Umeå universitets logga

Background: The evaluation of metastatic spread to locoregional lymph nodes in rectal cancer is essential for patient management. This thesis evaluated the added value of hybrid imaging using 18F-fluoro-2-deoxy-D-glucose-Positron Emission Tomography/Magnetic Resonance Imaging (FDG-PET/MRI) for nodal staging. To achieve this, we presented a method for the anatomical matching of individual mesorectal nodal structures between imaging and histopathology and evaluated the current European Society of Gastrointestinal and Abdominal Radiology (ESGAR) consensus criteria for nodal staging.

Methods: Sixty-five patients (38 men, median age 70 years, interquartile range: 61–74) were consecutively included in Region Västerbotten as part of the prospective observational study RECTOPET (Rectal Cancer Trial on PET/MRI/CT). All patients had preoperative FDG-PET/CT and FDG-PET/MRI examinations, MRI of the surgical specimen, and a histopathological finding-by-finding description that facilitated anatomical matching of individual nodal structures between imaging and histopathology. Mixed-effects logistic regression models were used to evaluate the ESGAR consensus criteria for nodal staging, with histopathology as the gold standard. Additionally, the accuracy of FDG avidity (uptake above background level) and the combination of FDG avidity with ESGAR criteria (FDG-PET/MRI) for malignant nodal assessment was compared to standalone MRI.

Results: Study I described the feasibility of incorporating FDG-PET/CT and FDG-PET/MRI into the clinical workflow for patients with rectal cancer. Study II illustrated that the described methodology enabled the matching of 47% (92/197) of the mesorectal nodal structures detected at preoperative MRI; 25% (23/92) were malignant, with 48% (11/23) histopathologically verified tumour deposits. Study III demonstrated the strongest associations for nodal malignancy with a short-axis diameter of ≥ 5 mm (odds ratio [OR] 21.43; 95% confidence interval [CI] 4.13–111.29) and heterogeneous signal (OR 9.02; 95% CI 1.33–61.08). Additionally, the OR of malignant outcomes for nodes with positive ESGAR was 8.23 (95% CI 2.15–31.50), with corresponding sensitivity and specificity of 54% and 85%, respectively. In Study IV, N-staging with FDG-PET/MRI was equal to the histopathological nodal stage in 45% of the patients, compared to 37% for MRI alone. Furthermore, 75% (78/104) of the malignant nodal structures were FDG-avid, with a sensitivity of 75% and a specificity of 79% for FDG-PET.

Conclusion: The added value of FDG-PET/MRI in N-staging mainly lies in its ability to reduce the prevalence of false-negative nodes, which can be beneficial for organ preservation.