FDG PET/MRI for evaluation of nodal mesorectal structures in rectal cancer: A matched comparison to histopathology

FDG PET/MRI is a promising imaging modality for nodal staging in rectal cancer; however, its role remains to be defined. We aimed to assess its performance in detecting mesorectal malignant lymph node involvement based on both metabolic and morphological criteria at PET/MRI versus at MRI alone.

Sixty-five patients (median age 70 years, IQR 61-74; 39 men) were examined with FDG PET/MRI followed by individual anatomical matching of mesorectal nodal structures between histopathology and MRI. PET N-stage assessment was evaluated using FDG uptake over background levels, MRI N-stage by the 2016 European Society of Gastrointestinal and Abdominal Radiology (ESGAR) consensus criteria and PET/MRI was evaluated using both in combination. Histopathological assessment served as gold standard, and the accuracy of identifying malignancy at both nodal and patient level was calculated. Furthermore, FDG PET/MRI and MRI using ESGAR criteria for nodal restaging after neoadjuvant treatment were also evaluated.

In total, 835 nodal structures were matched, of which 104 were malignant (12%); among these, 59/104 (57%) were histopathologically proven lymph node metastases. MRI alone yielded a sensitivity of 54% and specificity of 85% for nodal malignancy, while the corresponding estimates for FDG-avidity gave a 75% sensitivity and 79% specificity. The different combined FDG PET/MRI criteria for malignancy were evaluated: FDG-positivity or malignancy according to ESGAR criteria resulted in a sensitivity of 76%; while the combination of FDG-positivity and malignancy according to ESGAR criteria achieved a specificity of 90%.

Compared to MRI alone, FDG PET/MRI offers potential added value by reducing the risk of nodal understaging.