Rationale: Esophageal cancer (EC) is the seventh most frequently diagnosed cancer and the sixth leading cause of cancer-related death worldwide. As a result of the late onset of symptoms, most patients with EC present in an advanced stage with a corresponding poor prognosis. Poor disease outcome after surgery alone (5-yr overall survival between 25-40%) prompted many researchers to explore neoadjuvant chemoradiotherapy (nCRT) or neoadjuvant or perioperative chemotherapy (nCT/pCT) approaches.

In the Netherlands, neoadjuvant chemoradiation has become standard of care for esophageal cancer since publication of the CROSS trial showing a benefit of nCRT over surgery alone for both adenocarcinoma (AC) and squamous cell carcinoma (SCC) (van Hagen et al., 2012). However, the benefit of nCRT was less pronounced in AC, which was also reflected by pathologic complete response (pCR) rates: 23% in AC vs. 49% in SCC. Furthermore, SCC and AC differ in patterns of recurrence after nCRT or chemotherapy. AC is more likely to develop distant metastases while SCC has a predisposition for locoregional recurrences. This difference in response to nCRT and in recurrence pattern indicates that histology-tailored treatment strategies should be explored. In the modern multidisciplinary discussion on the optimal approach to locally advanced adenocarcinoma of the esophagus and junction, both a trimodiality approach or perioperative chemotherapy are acceptable and evidence based. Therefore both are viable options within current guidelines.

As mentioned above, patients with an AC of the esophagus are especially prone to develop distant recurrences. In addition, response to nCRT is only moderate in AC. Therefore, the investigators hypothesize that the ideal neoadjuvant treatment should consist of adding MR-guided radiotherapy to standard pCT in order to achieve maximum systemic control and achieve maximum local control.

Objective: The main objective of this study is to determine the maximum tolerated dose (MTD) of 5 fractions MRgRT for patients with AC following FLOT therapy. The secondary objectives are feasibility, non-dose limiting toxicity, oncological outcomes and to explore variables for early response evaluation.

Study design: 6+3 dose-escalation design with 4 radiotherapy dose levels. Study population: Patients with a resectable esophageal adenocarcinoma who are eligible for nCRT and surgery and who are eligible for MRgRT.

Intervention: 5 sequential, homogenous fractions of 4-8 Gy within 2 weeks on the gross tumor volume (GTV) following preoperative FLOT (as part of standard perioperative chemotherapy) using MR-guided online adaptive radiotherapy on the MR-linac. Start in dose level 0, of 5 x 5Gy per patient, and if safe this is increased step-wise to a maximum dose level 3 of 5 x 8Gy per patient.

Main study parameters/endpoints: The primary endpoint is the incidence of a dose limiting toxicity (DLT). Early DLT is defined as radiation induced esophageal fistula/ perforation/ hemorrhage/ necrosis or tracheal, bronchial or bronchopleural fistula/tracheal or bronchopulmonary hemorrhage grade ≥ 3 or any non-hematological grade ≥ toxicity, assessed clinically significant and related to the radiotherapy, according to Common Toxicity Criteria for Adverse Events (CTCAE) version 5.0 occurring within 16 weeks after the start of radiotherapy and before surgery or postponing of surgery > 16 weeks after the end of radiotherapy due to any grade of treatment-related toxicity. Subacute DLT is defined as peri- and/or postoperative complications occurring within 30 days after surgery, defined as postoperative anastomotic leakage or pneumonitis ≥ 3b according to Clavien-Dindo.

Secondary endpoints are non-DLT toxicity, the technical feasibility of dose delivery, perioperative complications. and oncological outcomes including R0 resection rate, histopathological tumor response, local and regional recurrence and death from any cause.

Nature and extent of the burden and risks associated with participation, benefit and group relatedness: The benefits for the patients may include higher probability of complete primary tumor and lymph node metastases response that initially lead to increased survival and could eventually result in organ-sparing treatment programs. Possible risks are mainly esophageal fistula/perforation and broncho-esophageal fistula or hemorrhage.

Inclusion Criteria:

• Histologically confirmed adenocarcinoma of the esophagus or GE- junction (Siewert I or II)
• Potentially resectable, locally advanced esophageal tumor (cT1bN+, cT2-3, N0-3, M0) based on standard primary staging by EUS and 18F-FDG PET-CT
• Eligible for neoadjuvant treatment: followed by esophagectomy (as judged by the multidisciplinary tumor board)
• Eligible for pCT FLOT
• Tumor length ≤ 10 cm
• Age ≥ 18 years
• WHO performance status 0-2
• Signed informed consent
• Tumor volume that can be defined on MRI at baseline (T2w and DW-MRI)
• Written informed consent must be given according to ICH/GCP, and national/local regulations.

Exclusion Criteria:

• Squamous cell carcinoma

• Non-resectable, inoperable or metastatic adenocarcinoma of the esophagus or GE junction

• Siewert type III tumors

• Prior (chemo)radiotherapy to the mediastinum

• Prior esophageal surgery that impedes the ability to perform an esophagectomy

• Patients with multiple primary carcinomas of the esophagus

• Patients who meet exclusion criteria for MRI according to the MRI contraindications screening list of the imaging and oncology division of the UMC Utrecht

  * Irradical endoscopic mucosal resection (EMR) or endoscopic submucosal dissection (ESD) of primary tumor prior to start of neoadjuvant chemoradiotherapy

• Pregnant or breast-feeding patients

• Presence of any psychological, familial, sociological or geographical condition potentially hampering compliance with the study protocol and follow-up schedule before patient registration. Patients in whom it is not in their best interest to participate (in the judgment of the PI)

Undecided. Plans for individual participant data (IPD) sharing have not been finalized at this time. Any future data sharing will depend on institutional policies, ethical approvals, and participant consent.