Assessment of metastatic colorectal cancer with hybrid imaging: comparison of reading performance using different combinations of anatomical and functional imaging techniques in PET/MRI and PET/CT in a short case series

• Published in: European journal of nuclear medicine and molecular imaging 2016-01, Vol.43 (1), p.123-132
• Main Authors: Brendle, C., Schwenzer, N. F., Rempp, H., Schmidt, H., Pfannenberg, C., la Fougère, C., Nikolaou, K., Schraml, C.
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Cardiology; Case studies; Child; Colorectal cancer; Colorectal Neoplasms - diagnosis; Colorectal Neoplasms - diagnostic imaging; Colorectal Neoplasms - pathology; Diffusion Magnetic Resonance Imaging; Female; Fluorodeoxyglucose F18; Humans; Imaging; Male; Medical imaging; Medicine; Medicine & Public Health; Metastasis; Middle Aged; Multimodal Imaging; Neoplasm Metastasis; Nuclear Medicine; Oncology; Original Article; Orthopedics; Positron-Emission Tomography; Radiology; Retrospective Studies; Tomography; Tomography, X-Ray Computed; Young Adult
• ISSN: 1619-7070; 1619-7089
• DOI: 10.1007/s00259-015-3137-z

Purpose The purpose was to investigate the diagnostic performance of different combinations of anatomical and functional imaging techniques in PET/MRI and PET/CT for the evaluation of metastatic colorectal cancer lesions. Methods Image data of 15 colorectal cancer patients (FDG-PET/CT and subsequent FDG-PET/MRI) were retrospectively evaluated by two readers in five reading sessions: MRI (morphology) alone, MRI/diffusion-weighted MRI (DWI), MRI/PET, MRI/DWI/PET; and PET/CT. Diagnostic performance of lesion detection with each combination was assessed in general and organ-based. The reference standard was given by histology and/or follow-up imaging. Separate analysis of mucinous tumours was performed. Results One hundred and eighty lesions (110 malignant) were evaluated (intestine n  = 6, liver n  = 37, lymph nodes n  = 55, lung n  = 4, and peritoneal n  = 74). The overall lesion-based diagnostic accuracy was 0.46 for MRI, 0.47 for MRI/DWI, 0.57 for MRI/PET, 0.69 for MRI/DWI/PET and 0.66 for PET/CT. In the organ-based assessment, MRI/DWI/PET showed the highest accuracy for liver metastases (0.74), a comparable accuracy to PET/CT in peritoneal lesions (0.55), and in lymph node metastases (0.84). The accuracy in mucinous tumour lesions was limited in all modalities (MRI/DWI/PET = 0.52). Conclusions PET/MRI including DWI is comparable to PET/CT in the evaluation of colorectal cancer metastases, with a markedly higher accuracy when using combined imaging data than the modalities separately. Further improvement is needed in the imaging of peritoneal carcinomatosis and mucinous tumours.