The role of digital and integrative pathology for the detection of translocations: a narrative review

Background and Objective: Digital pathology represents an invaluable source of information and a long-term investment with high returns, with the possible deployment of artificial intelligence (AI) tools for both the clinical and research activity. Moreover, the rising nosological complexity of the oncologic diseases, e.g., lung cancer, is stressing the need of integration among different subspecialities (e.g., radiology, molecular biology, and immuno-oncology) for the final characterization of cancer. In this setting, digital pathology can play a pivotal role in the "integration" of these different competencies, and the application of AI for prognostic/predictive purposes can represent a further "third" revolution in pathology. The objective of the present review is to provide an updated overview of the possible role of digital and integrative pathology in detecting gene mutations and, especially, translocations in different types of tumors, focusing on the promising implications that this advancement can have in lung cancer characterization. Methods: A systematic literature search was conducted on different research engines (PubMed, IEEE Xplore, dblp, ACM digital library, Inspec) over a 15-year interval from January 1, 2006 to October 31, 2021 selecting only English-language articles to highlight the possible diagnostic and prognostic role of digital pathology tools in detecting gene mutations and, especially, translocations in different types of tumors. Key Content and Findings: Digital pathology tools already demonstrated a role in the detection of specific mutations and translocations in different types of cancer, both in a targeted (e.g., liver/thyroid carcinoma) and in an agnostic (e.g., MSI) setting. In lung cancer, AI showed the capability of highlighting specific subset of mutations (STK11, EGFR, FAT1, SETBP1, KRAS, and TP53) from whole slide imaging (WSI), with the translocation field representing a promising frontier, with some gene rearrangement (e.g., NTRK) more prone to be detected with this approach as demonstrated in other organs. Conclusions: Digital pathology is becoming a valuable tool to predict gene aberrations in different types of tumors, with a promising role in lung cancer. Digital and integrative tools for the detection of translocations are progressively changing the paradigm of molecular testing, contributing to the switch from a manual to a fully automated workflow, requiring further investigation.