Photoacoustic imaging as a tool to probe the tumour microenvironment

Abstract

The tumour microenvironment (TME) is a complex cellular ecosystem subjected to chemical and physical signals that play a role in shaping tumour heterogeneity, invasion and metastasis. Studying the roles of the TME in cancer progression would strongly benefit from non-invasive visualisation of the tumour as a whole organ in vivo, both preclinically in mouse models of the disease, as well as in patient tumours. Although imaging techniques exist that can probe different facets of the TME, they face several limitations, including limited spatial resolution, extended scan times and poor specificity from confounding signals. Photoacoustic imaging (PAI) is an emerging modality, currently in clinical trials, that has the potential to overcome these limitations. Here, we review the biological properties of the TME and potential of existing imaging methods that have been developed to analyse these properties non-invasively. We then introduce PAI and explore the preclinical and clinical evidence that support its use in probing multiple features of the TME simultaneously, including blood vessel architecture, blood oxygenation, acidity, extracellular matrix deposition, lipid concentration and immune cell infiltration. Finally, we highlight the future prospects and outstanding challenges in the application of PAI as a tool in cancer research and as part of a clinical oncologist's arsenal.

Keywords: Cancer; Hypoxia; Optoacoustic imaging; PAI; TME; Vasculature.

Fig. 1.

The tumour microenvironment (TME). Schematic diagram illustrating the involvement of multiple cell types in a tumour, including endothelial cells and pericytes that make up blood vessels, as well as immune cells, fibroblasts and adipocytes, alongside the cancer cells. Lipids are synthesised by adipocytes and cancer cells. Hypoxia arises as the tumour grows beyond the limit of oxygen diffusion from the surrounding vessels. Fibrosis arises from excessive deposition of extracellular matrix (ECM) components without concurrent degradation. A supportive environmental niche of these chemical and physical signals evolves with the cancer cells to promote tumour development and progression.

Fig. 2.

Principles of photoacoustic imaging (PAI). (A) During PAI, pulses of light illuminate the tissue (1). When light is absorbed (2), a transient heating gives rise to ultrasound waves (3). The ultrasound waves are then detected and used to reconstruct an image of the optical absorption in tissue (4). (B) Absorption spectra of endogenous molecules that absorb light pulses and can provide insight into the tumour microenvironment (TME). Panel B is reproduced with permission from Weber et al. (2016). This image is not published under the terms of the CC-BY licence of this article. For permission to reuse, please see Weber et al. (2016). Hb, deoxyhaemoglobin; HbO₂, oxyhaemoglobin.

Fig. 3.

Example photoacoustic images of the vasculature. (A) Shown are x-y maximum intensity projections of a human colorectal tumour (SW1222) and the surrounding vasculature between day 7 and day 8 post-inoculation. Dashed white lines indicate tumour margins. Green arrows show common vascular features between images. Increasing tortuosity of normal blood vessels between day 7 and 8 is indicated by blue arrows. Reproduced with permission from Laufer et al. (2012) and the Journal of Biomedical Optics. This image is not published under the terms of the CC-BY licence of this article. For permission to reuse, please see Laufer et al. (2012). (B) Representative images of PC3 (left) and LNCaP (right) tumours showing the spatial distribution of ΔsO₂ measured using PAI at multiple wavelengths. PC3 tumours displayed lower ΔsO₂ compared to LNCaP tumours and had a core with low ΔsO₂ (black arrow). Reproduced with minor formatting changes from Tomaszewski et al. (2017). (C) Photoacoustic pH image of rat glioma tumours at 75 min after SNARF-5F nanoparticle injection. The pH in the centre area (i.e. the area within the solid line) and the peripheral areas (i.e. the area between the solid line and the dashed line) are averaged, respectively. Reproduced with minor formatting changes from Jo et al. (2017). (D) Depth-encoded PAI images of the breast acquired while the patient, a 49-year-old woman with a stromal fibrosis or fibroadenoma, held her breath. Dashed white lines indicate tumour margin. Reproduced with minor formatting changes from Lin et al. (2018).

Fig. 4.

Example photoacoustic images of lipid content and immune cell tracking. (A) Representative transverse B-mode ultrasound images (top row) and PAI performed at multiple wavelengths to reveal lipid content (bottom row) of four histologies [normal, hyperplasia, ductal carcinoma in situ (DCIS) and invasive carcinoma] from distinct animals of a transgenic mouse model of breast cancer progression. Lesion severity increases from left to right. Regions of interest are outlined in red on B-mode images. Scale bars: 2 mm. Reproduced with minor formatting changes from Wilson et al. (2014). (B) In vivo imaging of near-IR-797-labelled T cells in a mouse sarcoma model displaying infiltration of T cells over time with a peak at 12 h and subsequent decline up until 72 h post-adoptive transfer. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced from Zheng et al. (2018) with permission, with minor formatting changes. This image is not published under the terms of the CC-BY licence of this article. For permission to reuse, please see Zheng et al. (2018).