2170 - Predictors of Leptomeningeal Disease after Postoperative Stereotactic Radiosurgery for Resected Brain Metastases: A Multi-Institutional Experience

03:00pm - 04:00pm ET

Poster Hall - Exhibit Hall A

Screen: 5

POSTER

Presenter(s)

Mira Patel, MD, BS - UChicago Medicine, Chicago, IL

M. Patel¹, M. J. Gutman², D. Kim³, C. Ketchum³, A. P. Sivananthan⁴, R. R. Katipally², M. Koshy³, S. Pitroda², S. J. Chmura², and M. C. Korpics²; ¹UChicago Medicine, Chicago, IL, United States, ²Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL, ³University of Illinois at Chicago, Chicago, IL, ⁴University of Chicago, Chicago, IL

Purpose/Objective(s):

Postoperative stereotactic radiosurgery (SRS) reduces local failure (LF) after resection of brain metastases; however, increased risk of leptomeningeal disease (LMD) is a prominent limitation. Here, we evaluate the incidence of LMD and associated risk factors in patients undergoing resection followed by SRS at two academic institutions.

Materials/Methods:

We retrospectively reviewed patients with brain metastases treated with resection followed by SRS from 2014-2023. Development of LMD was determined with MRI imaging and verified by a diagnostic radiologist and a radiation oncologist. Covariates include sex, location of brain metastasis, number of brain metastases at time of surgery, new brain metastases after surgery, tumor diameter, prior whole brain radiotherapy (WBRT), and LF. Tumor size was stratified by <2 cm, 2-4 cm, and >4 cm. Outcomes were assessed using a postoperative 4-month landmark time to account for SRS at 1 month and MRI at 3 months. Univariable (UVA) and multivariable (MVA) analysis was performed using Cox proportional hazards modeling.

Results:

193 patients were treated with resection and postoperative SRS of which 140 patients had follow up at 4 months. Median follow up for patients in the landmark analysis was 14.9 months (interquartile range [IQR]: 8.50–28.97 months). The cumulative incidence of LMD was 15% (n=21) with a median time to LMD of 12.2 months (IQR: 7.85-27.38 months). The cumulative incidence of LF was 20.0% (n=28) and new brain metastases was 40.7% (n=57) after surgery. On UVA, patients who developed LF (p= 0.011) or new metastases after surgery (p= 0.016) had higher rates of LMD. The number of brain metastases at time of resection (hazard ratio [HR]: 1.02, 95% CI: 0.72-1.4, p= 0.902) and preoperative tumor diameter (HR: 1.25, 95% CI: 0.80-1.93, p= 0.315) were not associated with LMD. On MVA, developing new brain metastases after surgery was associated with higher rates of LMD (HR: 3.42, 95% CI: 1.13-10.36, p=0.030), but LF was not (HR: 1.96, 95% CI: 0.65-5.91). There was a non-significant trend that patients with infratentorial tumors were more likely to develop LMD (HR: 2.15, 95% CI: 0.75- 6.14, p=0.152). Patients with a tumor diameter > 4 cm (HR: 4.53, 95% CI: 1.06-19.29, p=0.041) and new brain metastases after surgery (HR: 4.29, 95% CI: 1.76-10.47, p=0.001) were associated with higher rates of LF.

Conclusion:

Landmark analysis showed that LMD developed in 15% of patients at a median of 12.2 months after surgery. No actionable risk factors for LMD after post-operative SRS were identified. New intracranial metastases following surgery were associated with LMD, which is consistent with natural disease progression. On MVA, LF was not associated with LMD, possibly suggesting that strategies to improve LF may not improve LMD rates. Ongoing prospective studies are evaluating brain penetrant systemic therapy and preoperative SRS (NRG-BN012) to further mitigate LMD risk.