07 Jun 2016
Sarah Bohndiek leads a team studying new ways to monitor tumor development.
Sarah Bohndiek at the University of Cambridge leads an international team using bioimaging techniques to better understand the topic, giving her a broad perspective on the potential for novel photonics technologies to tackle this key question in cancer studies.
"A developing tumor stimulates the growth of new blood vessels to support itself, called angiogenesis," commented Dr Bohndiek. "But that supply can initially be relatively poor, leading to some areas of the tumor suffering from hypoxia - insufficient levels of oxygen."
At the same time, other areas of the tumor can experience oxidative stress, when the proliferation of cancer cells in an environment of sufficient oxygen generates an excess of free radicals, which are themselves damaging to the cell's DNA.
"Both of these stresses are known to be linked with poor clinical outcomes," said Bohndiek. "Hypoxic stress in hormone-sensitive cancers such as breast, prostate or ovarian cancer tends to make clinical prognosis worse. So imaging modalities able to study how oxygen is delivered and utilized in a tumor could be very valuable. In particular, directly probing the absorption properties of hemoglobin allows us to determine how much blood is present in the tumor, and how well oxygenated it is."
Bohndiek's team is studying a number of modalities that might contribute to this understanding, predominantly using visible or near-infrared light. A key avenue of interest is multi-spectral optoacoustic tomography (MSOT), a variant of the optoacoustic technique performed at multiple wavelengths, using the differential absorption of oxy- and deoxy-hemoglobin to assess oxygenation in tissue.
"Our work on MSOT sits close to the clinic, in that we have a series of experiments under way to validate the technique and determine just how reliable and quantitative the resulting images are," Bohndiek commented "We are also working to see how the MSOT data relates to the physical changes in blood vasculature within a tumor, and how to connect the results with vascular function in a meaningful way."
Another topic of research is hyperspectral imaging (HSI), able to produce both spectral and structural information about a tumor simultaneously. One goal here involves integrating monolithic hyperspectral sensors into robust fluorescence HSI systems, and applying the multiplexed imaging of reflectance, autofluorescence and fluorescent contrast agents. These experiments are currently being performed in pre-clinical cancer models.
A third angle of attack exploits current advances in endoscopy, both as a means to improve early cancer detection and potentially for therapeutic purposes too.
"Endoscopy is helpful in assessing high-risk patients, such as those with a predisposition to esophageal or colorectal cancer, but finding early, curable, lesions is challenging" Bohndiek noted. "Having advanced optical imaging capabilities incorporated into the endoscope - perhaps including hyperspectral imaging, near-infrared fluorescence or optical coherence tomography – could provide significant advances in endoscopic surveillance of high-risk patients in the coming years."
As an example, Bohndiek points to her group's collaboration with Rebecca Fitzgerald, also of the University of Cambridge and developer of the Cytosponge - a "pill on a string." When swallowed, the device dissolves to reveal a sponge able to scrape off cells as it is withdrawn up through the gullet, collecting cells from along the length that can be analyzed to identify high risk patients.
"We are looking at partnering that type of test with an endoscopic imaging technique that looks directly at the lesions, to try to diagnose very early changes in the gullet that are not yet strictly cancerous but which indicate an increasing likelihood of progressing to cancer," she said. "That kind of partnership is where I can see optical techniques having a really strong role in clinics."
Translation also requires an understanding of exactly where an optics-based procedure might fit into the diagnostic pipeline and general clinical practices, which may well vary between geographic regions. In the UK, patients are highly likely to first see a local GP and then be referred to a specialist - such as a dermatologist for a skin lesion - where the more advanced technologies are usually brought to bear.
"Adoption depends on whether a technology solves a clinical need and can fit readily within the existing clinical paradigm, and this is where optical imaging modalities can often run into problems," noted Bohndiek. "If you look at the suite of clinically approved optical imaging methods, very few are actually used routinely in clinics. They need to pervade downwards and outwards from the hospital environment, and into a more local GP setting, to impact patient management. "
This in turn requires the new techniques to be incorporated into robust medical devices, able to withstand the realistic wear and tear of clinical environments - another hurdle where optical devices can sometimes fall short. Developers run the risk of investing too much at the early stages of product development, and overlooking how it can be packaged so as to survive life in the real world.
"A multidisciplinary approach is crucial in tackling these questions, and in the UK I am happy to say that we have been focused on driving interdisciplinary research forward in recent years," concluded Bohndiek. "Exactly how these programs develop in the future is still being discussed, but there is a strong interest in bringing more physical scientists from a broader range of backgrounds into biomedical studies. That's one reason why now is definitely a good time to be doing this kind of research."
About the Author
Tim Hayes is a contributor to Optics.org.