SUN-459 Analysis of the Evolution of Postoperative Pituitary Resection Cavities Assessed by Magnetic Resonance Imaging and Implications Regarding Choice of Radiation Therapy Modality

Surgical resection is the standard initial therapy for patients with symptomatic non-prolactin secreting adenomas and other pituitary tumors (1). After surgical resection select patients will require adjuvant radiation therapy. There are a number of radiation modalities available for treatment. One factor that influences the choice of radiation modality is proximity of tumor to local critical anatomy, namely the optic chiasm (2). MRI is the modality of choice for evaluating pituitary tumors, postoperative resection cavities and the relationship between tumor and the optic chiasm (3). To study the expected evolution of postoperative pituitary resection cavities and the relationship between residual sellar tissue and the optic chiasm we retrospectively reviewed MRI’s of patients with surgically resected pituitary tumors obtained during the preoperative (PreO), immediate postoperative (IPO) (<72 h) and delayed postoperative (DPO) (>3 mo) phases. 91 patients (45 F, 46 M, mean age 50.9 yrs) with functional adenomas (26.4%), non-functional adenomas (46.2%), Rathke cleft cysts (9.9%), craniopharyngioma’s (3.3%) and other lesions (14.2%) were studied. PreO estimated tumor volume was 4.3 cc (0-28.7), craniocaudal (CC) dimension 17.2 mm (2-49), and distance between tumor and optic chiasm 2.0 mm (0-10). 9.9 % of patients had cavernous sinus invasion by imaging criteria. On IPO imaging 83.5 % of patients had gross total resection. Average IPO and DPO resection cavity estimated volumes were 3.3 cc (0.1-16.3) and 1.6 cc (0.0-8.7), respectively. Average % decrease in cavity volume from IPO to DPO scans was 51.5%. Average distance between residual sellar tissue and chiasm was 2.5 mm on IPO and 4.7 mm on DPO scans. The distance between tissue and chiasm on IPO and DPO scans was greater for tumors <10 mm vs >10mm (p=0.019). Percent change in CC dimension of resection cavities from IPO to DPO scans was higher for cavities without fat packing (63.2%) vs cavities with fat packing (52.1%) (p=0.025). Tumor histology, cavernous sinus invasion, degree of PreO chiasm mass effect, and presence of fluid or blood within the IPO cavity did not correlate with distance between tissue and chiasm on DPO scans. Conclusion: There is a significant reduction in cavity volume and increased distance between the chiasm and residual sellar tissue on DPO vs IPO scans. Evaluation for radiosurgery as a treatment option, which has less total and chiasm dose, may be an option after reviewing DPO scans. References: (1) Rim et al., Radiat Oncol J. 2011;29(3):156-63. (2) Minniti et al., Radiat Oncol. 2016;(11)135. (3) Patel et al., World Neurosurg. 2014;82(5):770-80.