The combination of neoadjuvant chemoradiotherapy (CRT) and total mesorectal excision (TME) is considered the standard treatment for locally advanced rectal cancer in the western world. Appropriate preoperative treatment and margin free surgery are key-elements in reducing the local-recurrence of the tumor and consequently improving overall survival.

Nevertheless, the local recurrence of stage II and III rectal cancer is still high, with current levels of 5% to 10% even when R0 resection is achieved. Most of the cases of loco-regional recurrence are associated with lateral lymph nodes (LLN) spread of cancer cells, which is not always controlled by the preoperative chemotherapy. As a matter of fact, the incidence of LLD metastases has been estimated to range from 11% to 22% in patients with T3/4 rectal cancer below the peritoneal reflection.

In order to improve these poor outcomes, Japanese surgeons have adopted extended lymphadenectomy with the dissection of lateral extramesorectal lymph nodes as the standard of care for T2-3 low rectal cancer patients5. While this approach is widely used in Japan and Korea, western surgeons have preferred a less aggressive approach, indicating lateral lymph node dissection (LLND) only in presence of clinically highly suspicious lateral pelvic lymph nodes on baseline magnetic resonance imaging (MRI). Thus, it is essential to identify preoperative predictive factors of LLN metastasis.

Even if MRI is considered the optimal diagnostic tool in rectal cancer, its accuracy for LLN staging is considered poor, especially after neoadjuvant treatment. LLNs often change in both features and size after CRT, and this behaviour might not be in concordance with the response of the primary tumor.

To the best of our knowledge, no consensus exists on whether the risk of local recurrence should be determined by assessing the features of LLN on the primary MRI or on the restaging MRI. Moreover, the relation between LLN response and primary tumor regression grade after neoadjuvant CRT needs to be thoroughly explored.

This multicenter cohort study aimed to investigate factors on primary and restaging MRI associated with lateral nodal recurrence and to identify patients who may benefit from LLND after neoadjuvant treatment for locally advanced rectal cancer.

The present is an international multicenter observational retrospective study. All consecutive patients admitted with a histologically confirmed locally advanced rectal adenocarcinoma [clinical T3-T4N0M0 or T(any)N+M0] treated with neoadjuvant CRT followed by curative intent elective surgery between January 2014 and December 2019 at four academic high-volume institutions: the Division of Colon and Rectal Surgery, Mayo Clinic, Rochester (MN), USA; the Department of Visceral Surgery, Centre Hospitalier Universitaire Vaudois (CHUV), Lausanne, Switzerland; the Division of Minimally-Invasive Surgery, ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy, and the Division of Colon and Rectal Surgery, IRCCS Humanitas Research Hospital, Rozzano - Milan.

All included patients had a pelvic MRI at baseline and a second pelvic MRI after neoadjuvant CRT. Interpretation and reporting of MRI for baseline staging was performed through a structured report adopted by each center involved in the present study. All MRI reports were presented according to international standards and validated by board-certified specialized lower gastrointestinal radiologists.

Inclusion criteria

• Aged 18 years and older
• Histological diagnosis of locally advanced rectal adenocarcinoma [clinical T3-T4N0M0 or T(any)N+M0] treated with neoadjuvant CRT followed by curative intent elective surgery

Esclusion criteria

• No MRI or pathologic data
• Histological diagnosis other than adenocarcinoma
• Urgent or emergency surgery
• Surgery without curative intent

The study population includes patients diagnosed with rectal cancer locally advanced, subjected to Chemo-Radiotherapy and surgical treatment with intent curative between January 2013 and December 2019, in each of the Centers involved. The data they will be collected from medical records and institutional data collection and follow-up databases