Contributed by Ali Mahta, MD1#, Ewa Borys, MD2#, Scott R. Vandenberg, MD, PhD2, Bob Carter, MD, PhD3, Santosh Kesari, MD, PhD1
# co-first authors
1Department of Neurosciences, Moores Cancer Center, 2Department of Pathology, Neuropathology division, and 3Department of Neurosurgery, University of California, San Diego, La Jolla, CA
CLINICAL HISTORY AND IMAGING
A 72 year old female with no significant past medical history who was under treatment and follow up for her chronic low back pain was referred to our center for evaluation of an incidentally discovered intramedullary lesion in her thoracic spine after her recent fall. She didn't have any neurologic deficits at the time, including weakness or any bladder or bowl dysfunction and on exam she was neurologically intact. On her further work up, MRI of thoracic spine revealed a well-defined intramedullary lesion with an intradural component measuring about 1 cm at the level of T10. (Fig 1, left) There was a mildly increased T2 signal with enhancement after gadolinium administration. Chronic superior endplate compression fracture of T9 and L1 vertebral body was also noted. The intramedullary lesion was considered an incidental finding unrelated to her symptoms therefore the initial plan was close observation, given her unwillingness for any surgical intervention, in addition to the small size of the lesion, lack of neurological deficits and its benign appearance on imaging. Based on imaging, the differential was a low-grade astrocytoma, ependymoma or an intramedullary schwannoma. After about 3 years, in a follow up visit, we noticed a mild weakness in her left lower extremity and some gait unsteadiness. This was accompanied by a significant increase in the size of the lesion (Fig 1, right), so she agreed to an excisional biopsy. Intraoperatively the lesion was noted to be a bluish, firm mass with a good plane at the level of T10/11.
H&E stained sections showed a discrete non-infiltrative pigmented spindle cell neoplasm with moderate to high cellularity. Interwoven fascicles of neoplastic cells ranged from plumply spindle shaped with ovoid nuclei to more overtly fusiform with slender bipolar nuclei. In a few fields clusters of tumor cells had polygonal contours and appeared vaguely epithelial. (Fig 2) Melanin was distributed as a fine dusting within fusiform cells and clumpy aggregates in melanophages. (Fig 3) Nucleoli were distinct or slightly prominent only in a few fields, and some nuclei had longitudinal grooves. Prolonged scrutiny was required for the identification of even a rare mitotic figure and there was no necrosis. The tumor cells were positive for S-100, HMB-45 (Fig 4), and MART-1. An iron stain was negative confirming the melanotic nature of the brown granular pigment. A MIB-1 preparation labeled less than 1% of tumor nuclei. Electron microscopy showed that the melanin containing tumor cells lacked basal lamina. In addition, in the cytoplasm of these cells, a full range of melanized structures was observed in various stages of maturation ranging from the premelanosomes to fully melanized organelles. (Fig 5) Premelanosomes were abundant and had the characteristic "zipper"-like cores. (Figs 6 and 7) What is your diagnosis?