Abdominal recurrence was defined by one of the following events: local/abdominal or distant progressive disease during preoperative radiotherapy (as per RECIST 1.1), tumor or patient becoming inoperable (ASA score of 3 or involvement of superior mesenteric artery, aorta, or bone), peritoneal metastasis found at surgery, macroscopic residual disease left in at surgery (R2 resection), or local relapse (after macroscopically complete resection). Liver metastases were regarded as distant metastatic events. Patients with distant metastases were followed until local failure was detected. Patients without one of these events were censored at the date of last follow-up.

The acute toxicity was assessed in randomized patients who received at least one dose of preoperative radiotherapy (RT). It follows Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 in the RT safety population, assuming the following classification:

• Grade 0 No event reported
• Grade 1 Mild; asymptomatic or mild symptoms; clinical or diagnostic observations only; intervention not indicated.
• Grade 2 Moderate; minimal, local or noninvasive intervention indicated; limiting age-appropriate instrumental ADL.
• Grade 3 Severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated; disabling; limiting self care ADL.
• Grade 4 Life-threatening consequences; urgent intervention indicated.
• Grade 5 Death related to AE. For each item of the CTCAE, the worst grade of acute toxicity from the data of randomization prior to surgery was taken

Only patients in the "Preoperative radiotherapy followed by en-bloc surgery" that received at least one dose of radiotherapy were considered for this secondary endpoint.

Adverse events and side effects possibly related to surgery were assessed according to the Dindo's classification. This scales the observed side effects as Grade 0 No event Grade I Any deviation from the normal postoperative course without the need for pharmacological treatment or surgical, endoscopic and radiological interventions Grade II Requiring pharmacological treatment with drugs other than such allowed for grade I complications.

Grade III Requiring surgical, endoscopic or radiological intervention Grade IV Life-threatening complication (including CNS complications) requiring IC/ICU-management Grade V Death of a patient For each item, the frequency of the worst grade of the observed toxicity was tabulated by treatment group.

The perioperative period commence at the time of surgery (at the time of the induction anaesthesia) to the complete closure of the wound.

all randomized patients who are eligible and have started their allocated treatment (at least have been operated or received one fraction of irradiation followed by surgery)

The late toxicities occurring more than 60 days after surgery. These were reported using Common Terminology Criteria for Adverse Events (CTCAE) v4.0. This scales the observed toxicity from Grade 1 to 5, assuming the following classification:

• Grade 0 no event
• Grade 1 Mild; asymptomatic or mild symptoms; clinical or diagnostic observations only; intervention not indicated.
• Grade 2 Moderate; minimal, local or noninvasive intervention indicated; limiting age-appropriate instrumental ADL.
• Grade 3 Severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated; disabling; limiting self care ADL.
• Grade 4 Life-threatening consequences; urgent intervention indicated.
• Grade 5 Death related to AE. Full AE reporting can be found in the Adverse Event Section.

all randomized patients who are eligible and have started their allocated treatment (at least have been operated or received one fraction of irradiation)

For patients that received preoperative radiotherapy, the tumor response was assessed using RECIST 1.1. Response criteria were essentially based on a set of measurable lesions identified at baseline as target lesions, and followed at the end of the radiotherapy. Response was not the primary endpoint, so a confirmatory CT-scan was not mandatory.

Tumor response using RECIST 1.1 was only assessed for patients receiving preoperative radiotherapy.

Abdominal recurrence was defined in the ARFS section. The following are considered competing events:

• death in the absence of abdominal failure
• distant metastases diagnosed before abdominal failure Patients without one of these events were censored at the date of last follow-up.

Note that distant metastasis before or in the absence of abdominal recurrence was counted as a competing risk for this endpoint - whereas distant metastasis in the absence of abdominal recurrence was considered a non-event for the primary endpoint (ARFS). This explains the difference in non-events between the survival and interval endpoint.

Alive and metastases free patients will be censored at the date of last follow-up.

Alive patients were censored at the date of last follow-up. Causes of death were recorded and reported as a table.