The incidence of malignity in the gastroesophageal junction (GEJ) is rising in Denmark, 400 new cases annually 1. 60% of the patients with GEJ cancer have disseminated disease (M1-stage) at time of diagnosis and are consequently unable to undergo curative intended surgery. The prognosis is poor, with 5-year survival rates of approximately 2 % for these 60% and the treatment only consists of palliative therapy 1. For the remaining approximately 40 % who are assessed as candidates for curative intended surgery, the 5-year survival rate is 33 % 1. This emphasizes the need for further research and knowledge concerning tumour biology and spontaneous course of the disease.

In Denmark, all cancer patients are enrolled in a specific cancer program. The primary diagnostic work-up for GEJ cancers includes gastroscopy with biopsy, blood samples, ultrasonography scan, Positronemissionstomography (PET) and Computed tomography (CT) alone or PET/CT in combination 2. From these parameters physicians determine resectability and TNM-stage (tumour staging), which is substantial for the prognosis and future treatment. The primary goal is to achieve a macroscopically resection of tumour and lymphnodes in relation to the stomach and oesophagus (Esophagectomi a.m. Ivor Lewis and D1+ lymphadenectomy in the abdomen and thorax). In addition to surgery, patients receive perioperative chemotherapy, which consist of three series of chemotherapy preoperative and three series postoperative approximately 21-28 days after surgery.

Approximately 12.6 % of patients receiving perioperative chemotherapy prior to surgery will have disease progression due to chemotherapy resistance during the therapy 3. This unintentionally leads to shifting these patients from the resectable group to non-resectable group (palliative treatment). Thus, the possibility for detecting response to perioperative chemotherapy is of great interest. A paradigm shift towards an individualised tailored therapy form has emerged in recent years, which potentially require a higher need for diagnosis on molecular level. Today most molecular biological methods apply tissue samples for in-vitro analyses, but new radiological tools provide opportunity for non-invasive examinations; for example PET can with a radioactive sugar compound: Flour-18 deoxyglucose (18F-FDG). This compound is injected through a catheter in a larger vein (media cubiti vein) and absorbed in cells with increased metabolism - especially cancer cells. A PET scanner registers the absorption; this radiology modality can provide valid information, which is essential for non-invasive tumour staging and monitoring response under a specific therapy.

A new diagnostic modality is PET scan combinated with magnetic resonance (PET/MR) simultaneously. So far, no studies have conducted an evaluation of simultaneous PET/MR scan to assess the perioperative chemotherapy response in patients with GEJ cancer. However, some studies suggest that commercially available PET/MR scanner might contribute in a diagnostic elucidation 6. Simultaneous PET/MR scan might in theory minimize the misinterpretations of potential response changes after chemotherapy, which can appear in the interval between separate PET, CT and MR scans 4,5,7. Studies have found PET scan of GEJ cancer could be helpful as a prognostic tool to differentiate between responders and non-responders during chemotherapy 9. Standardized uptake value (SUV) is a unit that display the absorption of 18F-FDG and is used routinely to quantify tumour glucose metabolism in PET scan 8. A change of more than 35 % in SUV measurements before and after the induction of chemotherapy is considered as the definition of responders and non-responders in earlier studies 9.

The MR technique is based on magnetic fields and radio waves. Diffusion Weighted Imaging (DWI) is a non-invasive MR-modality, which measures the changes in water diffusion (Brownian movements) throughout tissue. These changes are measured in Apparent Diffusion Coefficient (ADC), a parameter derived from DWI and reflects the change in diffusion 7. ADC and DWI can be used to differentiate between benign and malignant tumours, due to a larger cell density in malignant tumours. Consequently, malignant tissue has a decreased diffusion relative to normal tissue. ADC has been used as a factor in some studies to predict the response to chemotherapy 10. A single study has shown a rise in ADC-value two weeks after initiation of chemotherapy in patients with GEJ cancer, and demonstrated that the percentage change in ADC-value between the groups (responders and non-responders) is significantly different 11. Simultaneous PET- and MR scan might be very useful to evaluate the response to chemotherapy in patients with GEJ cancer compared with these parameters alone. The opportunity for a more individualised tailored treatment in future might be possible with PET/MR.

Inclusion Criteria:

• Patients with histologically verified adenocarcinoma of the Gastroesophageal junction and assessed as operable.

Exclusion Criteria:

• Under 18 of age
• Not given informed consent
• Other histological cancers besides adenocarcinoma
• Metal in the body that would contraindicate MR-imaging
• Allergies for contrast
• Claustrophobia
• Not a candidate for perioperative chemotherapy and surgery

All patients referred to the Department of Surgical Gastroenterology, Rigshospitalet, with histologically verified adenocarcinoma of the Gastroesophageal junction and assessed as operable. Informed consent is required from each patient