A rapidly growing tumour can’t deliver oxygen to all its regions. The resulting oxygen-starved tumour regions, however, are difficult to treat with radiation therapy, a technique that relies on free radicals produced in the presence of oxygen to damage DNA in cancer cells.
Clinicians have been tackling this problem with a variety of approaches – from radiosensitizers that enhance the effects of radiotherapy in hypoxic tumours to techniques like proton therapy that deliver high radiation doses. Still, researchers want to be able to identify oxygen-starved tumours so that treatments can be adjusted to target such tumours more effectively. But current techniques to measure tumour oxygen levels are invasive, provide limited spatial information or require radiopharmaceuticals that as yet cannot be obtained in many clinical settings.
In an important step for non-invasive hypoxia imaging and future biology-guided adaptive radiotherapy studies, researchers have integrated a technique to measure tumour oxygenation with an MR-linac, a hybrid MRI scanner and radiotherapy delivery system.
Michael Dubec, a principal scientist in magnetic resonance imaging at The Christie NHS Foundation Trust and an MR research physicist at The University of Manchester, is first author on the study, which was published in Radiotherapy and Oncology.
“In this work we investigated the change in longitudinal relaxation rate (R1) in tumours induced by 100% oxygen gas breathing,” Dubec says. “Based on previous validation work against immunohistochemistry, we can say that the ΔR1 technique can be used to identify tumour regions associated with low oxygen levels.”
During an oxygen-enhanced magnetic resonance imaging (OE-MRI) scan, patients breathe pure oxygen, which initially binds to haemoglobin, maximizing blood oxygen saturation. Any additional oxygen then dissolves in blood plasma and tissues, increasing the concentration of oxygen molecules and leading to faster longitudinal net-magnetization recovery and a greater longitudinal relaxation rate (R1).
The researchers tested the hypoxia imaging technique using a diagnostic MR scanner, in healthy participants and then participants with head-and-neck cancers. They also performed phantom studies. They created images showing change in R1 throughout the head and neck, and used region-of-interest analyses to measure the magnitude of this change in tumours.
Dubec and colleagues repeated the study on an MR-linac system. They conclude that the OE-MRI methods are repeatable and reproducible on MR-linac systems and provide “equivalent quality data” to that acquired on diagnostic MR scanners.
“Oxygen-enhanced MRI offers a practical and readily translatable technique to assess oxygenation in normal tissues and tumours which we have, for the first time, shown can be incorporated onto MR-guided radiotherapy systems with no issues reported from healthy volunteers and patients,” Dubec says.
Photoacoustic mapping of tumour oxygenation predicts radiotherapy efficacy
Though the researchers used an MR imaging sequence that acquires 3D image volumes rapidly, they note that their protocol is still too long to fit into a standard MR linac workflow. Additional work will incorporate a perfusion sequence to identify necrotic regions and will evaluate the reproducibility of methods and results across clinics. Dubec says that validation work should also directly link changes in R1 value to changes in absolute oxygen concentration and then to specific oxygen levels in tumours.
“Essentially, we aim to develop and translate the OE-MRI technique so that it can be used for adaptive radiotherapy-based clinical trials in hospitals in the future,” Dubec says. “Having more institutions investigate, and collaborate on, OE-MRI techniques is important so that we can accumulate more evidence of the limitations and benefits of this technique, and assess its utility in different tumour types.”
