Skip to content
Main menu hidden.

The Halo-study

Research project Use of advanced imaging, liquid biopsies/blood-tests and neuropsychology for longitudinal diagnostic assessments of malignant glioma. Malignant glioma (WHO II-IV) is the most common form of primary brain cancer.

Standard of care for glioma treatment is today multimodal, including neurosurgery (for tissue diagnosis and with the further aim to perform the most radical surgery possible yet trying to avoid neurological deficits), followed by radiotherapy and chemotherapy or (particularly in some lower grade gliomas), watchful waiting. Some patients are also offered treatment with Tumour Treating Fields.

Head of project

Maria Sandström
Associate professor, senior consultant (attending) physician

Project overview

Project period:

2022-01-01 2025-12-31

Participating departments and units at Umeå University

Department of Radiation Sciences

Research area


Project description

Throughout the course of the disease, the ability of the clinician to longitudinally understand and predict its nature as well as the effects of therapeutic
interventions and their interactions with normal brain function is essential for clinical decision making.

The non-invasive diagnostic modality that is currently considered the gold standard for longitudinal and prognostic assessments of the disease is brain MRI, including T2-weighted, T2-weighted fluid-attenuated inversion recovery (FLAIR) sequences and 3D T1-weighted sequences before and after application of a gadolinium-based contrast agent. Such images are used both to understand the potential impact of the disease on brain function and the degree of disease activity/malignancy at different stages of the disease. However, despite its widespread use and utility this approach has several important limitations that will frequently force clinicians to make decisions where survival as well as potential neurological deficits lie in the balance based on limited information with an unacceptably high rate of uncertainty.

In this context a number of new diagnostic modalities aimed at increasing the understanding of the longitudinal progression of the disease have during later years become increasingly important. Some of the most promising of these modalities are:

  • Use of blood samples (i.e.”Liquid biopsies”) to predict disease progression using blood diagnostic markers.
  • O-(2-[18F]fluoretyl)-L-tyrosin) (FET-PET)-MRI to visualize metabolic tumor activity.
  • Resting state functional MRI (fMRI) and task-based fMRI using the Blood Oxygen Level (BOLD) signal to visualize areas with specific brain functions and activity in functional neural networks.
  • Advanced neuropsychological testing.

In the present project we will repeatedly apply all of these diagnostic modalities in order to follow a cohort of glioma patients from initial diagnosis through the first series of neurooncological interventions. The main purpose of this will be to increase our understanding of the utility of these procedures in the managing and understanding of the glioma disease and its interaction with normal brain function.

Research projects on liquid biopsies, fMRI and neuropsychological testing has already been initiated by our group within the frameworks of explorative studies in glioma patients. Here those studies will be integrated with one another and in addition, FET-PET-MR will be added as yet another diagnostic modality. This means that FET-PET-MR is the new modality added for which ethical permission has not been previously granted to our group in this context.

Survey of the field

Malignant gliomas are a heterogenic group of different diseases. Depending on tumour type, molecular pattern and clinical factors, survival differs between a couple of months up to twenty years. For patients with good prognosis there is always a risk of serious side-effects after treatment which commonly includes surgery, radiotherapy and chemotherapy.

In the design of this study, experience from research with brain functional MRI by Lars Nyberg and his group (2) together with Rickard Sjöberg (3) has been considered. We will therefore use a battery of imaging and neuropsychological tests. The rationale for this is strong theoretical reasons to assume that plasticity of function/rehabilitation potential is strongly related to plasticity in these networks. In parallel, brain function will be monitored by a battery of psychological tests.

FET-PET is a method with a tracer that is suitable for imaging the brain. FDG (fluorodeoxyglucose), the more common tracer used to look for high metabolic activity is less suitable since the normal brain has a very high uptake of FDG. This makes differentiation between normal brain tissue and tumor tissue much more difficult.

