Kristen W. Yeom
A few years ago we encountered a series of very rare and aggressive chordomas of the skull base and sacrum in four young children, referred to as poorly differentiated chordomas. While it is commonly known that chordomas are T2-bright, these were very dark. So dark, in fact, that we had initially considered them embryonal tumors. We also noted that they had a pattern of decreased water movement on diffusion-weighted imaging, which was also similar to embryonal tumors. They were, however, distinguishable at the molecular level, as they had unusual histologic features of macronucleoli and mitotic figures and lacked nuclear expression of SMRCB1/INI1, while demonstrating brachyury expression, a highly specific chordoma marker.
Classic chordomas and chondrosarcomas of the skull base, on the other hand, are often difficult to distinguish based on conventional imaging features, and the classic locations for these tumors do not always hold up. Recognizing the unique morphological features of poorly differentiated chordoma, myxoid stroma of classic chordoma, and cartilaginous matrix of chondrosarcomas, we surmised that water diffusion properties likely differ among these three tumor types. The hypothesis was supported by our data, and we were happy to see that, for the first time, an MR imaging modality could definitively distinguish the tumors.
The current therapeutic strategy for these lesions is complete surgical resection when possible, followed by therapies such as proton beam or stereotactic radiosurgery. As new technologies emerge, we eventually may see the field change with improved preoperative diagnosis. Thus, we approach these types of studies with a future clinical utility in mind, ie, thinking imaging advances such as this may bring us to “where the puck is going to be,” rather than “where the puck is now.” As for now,