Molecular Oncology (May 2019)

The immune microenvironment in non‐small cell lung cancer is predictive of prognosis after surgery

  • Åsa Kristina Öjlert,
  • Ann Rita Halvorsen,
  • Daniel Nebdal,
  • Marius Lund‐Iversen,
  • Steinar Solberg,
  • Odd Terje Brustugun,
  • Ole Christian Lingjærde,
  • Åslaug Helland

DOI
https://doi.org/10.1002/1878-0261.12475
Journal volume & issue
Vol. 13, no. 5
pp. 1166 – 1179

Abstract

Read online

The impact of the tumor immune microenvironment on overall survival in non‐small cell lung cancer (NSCLC) has been studied, but there is little information on its relevance for risk of relapse after surgery. Understanding more about the immune microenvironment in previously untreated NSCLC could help in identifying high‐risk patients and patients more likely to benefit from neoadjuvant/adjuvant immunotherapy. Here, we examined gene expression in 399 surgically derived NSCLC samples and 47 samples from normal lung, using Agilent microarray and RNA sequencing. In 335 of the tumor samples, programmed death‐ligand 1 (PD‐L1) expression was evaluated by immunohistochemistry. Gene expression was used to estimate content of immune cells and to calculate an immune score. Properties of the immune microenvironment, and its impact on prognosis, were compared in histological subgroups and gene expression subtypes. Tumors with an active immune microenvironment were found for both adenocarcinomas (AD) and squamous cell carcinomas (SCC). In AD, high immune score and high estimates of several immune cell types belonging to the adaptive immune system were associated with better progression‐free survival (PFS), while in SCC, no association between immune characteristics and PFS was found. The immune microenvironment, including PD‐L1 expression, and its impact on prognosis showed clear differences in AD and SCC gene expression subtypes. In conclusion, the NSCLC immune microenvironment is predictive of prognosis after surgery. Lung AD and SCC gene expression subtypes should be investigated as potential prognostic biomarkers in patients treated with immune checkpoint inhibitors.

Keywords