In neuro-oncology there is a long experience with another amino acid-tracer, 11C-metionin (MET) which mimics the protein uptake into the tumor and is strictly correlated to cell-density and to a lesser part to tumor cell proliferation and vascularization. Uptake of MET is also seen in tumours with intact blood-brain-barrier which makes it suitable also for low-grade glioma often lacking contrast enhancement.

During the last years, MET has been replaced by FET as tracer in many centers for brain tumor imaging, since it has a longer half-life, due to labelling with 18F and a more preferable manufacturing (4). Several studies have shown comparable results for the tracers FET and MET in gliomas. Dynamic FET-PET has also been shown to be able to differentiate between high- and low-grade glioma (5) and tumors with different IDH-genotypes, which is an important prognostic feature in gliomas (6). Until recently, amino-acid PET has not been available in clinical routine at Umeå University Hospital. Since 2015, however, a study with FLT-PET/MRI has been running at Umeå University/Umeå University Hospital and is now closed for inclusion with 50 patients included. Data is currently under processing and several research questions are to be analysed.

Only a few comparative studies of primary diagnostic work-up in gliomas using FLT-PET and FET-PET have been published (7, 8) showing superior diagnostic value of FET-PET in gliomas, mainly due to FET being able to detect tumours independently of intact or damaged bloodbrain-barrier. There are, to the best of our knowledge, no previous publications directing whether FET-PET or FLT-PET is the better choice in later stages of tumour (pseudo)progression/ recurrence.

Liquid biopsies may open up for the possibility to detect early signs of disease progression in blood rather than wait for tumour growth to be visible in MRI. One of the major problems with brain tumours is that tissue samples are hard to get and not without risks. In blood, specific mutations can be detected and followed in blood as circulating tumour DNA (ctDNA) or as tumour derived RNA sequestered by platelets (ptRNA) in order to find prognostic and predictive tumour markers (9). One research question is to find out if the pattern is in concordance with findings in FET-PET/MR. Blood will also be biobanked for later analyses regarding metabolomics.

Research questions

As outlined in the purpose and aims section above, the primary aim of the present project is to perform a prospective longitudinal study of glioma patients in which exploratory data from several different modalities are collected. This will allow us to increase our understanding of the disease using approaches where data from these different modalities are correlated in meaningful ways. Examples of research questions that will be explored with the help of FETPET-MRI:

  1. Does FET-PET/MRI predict outcome in patients with glioma?
  2. Can liquid biopsies and/or metabolomics in blood serve as a surrogate marker for FET-PET/MRI and predict prognosis and response to treatment
  3. Does FET-PET/MRI add relevant information in dose-planning of radiotherapy?

Project description

Patients: 40 patients with glioblastoma (WHO grade IV) and 30 patients with glioma/oligodendroglioma WHO grade II-III will be consecutively included preoperatively in this study.

We estimate an inclusion of 70 glioma patients over two years. In the Northern Region of Sweden approximately 60 non-pedriatic patients are admitted for glioma per year. Forty of these are glioblastoma (WHO grade IV). A very conservative estimate would be approximately 50% drop out (difficulties to adhere to protocol due to other illnesses, not eligible for surgery, not speaking Swedish, etc), giving an inclusion of totally 40 patients with glioblastoma for two years. In addition, approximately 20 gliomas of WHO-grade II-III are diagnosed per year.

These patients are often less affected by the disease and we estimate only 25% drop out giving 30 patients included into the study with gliomas of WHO grade II-III over two years. Using this FET-PET and these number of GBM-patients we expect to have an 80% chance of detecting medium level effect sizes (Z=0.5) between methods which should cover clinically meaningful effects.

Inclusion criteria: Patients > 18 years of age, scheduled for surgery for suspected glioma, and who give their informed consent to participate. Patients are eligible if they are planned for adjuvant treatment with radiotherapy and/or chemotherapy with or without tumour treating fields) TTF after surgery. Patients with glioma that is recommended active expectation instead of adjuvant treatment are also eligible (3b in sampling protocol).