When cells divide, mistakes happen. However, an intricate surveillance system has evolved to detect and eliminate anomalous cells before they become detrimental to the host organism. In cancer, abnormal cells manage to escape the immune system and grow uncontrollably. In this sense, cancer can be considered as an oversight of the immune system, as immune escape is a defining feature of clinically detectable cancers. The role of the immune system in fighting cancer is becoming increasingly indisputable as our understanding of its underlying mechanisms expand owing to technological advances in genomics, cancer biology, and computational sciences. In particular, significant research effort is undertaken in the field of cancer immunotherapy, where the immune system is stimulated to recognize and attack cancerous cells. In this thesis, I investigate certain aspects of the immune system in the context of cancer by computationally reconstructing and analyzing intratumoral B and T cell receptors.

Applying a novel immune cell receptor profiling protocol to original single-cell RNA sequencing (scRNA-seq) data obtained from melanoma patients, I present a complete computational reconstruction of intratumoral immune receptors in this cancer type. The scRNA-seq results are consistent with the presence of an ongoing intratumoral immune response, likely involving tertiary lymphoid structures and the cooperation between B and T cells.

Additionally, using a dataset of paired tumor biopsies collected pre- and post-treatment, I show that B cell infiltration increases after immunotherapy in pancreatic and colorectal cancer. This thesis includes the sequences of the most clonally expanded intratumoral antibodies expressed in these biopsies which I computationally reconstructed from bulk RNA sequencing reads.

Furthermore, by combining scRNA-seq and immune cell receptor profiling of samples collected from a novel mouse model, I present a comprehensive statistical analysis of gene expression in clonal tumor-reactive T cells. I also show the distribution of tumor-reactive clones across the tumor and spleen. This study forms a first proof-of-principle effort for the in-depth assessment of tumor-reactive T and B cell clones, in-vivo, and paves the way for further, more extensive experiments.