Exploring the use of optical coherence tomography and hyperspectral imaging in glioma detection during surgery

Surgery is a very important part of glioma treatment (followed with
radiotherapy and chemotherapy). The best survival is seen when the tumor is
resected as much as possible. Resection is hampered by difficulties in
visualizing tumour and differentiating from normal tissue. The use of HSI/OCT
seems promising especially for distinguishing LGG, since no other technique
exists to achieve this. In high grade tumors fluorescent techniques have
already shown to distinguish healthy brain from tumor tissue and improve
(progression-free) survival. This technique has several draw-backs: it is
costly (about ¤1000 per patient), involves ingestion of an agent
(5-aminolevulinic acid) with possible side effects and does not visualize the
infiltrative zone of the tumour very well.

The aim of this project is to develop an intra-operative, real-time
visualization, that surgeons can use for accurate detection of glioma in order
to extend survival and quality of life of the patient.

Primary Objective:
To assess whether both OCT (optical coherence tomography) and/or HSI
(hyperspectral imaging) detect differences in glioma tissue as compared to
healthy tissue as examined during brain-surgery.

Secondary Objective(s):
To assess if either technique can discern a characteristic pattern that enables
automatic detection of pathological tissue.
Can a combination of these techniques discern such a pattern?

Study design is that of a case control study in which healthy brain and in vivo
glioma tissue will be imaged during brain tumour surgery. This design
facilitates the best possible control group (same subject, same brain to
compare glioma with), but also it represents the exact circumstances in which
possible future application of the tested imaging techniques will be applied.
We aim at inclusion of 3 low grade and 3 high grade glioma patients and we want
to collect 5 images of normal brain surface, 5 images of glioma surface, 5
images of subcortical glioma and (when available) 5 images of normal
subcortical tissue in each subject. The spatial resolution of the HSI image
depends on the optics; a large field of view will lower its resolution. In
practice the resolution will be sub mm. Each OCT picture consists of vast
amounts of data points. This will yield enough data for extensive comparison of
imaging data.
Data collection takes place during surgery at the Radboud university medical
center. Each year about 80 glioma patients are operated here, half of whom
carry al low grade glioma. Since some patients will only be biopsied and glioma
incidence varies, if we take into account that half of the available patients
will or can participate, inclusion of 6 patients will most probably will take
less than 3 months.

There are no anticipated additional risks for the participants since the OCT
and HSI are non-invasive imaging methods based on the interaction of tissue and
light, which is harmless to the patient. However, the operation time is
prolonged with a few about 20 minutes.