PET/MRI Characterization of Mucinous Versus Nonmucinous Components of Rectal Adenocarcinoma: A Comparison of Tumor Metabolism and Cellularity

Abstract

OBJECTIVE. The purpose of this study was to evaluate whether FDG PET/MRI can be used to differentiate the mucinous from the nonmucinous components of primary rectal tumors and to compare the glycolytic metabolism on PET with tumor cellularity on DWI in both components. SUBJECTS AND METHODS. Ninety-nine patients who underwent FDG PET/MRI for staging of primary rectal cancer were included in this prospective analysis. MRI depicted the mucin component through the tumor volume. Separate volumes of interest were drawn on both mucinous and nonmucinous components and propagated to PET and apparent diffusion coefficient (ADC) mapping. Maximum and mean standardized uptake values (SUVmax, SUVmean) and maximum, mean, and minimum ADC values (ADC(max), ADC(mean), ADC(min)) were recorded and compared between areas with mucinous and nonmucinous components. Whole-body PET/MRI was also used to evaluate for the presence of distant metastases. Nonparametric testing was used to compare the two groups of patients: those with tumors with a mucinous component and those with tumors without a mucinous component. Logistic regression analysis was performed to calculate the association risk between mucinous component and metastatic disease. RESULTS. Seventeen patients (17.2%) had a mucinous component within the tumor on T2-weighted MRI. Most of these patients had advanced disease, the mucinous component tumors being in significantly higher T categories than the tumors without a mucinous component (88.2% vs 61.0%; p = 0.032). SUVmax (7.4 vs 16.7; p = 0.002) and SUVmean (5.4 vs 13.4; p = 0.001) were significantly lower in tumors with a mucinous component than in those without a mucinous component. Tumor ADC measurements were not different between tumors with and those without a mucinous component (ADC mean, 1.4 vs 1.6; p = 0.361). There was no association between presence of a mucinous component within the primary rectal tumor and presence of synchronous metastases (odds ratio, 1.1 [0.4-3.0]; p = 0.904). Moreover, the occurrence of metastases in patients with mucinous component tumors (7/17 [41.2%]) was not different from that in patients with tumors without a mucinous component (28/82 [34.1%]) (p = 0.887). CONCLUSION. PET/MRI can be used to differentiate the mucinous and nonmucinous components within primary rectal adenocarcinoma on the basis of metabolic status. The FDG uptake is significantly lower in the mucinous component, but tumor cellularity based on MRI and DWI findings is not. Despite being associated with a higher T category in the sample of patients in this study, the presence of a mucinous component seems not to be associated with increased risk of synchronous metastases.