AUTHOR=Varon Eli , Blumrosen Gaddi , Shefi Orit 

TITLE=A predictive model for personalization of nanotechnology-based phototherapy in cancer treatment

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.1037419

DOI=10.3389/fonc.2022.1037419

ISSN=2234-943X

ABSTRACT=A major challenge in radiation oncology is predicting and optimizing a clinical response on a personalized manner. Recently, nanotechnology-based cancer treatments are being combined with photodynamic therapy (PDT) and photothermal therapy (PTT). Machine learning predictive models can be used to optimize the clinical setup configuration, such as: laser radiation intensity, treatment duration, and nanoparticles features.  In this work we demonstrate a methodology to find the optimized treatment parameters for PDT and PTT by collecting data of in vitro cytotoxicity assay of PDT/PTT-induced cell death using a single nanocomplex. We examine three machine learning prediction models of regression, interpolation, and low degree analytical function to predict the laser radiation intensity and duration that maximize the treatment efficiency. To examine these prediction models accuracy, we built a  dedicated dataset for PDT, PTT, and a combined treatment that is based on cell death measurements after light radiation treatment, divided to training and test sets. The preliminary results show that all models offer sufficient performance with death rate error of 0.09, 0.15, and 0.12 for the regression, interpolation, and analytical function fitting. Nevertheless, the analytical function due to its simple form has a clinical application advantage that can be used for further sensitivity analysis of the treatment parameters on the performance. In all, the results of this work form a baseline for a future machine learning base personal prediction model in combined nanotechnology-based phototherapy cancer treatment.