Preoperative Image-guided Identification of Response to Neoadjuvant Chemoradiotherapy in Esophageal Cancer

Rationale: For locally advanced esophageal cancer the standard treatment consists of 5 weeks of neoadjuvant chemoradiotherapy (nCRT) followed by surgery. Surgery is currently performed independent of the response to nCRT and is associated with substantial morbidity. Prior knowledge of the eventual response to nCRT would greatly impact on the optimal care for many esophageal cancer patients for two imperative reasons:

Firstly, it is argued that patients who achieved a pathologic complete response (pCR, 29%) may not have benefitted from surgery. Consequently, proper identification of pathological complete responders prior to surgery could yield an organ-preserving regimen avoiding unnecessary toxicity.

Secondly, non-responders are exposed to the side effects of nCRT without showing any tumor regression. Early identification of the non-responders during nCRT would be beneficial for this group as ineffective therapy could be stopped, and for who altered treatment strategies could be explored.

Objective: To develop a multimodal model that predicts the probability of pathologic complete response to nCRT in esophageal cancer, by integrating diffusion weighted magnetic resonance imaging (DW-MRI) and dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) in conjunction with combined 18F-fluorodeoxyglucose positron emission tomography and computed tomography (18F-FDG PET-CT) scans acquired prior to, during and after administration of nCRT.

Study design: Multi-center observational study

Study population: Patients (>18 years) with potentially resectable locally advanced squamous cell- or adenocarcinoma of the esophagus or gastroesophageal junction, receiving nCRT prior to surgery.

Intervention: In addition to the standard diagnostic work-up for esophageal cancer that includes a 18F-FDG PET-CT scan at diagnosis and after nCRT, one 18F-FDG PET-CT scans will be performed during nCRT, as well as three MRI scans (before, during and after nCRT) within fixed time intervals. Furthermore, after response imaging after nCRT has been performed, but prior to surgery, patients will undergo (on an opt-out basis) an endoscopy and/or endoscopic ultrasonography (EUS) with biopsies of the primary tumor site, other suspected lesions and suspected lymph nodes. Furthermore, blood samples will be collected at three time points.

Main study parameters/endpoints: An accurate multimodal prediction model for the patients' individual probability of pathologic complete response after nCRT, based on the quantitative parameters derived from a longitudinal series of DW-MRI, DCE-MRI and 18F-FDG PET-CT datasets.