Classification for cancers revealed based on immune response

In the journal Immunity, researchers co-led by UNC Lineberger's Benjamin Vincent, MD, reported their results of the analysis of immune responses in approximately 10,000 tumor samples. They revealed a new method for categorizing cancers based on their findings about the characteristics of the immune “microenvironment,” which they discovered using different types of genomic analysis, including RNA-sequencing.

Classification for cancers revealed based on immune response click to enlarge UNC Lineberger's Benjamin Vincent, MD, and Jonathan Serody, MD.

In a project that makes unprecedented strides in the understanding of how the immune system responds to cancer, researchers from the University of North Carolina Lineberger Comprehensive Cancer Center and their colleagues from The Cancer Genome Atlas analyzed thousands of cancer samples across multiple different tumor types to classify them based on the body’s defense response.

In the journal Immunity, researchers reported their results of the analysis of immune responses in approximately 10,000 tumor samples. They revealed a new method for categorizing cancers based on their findings about the characteristics of the immune “microenvironment,” which they discovered using different types of genomic analysis, including RNA-sequencing.

“There are similarities in the underlying immune microenvironments of cancers that are much broader and deeper than if you had just looked at the cancer tissue types and known subtypes,” said the study’s co-corresponding author Benjamin Vincent, MD, of UNC Lineberger and an assistant professor in the UNC School of Medicine Division of Hematology/Oncology. “This hopefully will have therapeutic implications in that it will help us determine how patients will do on different immune-modulating treatments.”

For the study, researchers analyzed the immune “microenvironment” in the tumor samples, which is the collection of non-cancerous cells found in a tumor sample that are part of the immune system. Using different types of sequencing methods, including analyzing DNA, gene expression, and other genomic approaches, they identified different types of immune cells and signals within tumors.

They found that not all cancers are the same in terms of the types of immune system-related features that are present in the tumor environment. Based on their analysis, they determined there were six different subtypes of immune microenvironments.

One microenvironment, known as “wound healing,” has an enrichment of genes linked with blood vessel growth, which can help deliver nutrients to tumors, and high rates of growth of tumor cells. This type was seen frequently in colorectal cancers and lung squamous cell carcinoma.

The “inflammatory” subtype has higher expression of genes that promote inflammation, which helps to draw the immune system to the site, and these tumors showed low to moderate tumor cell proliferation. This was found most frequently in kidney cancer, prostate cancer, pancreatic cancer, and papillary thyroid carcinoma.

The “immunologically quiet” microenvironment, typically associated with a cancer of the nervous system, shows the lowest level of lymphocytes and the highest level of immune cells that can ingest damaged, dead or diseased cells.

“Genomics has provided us with a large number of tools that allow us to classify histologically different cancers into similar groups, which TCGA has shown,” said Jonathan Serody, MD, UNC Lineberger’s associate director for translational sciences, the Elizabeth Thomas Professor of Medicine, Microbiology and Immunology in the UNC School of Medicine and a paper co-author. “Here, we have demonstrated there is a whole new approach to cancer classification based on tumor cell extrinsic characteristics. Even more interesting is that this new classification is associated with cross-talk between the genetic makeup of the tumor and the immune response. These findings shed a lot of light on the interaction of the tumor and immune system and provide a rational approach for immune-targeted therapy.”

The researchers report that mapping the molecular complexity of tumors based on their immune responses is “critical” both to understanding the basic tumor biology and designing treatment strategies that can improve the immune system’s ability to recognize and attack tumors. Their findings suggest a potential new classification that could drive different approaches to treatment.

"Treatments that work by unleashing the immune system against tumors normally hidden from the body’s defenses have revolutionized care for melanoma and other diseases,” Vincent said “Not all patients respond, however. The next step will be to determine if patients respond differently to immunotherapies based on these subtypes, which will allow oncologists to further personalize immunotherapy choice to benefit patients.”

In addition to Vincent and Serody, the paper’s other authors were Vesteinn Thorsson, David L. Gibbs, Christopher L. Plaisier, Varsha Dhankani, Sheila M. Reynolds, and Ilya Shmulevich, the Institute for Systems Biology; Scott D. Brown, Reanne Bowlby and Robert A. Holt, Canada’s Michael Smith Genome Sciences Centre; Denise Wolf, University of California, San Francisco; Dante S. Bortone, Joel S. Parker and Lisle Mose, UNC Lineberger; Tai-Hsien Ou Yang and Dimitris Anastassiou, Columbia University Department of Systems Biology and Department of Electrical Engineering; Eduard Porta-Pardo, Barcelona Supercomputing Centre; Galen Gao and Andrew D. Cherniak, Eli and Edythe L. Broad Institute of the Massachusetts Institute of Technology and Harvard University; James A. Eddy, Sage Bionetworks; Elad Ziv, Helen Diller Family Comprehensive Cancer Center at the University of California, San Francisco; Aedin C. Culhane, Chandra Sekhar Pedamallu, Susan Bullman,  and Toni K. Choueiri, Dana-Farber Cancer Institute; Evan O. Paull and Andrea Califano, Irving Cancer Research Center; I.K. Ashok Sivakumar, Johns Hopkins University; Andrew J. Gentles, Stanford University; Raunaq Malhotra of Seven Bridges; Farshad Farshidfar, University of Calgary; Antonio Colaprico, Universite libre de Bruxelles; Nam Sy Vo, Yuexin Liu, Arvind Rao, Ken Chen,  John N. Weinstein, and Alex J. Lazar, MD Anderson Cancer Center; Jianfang Liu, Hai Hu, Chan Soon-Shion Institute of Molecular Medicine at Windber; Janet Rader, Medical College of Wisconsin; Davide Bedognetti, Sidra Medical and Research Center; Alexander Krasnitz, Cold Spring Harbor Laboratory; Tathiane M. Malta and Houtan Noushmehr, Henry Ford Hospital; Akinyemi I. Ojesina, University of Alabama at Birmingham; Joel Saltz, Stony Brook Medicine; Charles E. Rabkin, NCI; The Cancer Genome Atlas Network; Elizabeth G. Demicco, Mount Sinai Hospital; and Mary L. Disis, University of Washington.

The research was funded in by the National Cancer Institute.