Monte Carlo based light propagation models to improve efficacy of biophotonics based therapeutics of hollow organs and solid tumours including photodynamic therapy and photobiomodulation (Conference Presentation)

The majority of denovo cancers are today being diagnosed in low and middle-income countries, which often lack resources and a range of therapeutic options. Minimally invasive therapies such as Photodynamic Therapy (PDT) and photobiomodulation (PBM) could become treatment options, albeit widespread acceptance is hindered by multiple factors ranging from training of surgeons in optical therapeutic techniques, lack of easily usable treatment optimizing tools and prediction of the anticipated treatment outcome. Based on the publicly available FullMonte software in combination with other open source image processing tools, a work plan is proposed that allows for personalized treatment planning. Starting with, generating 3D in silico models, execution of the Monte Carlo simulation and presentation of the 3D fluence rate distribution a treatment procedure is presented. Calculation of the forward solution of photon transport in biological tissues is executed in less than a minute for 3D models comprising 106 tetrahedral elements. The ability of the program to find optimal source placements was demonstrated for in silico brain tumour models for solid tumours. In hollow organs the impact of non-isotropic cavities is demonstrated on bladder cancer patient data. For photodynamic therapy treatment optimization, the process considers the selective uptake ratio of the photosensitizer between the target, host tissues and organs at risk and establish PDT sensitivities of these tissues based on the photodynamic threshold values. Tumours are assigned only a minimum required dose, whereas host and organs at risk a maximum permissible dose. For PBM the target and the host tissue are assigned minimum and maximum permissible dose due to the well documented biphasic response effect in PBM. For PDT sources of errors are uncertainties in the contouring whereas for PBM the depth of the actual target in the tissue is unknown and need often to be estimated based on body mass Index, and other morphometric parameters. Both photo therapeutic applications suffer from unknown tissue optical properties. Hence, the proposed workflow includes a perturbation of the planning tissue optical properties, uncertainties in the photon source placement and contouring errors, to validate the invariance of the attained solution against these unknowns. This requires also the need to determine the patients actual tissue optical properties at the onset of therapy, which in turn can only be achieved when the appropriate placement of invasive or diffuse reflective sensors is provided for. Hence, the planning process needs to include also identification of the most responsive positions for these sensors in the planning volume.