My research has focused on advancing MR imaging techniques, especially noncontrast techniques, for various applications in musculoskeletal tumor imaging, including the detection, characterization, and assessment of treatment response and the differentiation of postsurgical nodular scar from recurrence. A particular challenge in our field is the characterization of peripheral nerve sheath tumors (PNSTs), as benign and malignant PNSTs share imaging and clinical features. PNST characterization is remarkably challenging in patients with peripheral nerve tumor syndromes, such as neurofibromatosis (NF) type 1 or 2 and schwannomatosis, as these patients often have a multitude of peripheral nerve tumors throughout their bodies, with a 10% lifetime risk of undergoing malignant degeneration (in the case of NF-1).
Proton MRS is a technique that is noninvasive, requires no intravenous contrast administration, and provides information regarding the metabolic content of a region of interest. Our research indicates that quantitative proton MRS can be used to differentiate neurofibromas from malignant peripheral nerve sheath tumors by quantifying the contents of biomarkers of malignancy (i.e., the trimethylamines or choline-containing metabolites), and is thus a useful method for characterizing isolated PNSTs and tumors in patients with NF-1. Hence, MR imaging performed with MRS is a potential method for triaging patients toward biopsy (if MRS shows high biomarker content) or toward follow-up (if MRS shows low or negligible biomarker content).
Currently, MR imaging is the mainstay technique for evaluating patients with isolated PNSTs or with PNST syndromes, and we have previously shown that quantitative DWI with ADC mapping is also a useful technique for differentiating benign from malignant PNSTs. For musculoskeletal tumor imaging, while quantitative DWI is increasingly incorporated into clinical MR imaging tumor protocols as a simple and fast technique, quantitative proton MRS has not evolved into a clinically routine procedure, as this technique requires 15 minutes for scanner acquisition (shimming time in addition to an approximately 5-minute acquisition time), as well as additional time for postprocessing to determine metabolite concentrations. One potential alternative is to utilize a qualitative approach to MRS analysis, in which metabolite content is not quantified, but rather qualitatively assessed as present or absent; with a qualitative approach, postprocessing time is reduced, as is acquisition time slightly.