Longitudinal monitoring of chemotherapy response in preclinical oncology models using Spatial Frequency Domain Imaging (SFDI)
Date
2020
DOI
Authors
Tabassum, Syeda
Version
OA Version
Citation
Abstract
Methods for frequent non-invasive surveillance of the in-vivo tumor state may assist in detecting whether a patient is responding or developing resistance at an early stage of treatment. This would allow physicians to adapt and personalize treatment strategies in real-time. Multiple studies have demonstrated that clinical diffuse optical imaging (DOI), which can provide structural and hemodynamic profile of tumor, can reveal treatment induced changes that correlate strongly with patient response determined by pathology. While encouraging, there are many unknowns as to how DOI optical markers manifest for different treatment regimens and dosing. Additionally, a deeper understanding of the underlying cellular and molecular changes that contribute to DOI markers is needed to provide a mechanistic context to these clinical observations. This project addresses these issues at the preclinical level with Spatial Frequency Domain Imaging (SFDI). SFDI is a wide-field and non-invasive DOI modality that provides the same optical and hemodynamic information as the clinical tools and is more suitable for preclinical imaging, and so is the right tool for such exploratory study. To this end, the work presented in this dissertation was focused on establishing SFDI as a new preclinical monitoring tool for cancer. For the first time, the feasibility of using SFDI for frequent longitudinal monitoring of chemotherapy and targeted therapy efficacy in small animal oncology models was established. The SFDI optical property extraction accuracy was then improved in subcutaneous tumors by the development of a new two-layer Monte Carlo based inversion model. SFDI optical and functional metrics were then validated in the context of cellular and molecular correlates using immunohistochemistry. The treatment prediction ability of SFDI was also compared to simple tumor volume measurements in multiple tumor models. Finally, a custom-made LED-based SFDI system was developed to measure tissue water content in addition to hemodynamic features. Overall, this body of work helps to establish SFDI in the field of preclinical cancer treatment monitoring. Knowledge gained from this work may assist in the clinical translation of DOI tools as important feedback methods in the applications of treatment monitoring, drug testing, and personalization of treatment strategies.