BackgroundBC screening by mammography is associated with a significant reduction in mortality (19 % overall reduction of the relative risk), however with significant limitations and debatable cost-effectiveness. Screening individuals for cancer using liquid biopsies (LB) represents an unmet need. We report the first prospective “proof of concept” study using the QO technology combined with an AI approach using the concept of “MPnV” applied to BC detection. MethodsQO (femto/atto-second infrared laser spectroscopy) on LB is a simple, non-invasive and reproducible method allowing to identify individualized molecular spectra (MS). These highly detailed MS can be correlated to physiological or pathological changes allowing detection and translation of differences. Integrated into a super-computational approach using a non-hierarchical deep data mining strategy (MPnV concept), MS could discriminate individuals with and without BC. The plasma of 68 controls and 27 BC patients, accrued at the King Saud University BC screening program, Riyadh, Saudi Arabia (KSA), were studied by QO (Max Planck Institute of Quantum Optics /Ludwig Maximillian University Munich, Garching, Germany). The MS were analysed on the Shaheen II supercomputer at King Abdallah University for Science and Technology (KAUST), KSA, in order to generate comparative algorithms. ResultsThe use of special feature selection, followed by class analysis allowed to differentiate profiles between the two groups with a sensitivity of 97% and a specificity of 72% (variables n=1,100). A more in-depth analysis led to 99% sensitivity, but with a lower specificity of 64%. Further analysis of the series, using an age-matched approach led to 97% sensitivity with 98% specificity in differentiating women with or without BC. ConclusionsThese results warrant a large scale prospective validation BC screening trial (ongoing) and “proof of concept trials” in other frequent cancers, in particular those without existing screening programs. Editorial acknowledgementProf. Ferenc Krausz, Director, and Dr. Mihaela Zigman, Leader of the Broadband Infrared Diagnostics, Max Planck Institute of Quantum Optics (MPQ), Faculty of Physics at Ludwig-Maximilians Universität München (LMU), Garching, Germany. Legal entity responsible for the studyJean Marc Nabholtz, Oncology Centre, King Saud University, Riyadh, Saudi Arabia. Funding1. Oncology Centre, King Saud University Medical City, King Saud University, Saudi Arabia. 2. Max-Planck-Institut für Quantenoptik and Faculty of Physics at Ludwig-Maximilians Universität München, Garching, Germany. 3. Computational Bioscience Research Centre, King Abdallah University for Science and Technology (KAUST), Thuwal, Saudi Arabia. DisclosureM.R.K. Bahadoor: Travel / Accommodation / Expenses, ASCO Participation Funded: Pfizer; Advisory / Consultancy, Pancreas Expert Opinion Advisory: Baxter. All other authors have declared no conflicts of interest.