Predictive modeling for singlet oxygen generated during Rose Bengal Photodynamic Antimicrobial Therapy

Abstract

Purpose : Rose Bengal Photodynamic Antimicrobial Therapy (RB-PDAT) has been shown to effectively treat antimicrobial resistant and atypical infectious keratitis. RB-PDAT works by generating antimicrobial singlet oxygen (¹O₂) by exciting Rose Bengal (RB) photosensitizer with 525 nm light. While the cumulative ¹O₂ dose distribution generated during RB-PDAT is currently unknown, this project aims to create a predictive model to calculate this distribution under clinical conditions. This model will eventually enable optimization of ¹O₂ dose by optimizing RB-PDAT treatment parameters.

Methods : A chemical kinetics model was previously developed (Wang et al., 2010) for modeling ¹O₂ dose distributions in tumors treated with photodynamic therapy. To describe ground-state oxygen (O₂) supply specific to the cornea, the oxygen supply term was adapted to include the model described by Larrea & Büchler (2009) and Castillo et al. (2014). RB distribution was modeled using Fick’s second law, and 525 nm excitation light distribution by a time-dependent application of Beer’s law. Rate constants for RB diffusion were then derived from experimental data describing RB corneal penetrance (Naranjo, Pelaez, et al., 2019; Peterson et al., 2018; Wertheimer et al., 2019). Lateral variation of all species was assumed to be minimal to allow for one-dimensional modeling. The resulting system of partial differential equations was solved numerically in MATLAB. Cumulative ¹O₂ dose was calculated for RB-PDAT during typical conditions (6 mW/cm², 5.4 J/cm², 0.1% RB), as well as for conditions of pulsed and increased light dose, and addition of 90% supplemental O₂.

Results : Under typical RB-PDAT conditions, calculated O₂ decreases to approximately 0 beyond 10 µm depth in the first 1 min of treatment. In this scenario, ¹O₂ dose distribution was calculated to be limited to the first 10 µm of the cornea. Adding supplemental O₂ increases ¹O₂ dose depth by approximately 5–10 µm. Increasing the light dose from 5.4 J/cm² to 10 J/cm² while adding 1s on, 1s off pulsing increases calculated ¹O₂ surface dose 1.85x. Adding supplemental O₂ to the 10 J/cm² scenario, ¹O₂ surface dose is 1.7x the dose seen in the 5.4 J/cm² condition.

Conclusions : We demonstrate a proof-of-concept predictive model for ¹O₂ dose generated during RB-PDAT, capable of testing a range of experimental parameters. This model can be used for optimizing RB-PDAT clinical efficacy.

This is a 2021 ARVO Annual Meeting abstract.