Hippocampal Radiation Exposure and Cognitive Outcomes After Single-Fraction Gamma Knife Radiosurgery for Benign Sellar and Parasellar Lesions: A Prospective Study.

Introduction: This study evaluates hippocampal dosimetry and neurocognitive outcomes in patients undergoing Gamma Knife radiosurgery (GKRS) for benign sellar, parasellar, and suprasellar lesions. While hippocampal protection has been well studied in whole-brain radiotherapy (WBRT), its relevance in stereotactic radiosurgery (SRS) remains underexplored. This study systematically investigates hippocampal radiation dose thresholds and associated neurocognitive outcomes in patients undergoing single-fraction Gamma Knife radiosurgery specifically for benign sellar and parasellar lesions. Although cognitive outcomes after stereotactic radiosurgery have been previously studied in various contexts, specific hippocampal radiation thresholds and their cognitive impacts in this particular patient population have not been clearly defined.

Methods: A prospective analysis was conducted on 41 patients who underwent GKRS, with hippocampal dosimetry assessed using dose-volume parameters (Dmin, Dmax, Dmean, D40, and D100) following RTOG 0933 contouring protocols. Neurocognitive function was evaluated at baseline and six months post-treatment using standardized neuropsychological tests, including assessments of memory, executive function, and emotional well-being. The relationships between hippocampal radiation exposure and cognitive changes were analyzed through correlation and regression models.

Results: The mean prescription dose was 12 Gy (range: 10-16 Gy), with a steep dose gradient facilitating hippocampal sparing. ROC curve analysis identified hippocampal D40 ≥ 2.5 Gy and D100 ≥ 2.4 Gy as critical thresholds for neurocognitive decline, and these cutoff values were used in multivariable logistic regression analysis. Significant associations were found between higher hippocampal dose exposure and cognitive decline, particularly in verbal and memory retention domains. A 20% decline in verbal learning (Oktem Verbal Learning Test) was significantly associated with D40 ≥ 2.5 Gy (OR = 3.04, p = 0.006) and D100 ≥ 2.4 Gy (OR = 4.3, p = 0.031). Similarly, a 20% decline in memory retention (WMS Immediate Recall) was significantly linked to D40 ≥ 2.5 Gy (OR = 3.10, p = 0.041) and D100 ≥ 2.4 Gy (OR = 5.3, p = 0.045). Other factors, including age, gender, education level, and hippocampal volume, were not significantly associated with cognitive decline.

Conclusion: This preliminary study suggests that even relatively low-dose hippocampal radiation exposure in GKRS may potentially contribute to memory impairment. These initial findings provide insights into possible hippocampal dose thresholds specifically for single-fraction GKRS in benign lesions. However, larger prospective studies with longer follow-up periods are essential to validate these observations before recommending routine incorporation of hippocampal-sparing strategies into GKRS planning.