Abstract

Purpose Non-invasive characterization of lipomatous tumors can be challenging as several histological types have similar imaging characteristics. In this study we examine the use of a new biomarker based on spin coupling related signal loss between two acquisitions of different echo spacing to differentiate between benign lipomas, well, intermediate and poorly differentiated liposarcomas (l, wdl, idl and pdl, respectively). This study was based on previous work showing differences between vegetable oils of different botanical origin using the same protocol [1] . Methods Fourteen patients (9 male, 5 female, age: 37–87, mean 58) with soft tissue masses (5 lipomas, 2 myxoid, 5 dedifferentiated, 2 pleiomorphic liposarcomas) underwent MRI prior to any therapeutic intervention. MRI protocol, among other sequences, included two Multi Echo Spin Echo CPMG sequences with different echo spacing, 13.4 and 26.8 ms respectively, i.e. above and below the approximate threshold of 20 ms in order to have bright and dark fat appearance on T2-w images. All surgically excised specimen were histopathologically examined to determine the kind of lipomatous tumor and to localize sites of well or poor differentiation in the cases of dedifferentiated liposarcomas as distance from the upper tumor limit (z) and distance from the center (x,y). Relative signal loss between bright and dark fat images on TE 80 ms was calculated in order to measure the spin coupling Ratio (Rsc), defined as mean ROI value in the lesion divided by the same value in normal subcutaneous fat for the same patient. Results Mean (SD) of Rsc for l, wdl, idl and pdl was 1.036 (0.06), 0.77 (0.18), 0.055 (0.06) and −0.16 (0.57), respectively. Conclusions A new biomarker related on spin coupling signal loss is indicative of the differentiation grade of lipomatous tumors, with special interest regarding the clinically challenging question of benign lipomas vs. well differentiated liposarcomas. It is of note that Rsc decreases with increased differentiation grade (1–3).

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call