Preliminary Results of PET/MRI in Assessing Gastrointestinal Cancer: New Technology to Advance Our Staging Ability?

Sasan Partovi, MD, Deborah S Keller, MD, Raj Paspulati, MD, Rodney Ellis, MD, Brian Traughber, MD, Peter Faulhaber, MD, Conor P Delaney, MD, MCh, PhD

University Hospitals-Case Medical Center

Background: MRI and PET are excellent options for staging gastrointestinal malignancies. Hybrid PET/MRI is a novel option, with few scanners available in the US. PET/MRI leverages the superior spatial resolution of an MRI with the high sensitivity of a PET scan to optimize tumor diagnosis and staging. The goal of this preliminary study was to determine the feasibility of PET/MRI in gastrointestinal cancer imaging, and to compare its diagnostic ability with PET/CT for diagnosis and staging.

Materials and Methods: 12 patients (50% males, mean age 66±11 years), with gastrointestinal cancer (2 esophageal, 2 pancreatic, 5 colon, 4 rectal cancers, and 1 anal) had their tumors sequentially evaluated with PET/CT and a combined PET and 3.0T MRI system. Two rectal cancer patients received neoadjuvant therapy before imaging. The main outcome measures were the maximal and mean standard uptake value (SUVmax and SUVmean) and the longest and shortest tumor diameters for PET/CT and PET/MRI images. Both the primary tumor and metastatic lesions were evaluated. Pearson R correlation coefficients were used to compare the PET/CT and PET/MRI SUV values, and students t-test was used to compare the longest and shortest diameters.

Results: PET/MRI detected all primary (n=8) and metastatic (n=16) lesions found with PET/CT. The mean longest/shortest diameters were not statistically different for the primary (p=0.81 /p=0.62) or metastatic lesions (p=0.98/p=0.90) between the 2 methods. The SUVmax and SUVmean correlations for the primary and metastatic lesions between PET/MRI and PET/CT were also high (R=0.96 and R=0.89, respectively).

Conclusion: Our preliminary study revealed diagnostic ability equivalent to PET/CT in primary and metastatic lesions, even after neoadjuvant therapy. Based on our pilot study results, PET/MRI is feasible for assessing gastrointestinal cancer, and may be particularly suitable for rectal cancer if it offers the metastatic assessment of PET/CT in conjunction with the higher resolution of circumferential margin assessment in the pelvis. Further research is needed to validate results in a larger population and define a precise and appropriate role for PET/MRI in staging.

Session: Poster Presentation

Program Number: P374

« Return to SAGES 2013 abstract archive