Multiparametric profiling of non–small-cell lung cancers reveals distinct immunophenotypes
Patrick H. Lizotte, … , Mark A. Bittinger, Kwok-Kin Wong
Patrick H. Lizotte, … , Mark A. Bittinger, Kwok-Kin Wong
Published September 8, 2016
Citation Information: JCI Insight. 2016;1(14):e89014. https://doi.org/10.1172/jci.insight.89014.
View: Text | PDF
Clinical Medicine Immunology

Multiparametric profiling of non–small-cell lung cancers reveals distinct immunophenotypes

Abstract

BACKGROUND. Immune checkpoint blockade improves survival in a subset of patients with non–small-cell lung cancer (NSCLC), but robust biomarkers that predict response to PD-1 pathway inhibitors are lacking. Furthermore, our understanding of the diversity of the NSCLC tumor immune microenvironment remains limited.

METHODS. We performed comprehensive flow cytometric immunoprofiling on both tumor and immune cells from 51 NSCLCs and integrated this analysis with clinical and histopathologic characteristics, next-generation sequencing, mRNA expression, and PD-L1 immunohistochemistry (IHC).

RESULTS. Cytometric profiling identified an immunologically “hot” cluster with abundant CD8+ T cells expressing high levels of PD-1 and TIM-3 and an immunologically “cold” cluster with lower relative abundance of CD8+ T cells and expression of inhibitory markers. The “hot” cluster was highly enriched for expression of genes associated with T cell trafficking and cytotoxic function and high PD-L1 expression by IHC. There was no correlation between immunophenotype and KRAS or EGFR mutation, or patient smoking history, but we did observe an enrichment of squamous subtype and tumors with higher mutation burden in the “hot” cluster. Additionally, approximately 20% of cases had high B cell infiltrates with a subset producing IL-10.

CONCLUSIONS. Our results support the use of immune-based metrics to study response and resistance to immunotherapy in lung cancer.

FUNDING. The Robert A. and Renée E. Belfer Family Foundation, Expect Miracles Foundation, Starr Cancer Consortium, Stand Up to Cancer Foundation, Conquer Cancer Foundation, International Association for the Study of Lung Cancer, National Cancer Institute (R01 CA205150), and the Damon Runyon Cancer Research Foundation.

Authors

Patrick H. Lizotte, Elena V. Ivanova, Mark M. Awad, Robert E. Jones, Lauren Keogh, Hongye Liu, Ruben Dries, Christina Almonte, Grit S. Herter-Sprie, Abigail Santos, Nora B. Feeney, Cloud P. Paweletz, Meghana M. Kulkarni, Adam J. Bass, Anil K. Rustgi, Guo-Cheng Yuan, Donald W. Kufe, Pasi A. Jänne, Peter S. Hammerman, Lynette M. Sholl, F. Stephen Hodi, William G. Richards, Raphael Bueno, Jessie M. English, Mark A. Bittinger, Kwok-Kin Wong

×

Figure 7

Limited CD8+ T cell markers predict NSCLC immunophenotypes.

Options: View larger image (or click on image) Download as PowerPoint
Limited CD8+ T cell markers predict NSCLC immunophenotypes. (A) NSCLC ca...
(A) NSCLC cases were reanalyzed using t-distributed stochastic neighbor embedding (t-SNE) algorithm based on only 3 parameters: percentage of CD8+ T cells of CD3+ lymphocytes, percentage of TIM-3 expression on CD8+ T cells, and percentage of PD-1 expression on CD8+ T cells. The structure of “hot” and “cold” clusters matched what was observed for multiparameter clustering. (B) Proposed model of immunotherapy-favorable immunophenotype of NSCLC based on limited T cell–intrinsic factors.