Whole-body MRI with diffusion-weighted sequences compared with 18 FDG PET-CT, CT and superficial lymph node ultrasonography in the staging of advanced cutaneous melanoma: a prospective study
Conflict of Interest
Conflict of Interest
- None for J-C J., L.T., V.T., M.Y., C.J., I.M., F.D., F.G. and Y.B.; Luc Bracoud is technical engineer in ‘Bioclinica Company’ (Lyon France) that created the software used in the study to compare the different imaging modalities. This point is clearly stated in the ‘affiliation chapter’ of this co-author. However, we believe that it did not influence the results of the study as his company does not favour any of the imaging modality compared during this study.All authors declare that they have sufficiently participated in the submitted work to deserve authorship, all have had access to clinical material and have revised the manuscript before submission. Corresponding author (Pr Yves Berthezene) endorses the scientific responsibility of the reported work herein.
The aim of our study was to compare the diagnostic performances of non-radiating whole-body magnetic resonance imaging (wbMRI), either volumetric, with Volumetric interpolated breath-hold examination (VIBE) or metabolic, with diffusion-weighted sequences (wbMRI), with classical irradiating techniques such as PET-CT, CT and with lymph node ultrasonography (US) for the staging of advanced melanoma.
Patients and methods
Thirty-seven melanoma AJCC stage IV patients were prospectively included. All images were independently interpreted without prior knowledge of the results of studies performed with concurrent techniques, and all imaging techniques were scheduled within a mean interval of 7 days. The overall and site-specific diagnosis performances of each imaging modality were studied, as well as the interest of combined MRI VIBE and diffusion sequences.
The number of visceral or lymph node metastases spotted was, respectively, 218, with 125 metastases for wbMRI, 191/103 for PET-CT, 209/115 for CT and 33/13 for lymph node US. No statistically significant difference (P < 0.05) of overall diagnostic performances between wbMRI (Se 84%, Sp 87.1%, PPV 89.8%, NPV 80.2%) and PET-CT (Se 79.8%, Sp 93.1%, PPV 93.2%, NPV 79.4%) was observed. No statistically significant difference was found between wbMRI and PET-CT with two channels for CT with respect to different metastatic sites. Compared with the CT, wbMRI had significantly better overall specificity (P = 0.0011) and PPV (P = 0.02). For lung exploration, sensitivity of wbMRI (51.6%) was inferior to CT (71.4%). To detect superficial metastatic lymph nodes, wbMRI and US both showed high diagnostic accuracy with no statistically significant difference. Intra-observer agreement was almost perfect for all imaging modalities considering the overall staging. Inter-observer agreement for wbMRI and diffusion alone was almost perfect except for bone and lymphatic sites. Overall diagnostic performance of diffusion alone was significantly inferior to those of combined VIBE and diffusion sequences.
Whole-body MRI, using diffusion weighted sequences, was a reliable non-radiating imaging for staging of melanoma and offers the same diagnostic performances than combined CT, PET-CT and lymph node US.