In vivo characterisation of soft tissue tumours by 1.5-T proton MR spectroscopy

Objectives To determine whether proton magnetic resonance spectroscopy (1H-MRS) can help differentiate between benign and malignant soft tissue lesions, and to assess if there is a correlation between 1H-MRS data and the mitotic index. Methods MR measurements were performed in 43 patients with soft tissue tumours >15 mm in diameter. Six cases were excluded for technical failure. Examinations were performed at 1.5 T using a single-voxel point resolved spectroscopy sequence (PRESS) with TR/TE=2000/150 ms. The volume of interest was positioned within the lesion avoiding inclusion of necrotic regions. In all patients, a histological diagnosis was obtained and the corresponding mitotic index was also computed. 1H-MRS results and histopathological findings were compared using the chi-squared test and correlation coefficient. Results Choline was detected in 18/19 patients with malignant tumours and in 3/18 patients with benign lesions. The three benign lesions included one desmoid tumour, one ossificans myositis and one eccrine spiradenoma. Choline was not detected in 15 patients with benign lesions or in one patient with dermatofibrosarcoma protuberans. Resulting 1H-MRS sensitivity and specificity were 95% and 83% respectively. Conclusions Absence of choline peak is highly predictive of benign tumours suggesting that 1H-MRS can help to differentiate malignant from benign tumours. Key Points 1H-MRS may allow differentiation between benign and malignant soft tissue lesions Absence of choline peak is highly predictive of benign soft tissue lesions Malignant tumours with a mitotic index >2/10 HPF had a positive choline peak A choline peak may still be identified in some benign tumours