The research ambition of the department is to establish and expand research lines in oncologic imaging and interventional oncology addressing various cancer types. The lines will follow the priorities of the research themes in our institute (personalised medicine, immunotherapy, image-guided treatment, and fundamental research). Hence, the research priorities of the institute and will focus on the following 6 main streams:
MRI and organ preservation
Multi-parametric imaging for personalized medicine and
Radiomics and immunology
This line aims to investigate the safety of the practice of organ preserving treatment for rectal and oesophageal cancer patients with complete response after preoperative chemoradiotherapy and develop and validate modern MR techniques for better selection and follow up (multidisciplinary collaboration with the group of Prof Geerard Beets, a surgical oncologist).
Multi-parametric imaging for personalized medicine and image guided treatment .The combination of information from multiple modalities (MRI, functional MRI and FDG PET) has the potential to predict response to chemoradiotherapy in rectal cancer. Combining this potentially complementary data will lead to a better understanding of treatment response and thus provide better diagnostic value for the prediction of how a patient will respond to chemoradiotherapy. Such models can support other clinical and histological markers in predicting response and thereby further optimize personalized treatment in rectal cancer.
The prognosis of patients with peritoneal spread has dramatically improved since the launch of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS-HIPEC). To select patients who could benefit from CRS-HIPEC the Peritoneal Cancer Index (PCI) is used. The use of whole-body DWI-MRI has the potential to improve the detection performance of patients with peritoneal carcinomatosis (PC) of colorectal and ovarian carcinoma. A recent study demonstrated that DWI MRI was able to identify patients who may benefit from cytoreductive surgery and HIPEC with an accuracy of 94%.
Voxel-wise detection and classification of malignancies in multi-parametric images are all time-consuming operations, which often require assessing multiple imaging sequences in parallel. Our aim is to adopt artificial intelligence methods to facilitate and accelerate the work of researchers and clinicians at our department. Examples of projects in this line include fully automatic methods for image segmentation, assessment of treatment follow-up CT scans, and automatic differential diagnosis of granulomatous disease, pneumonitis, vasculitis and active lymph nodes.
Our immunotherapy imaging research aims to follow a core concept wherein the role of radiomics is included as part of the integrated diagnosis of cancer. This includes development of radiomics-based machine learning models to predict the response of each individual metastasis in lung cancer and melanoma patients treated with anti-PD1 checkpoint inhibitors. We aim to develop models to better understand the immunoprofile of the tumour, in addition to the tumour microenvironment (i.e. viable tumour cells, necrosed tumour cells, immune infiltrates, and edema) based on MRI/CT images.
Interventional radiology (IR) is evolving rapidly in the oncologic setting and becoming essential in the modern oncologic practice. Our ambition is to become a leading provider of oncologic IR by means of fostering research and implementing new techniques that embrace the current strategic vision of NKI-AVL (multidisciplinary approach to immunotherapy, image-guided intervention and image-guided navigation surgery, personalized treatment). Such interventions include cryoablation, CT-fluoroscopic guided thermoablation, and interventional treatment and biopsy guided by ultrasound fusion.