Radiological evaluation has been positioned as vital in clinical management since the treatment of rectal cancer has evolved rapidly and has now entered the era of individualization [1], while the regional nodal status, which is of great prognostic importance, has been omitted in most studies.

    Extranodal extension (ENE) refers to the extension of tumor cells from metastatic lymph nodes into perinodal fatty tissue through the thin fibrous capsule that envelops the entire lymph node on histopathology. With an incidence ranging between 22% and 68% in colorectal cancer, ENE has been broadly proven to indicate a worse prognosis [2-4] and to be associated with distant lymph node metastasis [5] and poorer outcomes following adjuvant chemotherapy [6], so a new method to earlier evaluate ENE on pretreatment MRI would be of substantial value in guiding treatment.

    Although there is a well-established characteristic for detecting ENE in cervical lymph nodes in fat-suppressed T2-weighted imaging [7], its application was limited in rectal cancer as it is not a requested sequence. Besides, the manifestation of the chemical shift effect (CSE) in T2-weighted imaging, has been shown to help determine nodal status [8, 9], but is generally perceived as unsuitable for diagnosis [10]; theoretically, opposed-phase MRI could more robustly provide similar information [11], but still underexplored in rectal MRI. The potential characteristics for ENE detection are those already applied in determining the benignity or malignancy of lymph nodes, but which and how they contribute need to be further studied.

    Therefore, the purpose of this study was to evaluate the performance of pretreatment MRI in detecting ENE using a node-by-node matched pathological reference standard and to assess its prognostic value in rectal cancer patients.