Optoacoustic (OA) imaging is an emerging low-cost hybrid imaging investigation/technique currently in clinical feasibility studies for breast cancer diagnosis and staging. The technique applies pulsed light to the tissue of interest where molecules absorb the light photons and generate acoustic pressure waves. The resulting acoustic responses are detected using ultrasound transducers and converted into images. Image contrast within a pixel is dependent on the relative concentration and absorption characteristics (i.e. spectrum) of the chromophores within the illuminated tissue. Thus, tissue responses from illumination using multiple wavelengths, chosen to reflect the differential absorption of oxy-/deoxy- and total haemoglobin, can be measured. In turn, these signals can be regarded as surrogate measures of tissue hypoxia and neoangiogenesis, hallmarks of cancer associated with adverse outcomes.

The aim of this PhD was to translate optoacoustic imaging into the breast clinic to try and fulfil some of the unmet clinical needs in breast cancer imaging using the imaging biomarker roadmap by O'Connor et al. Translation of this new technology to the clinical environment required extensive preparatory work, including the procurement and installation of a scanner prototype, liaison with UK regulatory bodies to secure ethical and MHRA approval, as well as several technical developments (performed during the course of the PhD) to make the technology suitable for breast cancer imaging. The first chapter of the thesis reviews the unmet needs of breast cancer imaging, being followed by a summary of recent techniques and technologies that may potentially fulfil gaps in knowledge and address some of the specific diagnostic challenges in breast cancer imaging.

The capabilities of optoacoustic imaging are then discussed in the context of this evolving landscape of new imaging techniques and technologies with a particular focus on the tumour biology (neoangiogenesis and hypoxia) that can be measured in humans using multimodality and multi parametric imaging. Chapter 2 reviews of the current state of clinical translation of optoacoustic imaging, highlighting the particular areas in which clinical translation has advanced the most (breast cancer, melanoma and inflammatory bowel disease). Chapter 3 discusses the logistical, regulatory and technical challenges and solutions involved in translating optoacoustic imaging to the clinic and setting up a clinical service.

Chapter 4 presents a series of validation experiments of oxygen saturation aimed at establishing the relationship between the optoacoustic signal and invasive pO2 measurements with an OxyLite probe in a porcine kidney model. This work was conducted in close collaboration with leading clinicians from the local transplant team. The following chapter describes the results of the first stage of our clinical work in the breast, namely the healthy volunteer study. This part had several aims: to perform qualitative assessment of the optoacoustic features of the normal breast under physiological conditions; to establish a robust scanning technique and identify technical and image interpretation pitfalls; and to perform qualitative evaluation of the hormonal changes that occur during the menstrual cycle and menopause, which, in turn, were used to validate surrogate measures of oxy-, deoxy and total haemoglobin.

Chapter 6 then focuses on the qualitative assessment of benign and malignant breast lesions and their appearances on optoacoustic imaging. The patient study was divided into three phases. Phase 1 created a feature set to differentiate benign from malignant lesions, while Phase 2 was a transition between the prototype scanner and the installation of the first-generation clinical scanner. In Phase 3 the feature set created in Phase 1 was validated in a reader study. The sensitivity and specificity of optoacoustic imaging for lesion detection and differentiation of benign from malignant lesions was compared with mammography and ultrasound.

Chapter 7 then deals with the quantitative analysis of the Phase 1 and Phase 3 data acquired in Chapter 6, assessing the relationships between the use of single wavelengths, spectral unmixing, vascularity versus receptor status, heterogeneity of signal intensity in relation to tumour stage and grade. This chapter also discusses the potential and limitations of quantifying the optoacoustic signal and leads to the final chapter, a discussion of future directions in optoacoustic imaging in breast cancer. At the end of this thesis, chapter 8 briefly discusses the potential future directions for the use of optoacoustic imaging as a clinical and scientific tool.