The Crohn's-like lymphoid reaction (CLR) to colorectal cancer (CRC), a CRC-specific ectopic lymphoid reaction, is thought to play an important role in the host response to CRC. CLR is characterized by peritumoral lymphocytic aggregates that are found at the advancing edge of the tumor. Spatial and molecular characterization of CLR within the tumor microenvironment (TME) have uncovered a spectrum of peritumoral lymphoid aggregates with varying levels of organization and maturation. In early stages of CLR development, CD4+ T-cells cluster predominantly with mature antigen presenting dendritic cells. As CLR matures, increasing numbers of B-cells, as well as follicular dendritic cells are recruited to create lymphoid follicles. When highly organized, CLR resembles functional tertiary lymphoid structures (TLS), allowing for lymphocyte recruitment to the TME and promoting a tumor-specific adaptive immune response. CLR has been consistently associated with favorable prognostic factors and improved survival among CRC patients, often providing more prognostic information than current clinical staging systems. However, consensus is lacking regarding CLR scoring and it is not clinically assessed or reported. Differences between CLR and other cancer-associated lymphoid structures exist both in primary and metastatic disease, underscoring the need to characterize organ-specific TLS. Further research is needed to explore the role of CLR in predicting response to immunotherapy and to leverage CLR to promote immunotherapeutic strategies in CRC.

Microsatellite instability (MSI) is a molecular marker of deficient DNA mismatch repair (dMMR) that is found in approximately 15% of colorectal cancer (CRC) patients. Testing all CRC patients for MSI/dMMR is recommended as screening for Lynch Syndrome and, more recently, to determine eligibility for immune checkpoint inhibitors in advanced disease. However, universal testing for MSI/dMMR has not been uniformly implemented because of cost and resource limitations. Artificial intelligence has been used to predict MSI/dMMR directly from hematoxylin and eosin (H&E) stained tissue slides. We review the emerging data regarding the utility of machine learning for MSI classification, focusing on CRC. We also provide the clinician with an introduction to image analysis with machine learning and convolutional neural networks. Machine learning can predict MSI/dMMR with high accuracy in high quality, curated datasets. Accuracy can be significantly decreased when applied to cohorts with different ethnic and/or clinical characteristics, or different tissue preparation protocols. Research is ongoing to determine the optimal machine learning methods for predicting MSI, which will need to be compared to current clinical practices, including next-generation sequencing. Predicting response to immunotherapy remains an unmet need.