrospective analysis.
Immunotherapy targeting the PD-1/PD-L1 pathway should be considered for patients with PD-L1–positive diffuse large B-cell lymphoma (DLBCL), the researchers wrote.
PD-L1 is expressed on select DLBCL tumor cells and on tumor-infiltrating nonmalignant cells. The PD-1/PD-L1 pathway inhibits antitumor responses; however, little is known about how the pathway functions in tumor environments.
Koichi Oshima, MD, PhD, of the department of pathology at Kurume University’s School of Medicine in Japan, and colleagues performed PD-L1/PAX5 double immunostaining on 1,253 biopsy samples from patients with DLBCL and DLBCL subtypes.
The researchers defined PD-L1–positive DLBCL as 30% or more lymphoma cells expressing distinct membranous and/or cytoplasmic staining of PD-L1 and nuclear staining of PAX5, regardless of the PD-L1 positivity of nonmalignant stromal cells.
They defined microenvironmental PD-L1–positive DLBCL among PD-L1–negative samples as those in which PD-L1–positive nonmalignant stromal cells comprised 20% or more of total tissue cellularity.
In the entire patient cohort, 10.5% (n = 132) of patients had PD-L1–positive DLBCL and 15.3% of patients (n = 172) had microenvironmental PD-L1–positive DLBCL.
Thirty-seven percent of patients (n = 461) received a diagnosis of germinal center B-cell–like (GCB) DLBCL and 9% of patients (n = 114) had Epstein-Barr virus positivity. PD-L1 positivity and microenvironmental PD-L1 positivity served as significant associative factors for these subtypes (P < .0001 and P = .0014, respectively).
The researchers performed quantitative analyses of PD-L1–positive tumor-infiltrating lymphocytes on 273 patients with available clinical data.
GCB-type tumors had a significantly higher number of PD-1–positive tumor-infiltrating lymphocytes (P = .034). The researchers found significantly lower PD-1–positive tumor-infiltrating lymphocytes in microenvironmental PD-L1–positive DLBCL (P = .017) and PD-L1–positive DLBCL (P < .0001).
Overall, patients with PD-L1–positive DLBCL had inferior OS outcomes compared with patients with PD-L1–negative DLBCL (P = .0009). Researchers observed no significant differences in OS between patients with microenvironmental PD-L1–positive and PD-L1–negative DLBCL.
In multivariate analyses that accounted for age, sex, clinical stage, performance status and other variables, PD-L1 positivity remained a significant prognostic factor (P = .0323).
The researchers acknowledged the small sample size and the lack of a standard treatment regimen as study limitations.
“Our findings support the hypothesis that immunotherapy targeting the PD-1/PD-L1 pathway may benefit patients with DLBCL, although therapeutic responses may differ according to PD-L1 expression profiles,” the researchers wrote. “Future investigations and clinical trials are needed to develop therapeutic strategies based on PD-L1–induced tumor immune evasion mechanisms and modulation of host immune responses.” – by Cameron Kelsall
Disclosure: The researchers report no relevant financial disclosures.
Perspective
Back to Top
The PD-1 receptor on T cells is a negative regulator of T-cell function and its ligands PD-L1 and PD-L2 provide an inhibitory signal when T cells encounter antigens. PD-1 is upregulated when T cells are activated, but some PD-1–positive T cells may actually be functionally exhausted and unable to respond to immune stimuli. PD-1 expressing T cells have been identified within the tumor microenvironment in diffuse large B-cell lymphoma (DLBCL) and previous studies have shown that soluble PD-L1 is associated with a poor prognosis in DLBCL. PD-L2 has been shown to be upregulated in some diffuse large B-cell lymphomas and the malignant B cell appears to utilize overexpression of PD-L1 and PD-L2 as a mechanism to suppress the T cell-mediated antitumor immune response. Therapeutically, blocking the interaction between PD-1 and its ligands has shown clinical promise and a clinical trial utilizing pidilizumab (CT-011, CureTech) to block PD-1 signaling showed activity in DLBCL.
In a recently reported study by Kiyasu and colleagues, it was shown that not only was tumor expression of PD-L1 associated with a poor prognosis in DLBCL, but the location of PD-L1 expression was of significant importance. PD-L1 expression was present on malignant B cells and the expression on lymphoma cells was confirmed by co-expression of Pax-5. However, PD-L1 expression was also seen on infiltrating non-malignant cells. The prognostic significance of PD-L1 expression was then tested by comparing the outcome of patients with PD-L1 expression in the tumor to those in whom no expression of PD-L1 was seen. They found that expression of PD-L1 on malignant B cells in DLBCL predicted for a poorer outcome, but there was no significant difference in OS between patients where PD-L1 positivity was on the intratumoral microenvironment cells compared with patients with tumors which had no staining for PD-L1. The expression of PD-L1 on the malignant lymphoma cells, but not the expression on non-malignant cells, remained statistically significant for OS when tested in a multivariate analysis that included other variables associated with prognosis.
This finding is important in view of the fact that clinical trials are in progress utilizing agents which block either PD-1 or PD-L1. Current studies do not discriminate between patients where PD-L1 expression is on the tumor cells compared with patients where PD-L1 is on the non-malignant cells in the tumor microenvironment. It may well be that the function of PD-L1 on these two cell types is different and therefore blockade of PD-L1/ PD-1 signaling may result in a different outcome based on the cells involved. Currently, a reliable biomarker to predict response to PD-1–directed therapy in lymphoma is not available. It might therefore be useful to assess whether PD-L1 expression on malignant B cells or in the microenvironment is a predictor of benefit when PD-1 or PD-L1 blockade is utilized. Additional studies will need to be done to assess this.
References:
Ahearne MJ, et al. Virchows Arch. 2014;doi:10.1007/s00428-014-1615-5.
Andorsky DJ, et al. Clin Cancer Res. 2011;doi:10.1158/1078-0432.CCR-10-2660.
Armand P, et al. J Clin Oncol. 2013;doi:10.1200/JCO.2012.48.3685. .
Chen BJ, et al. Clin Cancer Res. 2013;doi:10.1158/1078-0432.CCR-13-0855.
Rossille D, et al. Leukemia. 2014;doi:10.1038/leu.2014.137.
Shi M, et al. Am J Surg Pathol. 2014;doi:10.1097/PAS.0000000000000297.