Article Abstract

Characterization of PD-L1 expression and its prognostic value in patients with ovarian cancer

Authors: Bi-Heng Cheng, Hua Liang, Tao Jiang, Jian-Hua Chen, Gao-Hua Wang

Abstract

Background: The results in regards to the role in which programmed cell death ligand 1 (PD-L1) plays in the prognostic expression of ovarian cancer remains controversial.
Methods: Aiming to clarify the demographic incidence rate and the prognostic value of the PD-L1 expression in the patients with ovarian cancer, we performed a systematic search on PubMed, Web of Science, EMBASE and the Cochrane library to identify the relevant studies dating up to November 2017.
Results: We identified a total of 7 studies comprising 1,002 patients that met our criteria. Within these seven studies, PD-L1 was expressed in about 46% of patients with ovarian cancer (95% CI, 0.28–0.65; I2=97.9%; P<0.001). A pooled analysis showed that the positive PD-L1 expression was significantly associated with the prolonged OS (HR =0.63, P<0.001) in the patients with ovarian cancer. Subgroup analysis indicated that the positive rate was significantly higher in the patients suffering with the high-grade serous ovarian cancer (HGSOC) than the patients suffering with the less common ovarian histopathology (LOCH) (58% vs. 38%, P<0.05), while being comparable between the different ethnicities, i.e., East-Asian vs. Caucasian. The PD-L1 expression was significantly associated with the prolonged OS among the Caucasian population group rather than the East-Asian population group. Interestingly, there was a positive PD-L1 expression which predicted a longer OS time in the patients with HGSOC, whereas it significantly predicted a poor OS time for the patients with LOCH.
Conclusions: This meta-analysis found PD-L1 expressed in the part of the patients group with ovarian cancer to have a significantly higher rate in HGSOC than LOCH. As a result of this higher rate, PD-L1 expression could be used as a biomarker for patients of ovarian cancer, while it had a converse relationship between HGSOC and LOCH.