Biomarkers and imaging


The inclusion of prognostic and predictive factors that govern treatment outcome is essential for informed treatment decision and possible stratification in the effective management of radiotherapy of cancer patients. Biomarkers represent an attractive approach to identify such factors, in particular imaging biomarkers, as they can be obtained non-invasively using techniques that are available in radiotherapy. The WP aims to identify imaging metrics which can act as surrogates for the underlying biological mechanisms which affect the tissue response to radiation therapy. This response includes both disease control and normal tissue toxicity. The main imaging modalities used clinically and in radiation oncology are CT, MRI and PET why these modalities are the focus in the WP. However, there are several translations gaps to be bridged before clinical utility, including preclinical- technical and clinical validation.


Reproducibility of candidate biomarkers is critical for the translation of such markers into clinical routine. In the previous DCCC-RT WP 2, one of the focus areas was reproducibility of specifically diffusion weighted MRI (DW-MRI) derived parameters, and led to publications in national and international collaborations. This work is expected to continue in the new WP since several aspects of reproducibility are still under investigation, and focus is expanding to other imaging modalities as well. All Danish radiotherapy centres, as well as international centres across Europe and Northern America, are current collaborators.

Previous work also investigated the potential of DW-MRI as a prognostic marker in rectum cancer, glioblastoma, and pancreas cancer. We expect to continue this work improving DW-MRI methods, and expanding the portfolio with new candidate biomarkers both within MRI (e.g. oxygen or tissue oxygen enhanced MRI, TOLD, and relaxometry), and within PET and spectral CT, which are currently being investigated as hypoxia surrogates. Biomarker approaches (based on tissue- and liquid biopsies) will be integrated when feasible, for validation and correlation. Pre-clinical investigations (PET-MRI) are being planned as well.


WP 2.1 Hypoxia in radiotherapy
Hypoxia is a well-established negative prognostic factor for numerous cancer sites. Although being intensively studied in pre-clinical and clinical trial settings for many decades, no method for identifying and compensating for the presence of hypoxia has found its way into clinical routine practice. The aim of the WP is to collect and share ideas and experiences from the pre-clinical and clinical research groups working on this to facilitate the establishment of such methods. The main approach will be the use of existing procedures and technology in the clinic applied in novel ways. Below is described some ongoing and potential future projects that could exist within such a WP structure.


WP 2.1.1 Multi-modal hypoxia imaging in head and neck cancer
Prior and ongoing investigations have indicated that the use of imaging biomarkers derived from DW-MRI (prostate cancer) and dual-energy CT together with FDG PET (head and neck cancer, HNC) can potentially visualize hypoxia. A common hypothesis for these observations could be the so-called consumption and supply-based hypoxia model. The project therefore aims to jointly apply these multi-modal images commonly available in radiotherapy of HNC to clinically validate the predictive power of the biomarkers alone or in combination with each other in a multi-institutional setting.


WP 2.1.2 Multi-modal hypoxia imaging in oligometastatic disease
The treatment resistant effects of hypoxia seem more pronounced for squamous cell carcinoma (SCC) receiving high doses per fraction. Patients with such tumors and treatment strategies are often presented with oligometastatic decease. Patients with oligometastatic tumors can currently be treated on a hybrid MR-linac system capable of performing online MRI. Further, many of these patients have pre-treatment FDG PET and spectral CT available. By inhalation of oxygen-enriched air, the different magnetic properties of oxy- and deoxyhemoglobin can be detected and quantified using TOLD MRI.


WP 2.1.3 Preclinical PET/MRI study of hypoxia
The project involves leveraging diverse translations imaging techniques (correlative imaging, animal 7 T PET-MRI) and biological models (organoids, animals) to elucidate tumour- and normal tissue radiobiology, and its relevance to radiotherapy efficacy and toxicity, while WP 2.1.1 -2 translates validated protocols to human MR-, MRI-linac- and PET/MR systems. 

WP 2.2 Histological validation of DW-MRI based biomarker in glioblatoma
Characterization of radiotherapy resistant regions within glioblastomas can provide information regarding informed decisions for surgery and the adjuvant radiotherapy, and potentially enable personalized treatment. In this project a refined method for diffusion measurement will be implemented on radiotherapy MRI simulators. Model for diffusion MRI will be tested using Monte Carlo simulations, tested in phantoms and healthy volunteers, before going into patients for pilot testing. The project will be initiate as a national collaboration between Odense University Hospital (OUH), Technical University of Denmark (DTU) and Hvidovre Hospital.

WP 2.3 Outcome prediction in pancreas cancer
Diffusion weighted MRI (DW-MRI) has in a pilot study initiated from this WP shown a potential to predict overall survival in patients with pancreatic cancer. In a follow-up study the model will be tested in a national and/or international multi-center validation cohort, using longitudinal DW-MRI derived parameters. A cooperation with IP9 would make sense and further in silico studies for possible adaptive treatment strategies based on the images available could be incorporated in WP4.

WP 2.4 Toxicity and response assessment using spectral CT
Dual-energy and photon counting CT derived iodine maps can visualize the perfused blood volume as presented by the neovascular structure within tumors and organs-at-risk (OARs). The project will investigate if iodine maps obtained longitudinally could act as a potential surrogate for decease control, response assessment and toxicity, and further aid to quantify the development or destruction of the vascular structure in response to radiotherapy. Minor patient cohorts across multiple cancer diagnoses will be included in the investigation.

WP leaders

Faisal Mahmood (OUH), Jens Edmund (HGH), Jens Overgaard (AUH)

  • Brita Singers Sørensen

    Professor, PhD

    Aarhus University Hospital
  • Jens Edmund

    Hospitalsfysiker, PhD

    Herlev Hospital
  • Thomas Ravkilde

    Hospitalsfysiker, MSc, PhD

    Aarhus University Hospital
  • Faisal Mahmood

    Professor, Research unit of Oncology

    Odense University Hospital
  • Jens Overgaard


    Aarhus University Hospital
  • Laura Patricia Kaplan

    Physicist, PhD

    Zealand University Hospital, Næstved Sygehus
  • Mette van Overeem Felter

    MD, PhD

    Herlev Hospital