Liver MRI is increasingly used for staging of patients prior to local treatment of colorectal liver metastases(1). PET/CT is still necessary to exclude extra hepatic disease(2). A whole body MRI protocol, which can be added to a dedicated liver MRI, could alleviate the need for the PET/CT and offer a single modality one-stop-shop approach to staging patients before treatment of colorectal liver metastases. Prior to a direct comparison of whole body MRI and PET/CT, the appropriate MRI protocol should be decided from a variety of available whole body MRI sequences. It is recognised that a whole body MRI protocol should be tailored to the specific malignancy(3), hence a prospective study in colorectal cancer patients was decided.The overall purpose of the study was to define a proper whole body MRI protocol for detection of extra-hepatic colorectal cancer metastases with PET/CT as the reference standard. In this preliminary analysis we assess the effect of adding DWIBS to a standard whole body MRI protocol.

30 patients with extra-hepatic metastases from colorectal cancer were prospectively recruited and informed consent was obtained. All patients were scanned on a 1.5 T scanner (Philips 1.5 T Ingenia, Philips Medical Systems, Best, The Netherlands) using the dS Flex Coverage posterior coil, dS HeadNeck Coil and two Flex Coverage anterior coils. The patients were scanned from shoulders to upper thighs using transverse and coronal T2W single shot respiratory triggered TSE, breath hold coronal STIR, post-contrast 3D GRE mDixon and free-breathing DWIBS (B=0, 50, 900). The MRI contrast used was gadoxetic acid (Primovist, Bayer Pharma, Berlin, Germany) when the participant had an estimated glomerular filtration rate (eGFR) > 60 ml/min/1.73 m2 body surface area and gadoteric acid (Dotarem, Guerbet, Roissy CdG Cedex, France) when eGFR was between 30 and 60. All patients were scanned on an integrated PET/CT system, (Siemens Biograph Truepoint 64-slice, Siemens Healthcare, Erlangen, Germany) according to the standard clinical protocol, as reference.

In this preliminary analysis two MRI sets were constructed for each patient: One traditional set with all sequences excluding DWIBS and one set with all sequences including DWIBS. Case report forms were completed during the blinded reporting of all 60 MRI sets. On the case report forms lesions were categorized into 13 anatomic regions and a maximum of three lesions were reported for each region. For each lesion location, size (mm) and confidence of malignancy (1 – 5) was recorded.

Similar case report forms were completed during the reporting of the PET/CTs, which were reported with knowledge of all prior imaging and medical records.

Per lesion sensitivity and false positive rate was calculated and per region sensitivity and specificity was calculated. Findings were compared using McNemars Chi²-test.

175 lesions were detected in 90/390 anatomic regions by PET/CT in the 30 patients. Mean interval between PET/CT and whole body MRI was 1.33 days (range 0-19 days). The results of the preliminary analysis are displayed in table 1.

The preliminary results show that an MRI protocol incorporating DWIBS is superior compared to a standard whole body MRI protocol for detecting extra-hepatic colorectal cancer metastases. This illustrates the huge benefit of using diffusion weighted imaging to localize areas of impeded water diffusion to detect malign lesions(4).

The cost of a higher sensitivity is as expected an increase in the false positive rate and correspondingly a decrease in specificity (although not statistically significant). This could be attributed to lymph nodes, which present as high signal lesions on diffusion weighted sequences regardless of their nature. Further statistical analysis will reveal if this explains a large proportion of the higher false positive rate in the sets containing DWIBS.

The choice of PET/CT as the reference standard entails the possibility that false positive PET/CT findings are recorded as false negative findings on MRI, resulting in a lower sensitivity. Equivalently, lesions appearing as false negative on PET/CT e.g. due to size under the detectability limit of PET/CT, would result in an increased false positive rate / reduced specificity of the most sensitive MRI sets. Nevertheless, we chose to consider PET/CT as the reference standard in accordance with the purpose of defining an MRI protocol. Therefore, this study cannot be considered a comparison between whole body MRI and PET/CT.

A further subgroup analysis is planned to decide if it is possible to use DWIBS in combination with fewer sequences (w/wo iv. contrast) to allow for a decreased examination time and maintain acceptable sensitivity and specificity.

Preliminary results reveal the superiority of using DWIBS in detecting extra-hepatic colorectal cancer metastases.