Works in progress for cancer early screening from blood testing

What is it about?

We developed a system that uses flow cytometry and artificial intelligence to discern between blood cells and circulating tumour cells (CTCs), to be used in the future in the framework of liquid biopsy. We rely on a cutting-edge technique called holographic flow cytometry to collect images and the associated biophysical information from cells in high-throughput stream inside the so-called Lab-on-a-Chip devices. This technique allows obtaining quantitative information about the sample rather that plain images, which is used to boost artificial intelligence-based classification techniques with optimal performance. In liquid biopsy, the biggest challenge is to distinguish between the white blood cells and CTCs, since they are comparable in size. In particular, we classified with high accuracy monocytes vs. model cells from ovarian cancer, and we also minimized the flase negatives, which is a non-neglectable advantage considering the importance of a correct classification for prescreening programs.

Featured Image

Photo by Logan Moreno Gutierrez on Unsplash

Photo by Logan Moreno Gutierrez on Unsplash

Why is it important?

CTCs identification is of primary concern in the perspective of liquid biopsy used for cancer diagnosis, where detection of tumour-derived circulating material might lead to an early detection in a less-invasive fashion for the patient. For instance, in the future this analysis could be part of routine blood screening programs, as an alternative to the more invasive tissue biospy. Considering the high rarity of this material in the bloodstream, having really accurate classifiers is a crucial need, together with the minimization of the false negative classification results that could provoke the “loss” of a rare CTC. Above all, this method is completely marker-free. So far, cancer cells identification has been based on the the search for cancer-specific antigens, i.e. one has to know specifically the mutation to look for. Conversely, with the aim of implementing early screening programs, marker-free blind identification of cancer cells without knowing any prior information about the subjects/patients is a priceless added-value brought by the novel AI-aided holographic method.

Perspectives