Programmed Death-Ligand 1 Heterogeneity and Its Impact on Benefit From Immune Checkpoint Inhibitors in NSCLC

Lingzhi Hong(The University of Texas MD Anderson Cancer Center), Marcelo V. Negrão(The University of Texas MD Anderson Cancer Center), Seyedeh Dibaj(The University of Texas MD Anderson Cancer Center), Runzhe Chen(The University of Texas MD Anderson Cancer Center), Alexandre Reuben(The University of Texas MD Anderson Cancer Center), Jadi M. Bohac(The University of Texas MD Anderson Cancer Center), Xiaoke Liu(The University of Texas MD Anderson Cancer Center), Ferdinandos Skoulidis(The University of Texas MD Anderson Cancer Center), Carl M. Gay(The University of Texas MD Anderson Cancer Center), Tina Cascone(The University of Texas MD Anderson Cancer Center), K. G. Mitchell(The University of Texas MD Anderson Cancer Center), Hai T. Tran(The University of Texas MD Anderson Cancer Center), Xiuning Le(The University of Texas MD Anderson Cancer Center), Lauren A. Byers(The University of Texas MD Anderson Cancer Center), Boris Sepesi(The University of Texas MD Anderson Cancer Center), Mehmet Altan(The University of Texas MD Anderson Cancer Center), Yasir Y. Elamin(The University of Texas MD Anderson Cancer Center), Frank V. Fossella(The University of Texas MD Anderson Cancer Center), Jonathan M. Kurie(The University of Texas MD Anderson Cancer Center), Charles Lu(The University of Texas MD Anderson Cancer Center), Frank E. Mott(The University of Texas MD Anderson Cancer Center), Anne S. Tsao(The University of Texas MD Anderson Cancer Center), Waree Rinsurongkawong(The University of Texas MD Anderson Cancer Center), Jeff Lewis(The University of Texas MD Anderson Cancer Center), Don L. Gibbons(The University of Texas MD Anderson Cancer Center), Bonnie S. Glisson(The University of Texas MD Anderson Cancer Center), George R. Blumenschein(The University of Texas MD Anderson Cancer Center), Emily Roarty(The University of Texas MD Anderson Cancer Center), P. Andrew Futreal(The University of Texas MD Anderson Cancer Center), Ignacio I. Wistuba(The University of Texas MD Anderson Cancer Center), Jack A. Roth(The University of Texas MD Anderson Cancer Center), Stephen G. Swisher(The University of Texas MD Anderson Cancer Center), Vassiliki A. Papadimitrakopoulou(The University of Texas MD Anderson Cancer Center), John V. Heymach(The University of Texas MD Anderson Cancer Center), J. Jack Lee(The University of Texas MD Anderson Cancer Center), George R. Simon(The University of Texas MD Anderson Cancer Center), Jianjun Zhang(The University of Texas MD Anderson Cancer Center)
Journal of Thoracic Oncology
May 7, 2020
Cited by 190Open Access
Full Text

Abstract

INTRODUCTION: Programmed death-ligand 1 (PD-L1) expression may vary in different disease sites and at different time points of the disease course. We aimed to investigate PD-L1 heterogeneity and its usefulness as a predictive value for immune checkpoint inhibitor (ICI) therapy in patients with NSCLC. METHODS: PD-L1 expression was analyzed in 1398 patients with NSCLC. The predictive value of PD-L1 for ICIs in 398 patients with metastatic NSCLC was assessed. RESULTS: PD-L1 was significantly associated with biopsy sites (p = 0.004). Adrenal, liver, and lymph node (LN) metastases had the highest PD-L1 expression as a continuous variable and at 1% or 50% cutoff. PD-L1 expression was lower in bone and brain metastases. Among 112 patients with two specimens tested, 55 (49%) had major changes in PD-L1 falling into different clinically relevant categories (<1%, 1%-49%, ≥50%) at different time points. Previous ICI therapy was associated with significant decrease in PD-L1 compared with treatment-naive counterparts (p = 0.015). Patients with metastatic NSCLC treated with ICI (n = 398) were divided into three cohorts on the basis of biopsy sites: lung (n = 252), LN (n = 85), and distant metastasis (n = 61). Higher PD-L1 in lung or distant metastasis specimens was associated with higher response rate, longer progression-free survival, and overall survival. However, PD-L1 in LN biopsies was not associated with either response or survival. CONCLUSIONS: PD-L1 varies substantially across different anatomical sites and changes during the clinical course. PD-L1 from different biopsy sites may have different predictive values for benefit from ICIs in NSCLC.


Related Papers

No related papers found

Powered by citation graph analysis