MEDPNC

In recent years, the lack of efficacy of conventional cancer treatments has become increasingly apparent to the medical community. Patients tumor heterogeneity has been established, and it is now known that if two patients suffer from cancer at the same anatomical site, from the molecular point of view these two tumors may be different. Thus, evaluation of the protein and biochemical processes in the body together with their relationship to the disease is crucial for deciding on the optimal treatment for the individual patient.

Targeting central proteins for suitable treatment is essential for the purpose of restoring the required balanced state in the tissue that will result in increasing patient survival.

Dr. Nataly Kravchenko-Balasha, from the Hebrew University of Jerusalem, has developed an approach based on the information-theoretic approach, that facilitates the detection of unbalanced protein networks and creates a personalized “barcode” for every patient, capable of predicting the efficacy of proposed treatments and their optimal combination, helping medical staff to select the best treatment for each patient, thus increasing successful treatment and reducing mortality. The approach is based on Surprisal Analysis, which enables the prediction of the direction of processes in the world of physics, chemistry and engineering. The analysis was first formulated in 1972 by Prof. Raphael Levine from the Hebrew University. This approach will significantly improve the accuracy of the design of patient-specific combination treatments.

Tumors are biological systems in which the balanced state has been disturbed due to genomic and environmental constraints.

MEDPNC utilizes thermodynamic-based surprisal analysis to detect and quantify the constraints for each protein in an individual patient by identifying how the protein expression levels deviate from their expected levels at the balanced state. These constraints induce unbalanced networks operating in the diseased tissue. The technology integrates cancer biology into a thermodynamic-based information-theoretical framework, aiming to model tumor biology by applying fundamental physical laws, and thus to find an order underlying inter-tumor heterogeneity and patient specific protein network rewiring.

Large-scale cancer datasets have been rapidly accumulating over the past years. MEDPNC uses this information from these data bases, enabling to determine which combination of FDA-approved drugs is best suited to treat the specific tumor.