The role of PD-L1 expression as a predictive biomarker in advanced renal cell carcinoma: a meta-analysis of randomized clinical trials.

Date 30 September 2019
Event ESMO 2019 Congress
Session Poster Display session 3
Topics Renal Cell Cancer
Translational Research
Presenter Alberto Carretero-Gonzalez
Citation Annals of Oncology (2019) 30 (suppl_5): v25-v54. 10.1093/annonc/mdz239
Authors A. Carretero-Gonzalez1, I. Martín Sobrino2, I. Sáez Sanz2, D. Castellano Gauna1, G.A. De Velasco Oria de Rueda1
  • 1Medical Oncology, University Hospital 12 De Octubre, 28041 - Madrid/ES
  • 2School Of Medicine, University Hospital 12 De Octubre, 28041 - Madrid/ES



Immune checkpoint inhibitors (ICIs) are beneficial in a subset of metastatic renal cell carcinoma (mRCC) patients. However, no biomarker has been shown to be useful to select which patient benefits and the role of programmed death-ligand 1 (PD-L1) expression on tumor samples is controversial.


We assessed the potential role of PD-L1 expression according to Cochrane Collaboration’s Guidelines. Search of randomized clinical trials (RCTs) comparing ICIs (monotherapy or in combination with other therapies) to standard of care (SoC) in mRCC patients was performed. Trials must have included subgroup analysis evaluating the selected outcomes (progression-free survival-PFS- and overall survival-OS-) in different subsets of patients according to PD-L1 expression on tumor samples. Hazard ratios (HR) with confidence intervals (CI) were used as the measure of efficacy between groups.


A total of 3,720 patients (5 studies) were included (ICIs arm: 1,913 patients; SoC arm: 1,807 patients). Globally, PFS and OS results favored ICIs. Differential expression of PD-L1 on tumor samples could select a subset of patients who could benefit more in terms of PFS (those with higher levels; p-value for difference between subgroups: 0.003) but it did not seem to impact in OS results (p-value for difference: 0.29).Table: 129P

Value of PD-L1 expression as a predictive biomarker for ICIs. HR: Hazard ratio. CI: Confidence interval

Total populationHigh PD-L1Low PD-L1p-value for difference
PFSHR 0.72; 95% CI 0.58-0.90HR 0.63; 95% CI 0.51-0.77HR 0.96; 95% CI 0,79-1.160.003
OSHR 0.69; 95% CI 0.61-0.78HR 0.64; 95% CI 0.54-0.77HR 0.73; 95% CI 0.62-0.870.29


PD-L1 could represent a biomarker to test PFS in clinical trials but its value for OS is less clear.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.


Has not received any funding.


All authors have declared no conflicts of interest.