The tumor-immune interface of human lung cancer

Lung cancer is one of the most common human cancers and the one with the highest mortality rate worldwide. Immunotherapies that use the bodys own immune system to fight cancer have shown unprecedented clinical effects for different cancer types and are now approved also for the treatments of lung cancer. However, only a limited fraction of patients respond to immunotherapy, suggesting the persistence of different mechanisms of immune evasion, i.e. inhibitory strategies used by the tumors to keep the immune system in check. Tumors are not merely masses of malignant cells, but complex and dynamic networks of different immune and tumor cells that interact in the tumor microenvironment. Next- generation sequencing (NGS) technologies can provide the information necessary to portray the different layers of interaction of this tumor-immune interface. However, as of today, there are no computational tools available to reconstruct the tumor-immune interface from NGS data and there is an urgent need to develop novel bioinformatics methods and analytical pipelines. This project is aimed at developing bioinformatics methods to reconstruct the different facets of the tumor-immune interface in human lung cancer using NGS data. Among these methods, major efforts will be dedicated to the development and validation of a novel bioinformatics tool for the quantification of the different immune cell types infiltrating lung tumors. Within this project, NGS data and digital pathology images will be generated from ten lung tumors to develop and optimize the method. After successful validation, the methods will be used to analyze NGS data from fifty lung cancer patients subjected to immunotherapy with checkpoint blockers, so to reconstruct a high-resolution map of the tumor-immune interface. The results of this analysis will reveal the mechanisms of immune evasion that hamper antitumor immune responses and might lay the foundations for the rational design of more effective therapies for lung cancer.

Lung cancer is a leading cause of death and a major public health problem worldwide. In recent years, immunotherapies that leverage the body's own immune system to fight cancer have emerged as powerful treatments. Immunotherapy has transformed the practice of medical oncology, showing unprecedented clinical results and bringing new hope to patients with different cancer types. However, the majority of cancer patients still do not respond to immunotherapy and the mechanisms of resistance remain elusive. Sequencing technologies can now generate rich data from tumor samples that can be mined computationally to investigate the tumor microenvironment and the interactions taking place between tumor and immune cells (i.e., the "tumor-immune interface"). However, the possibility to use this data to inform immunotherapy depends upon the development of advanced computational techniques that are able to reconstruct from it the mechanisms taking place at the tumor-immune interface. The aim of this study was to develop new computational methods to chart the tumor-immune interface in lung cancer. We developed an extensive set of computational tools to reconstruct from patients' sequencing data all the different facets of the tumor-immune interface: from the cellular composition of the tumor microenvironment to the intra- and inter-cellular mechanisms that orchestrate anticancer immune responses. We have then applied this unique toolkit to the analysis of tumor sequencing data from patients with lung cancer and other cancer types to characterize their tumor-immune cell interface. Based on this, we have developed a machine learning method to predict patients' response to immunotherapy from sequencing data generated pre-therapy from tumor samples. By using data from different cohorts of patients, we could demonstrate that our approach can anticipate which patients are more likely to respond to immunotherapy and further pinpoint which components of the tumor-immune interface are associated with the success or defeat of this therapeutic approach. The resources and insights generated by this project can inform immunotherapy by identifying which patients are most likely to benefit from immunotherapy, but also provide the means to rationally design novel combinations therapies with higher clinical efficacy.