Sep 28

QP 08 - CNS Metastases and LMD: Patient Selection, Treatment Strategy and Outcomes

1044 - Proton vs. Photon Craniospinal Irradiation for Leptomeningeal Disease: Comparative Clinical Outcomes and Acute Toxicities

03:15pm - 03:20pm ET

Room 254

Presenter(s)

Audi Hoang, BS - Tilman J Fertitta Family College of Medicine, Houston, TX

A. Hoang^1,2, F. Kwong¹, D. Swanson³, M. C. Tom¹, T. Beckham¹, A. J. Ghia⁴, D. N. Yeboa¹, M. F. F. McAleer¹, T. A. Swanson¹, C. Wang¹, C. Chung¹, B. De¹, J. J. Chen¹, S. L. McGovern¹, and S. Perni¹; ¹Department of Radiation Oncology, Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, ²University of Houston Tilman J Fertitta Family College of Medicine, Houston, TX, ³Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX, ⁴Department of CNS Radiation Oncology, Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX

Purpose/Objective(s): Leptomeningeal disease (LMD) is an advanced manifestation of malignancy associated with poor prognosis and significant symptom burden. Proton craniospinal irradiation (pCSI) has been shown to prolong survival in patients with LMD, particularly those with breast and non-small cell lung cancers (NSCLC). However, volumetric modulated arc therapy (VMAT) CSI may be more widely accessible. We hypothesized VMAT CSI would demonstrate comparable survival and toxicity to pCSI.

Materials/Methods: We conducted a retrospective review of patients treated at our institution who completed CSI (30 Gy in 10 fractions) for LMD between 8/14/2021 and 8/13/2025. Proton therapy included both passive scatter (PS) and intensity-modulated proton therapy (IMPT) techniques. Hematologic trends were assessed using complete blood counts at baseline, during treatment, and post-treatment. Clinical parameters were compared using logistic regression analyses. Overall survival (OS), time to LMD progression, and initiation of additional LMD-directed intrathecal chemotherapy, radiotherapy, or surgery were estimated using Kaplan-Meier methods. We performed univariable and multivariable Cox regression examining age, gender, Karnofsky performance status (KPS), and histology.

Results: Seventy patients met our inclusion criteria, of which 47 received pCSI (IMPT = 21; PS = 26) and 23 received VMAT CSI. Breast cancer (40.0%) and NSCLC (31.4%) were the predominant histologies. The median age was 54.5 (range 17 - 79) years. Thirty percent were male, and 92.9% had a KPS of ≥70. Median time from consultation to first fraction was 20 days (IQR 14 – 26) for pCSI and 14 days (IQR 11 – 21) for VMAT. OS differed significantly between breast/NSCLC versus other histologies (Median OS: 17.2 [95% CI 12.7 - 26.0] versus 4.1 [95% CI 2.2 - 7.9] months, respectively). Among breast and NSCLC patients, CSI modality was not significantly associated with OS on adjusted analyses. Time to LMD progression, time to initiation of additional therapy, acute toxicities, and hematologic parameters were not significantly different by modality. VMAT was associated with lower odds of hospitalization during treatment compared to PS pCSI (OR 0.2, 95% CI 0.03 - 0.95; p = 0.04) but not IMPT.

Conclusion: Survival in LMD was influenced by underlying histology, with breast and NSCLC demonstrating improved OS compared to other malignancies. Among these patients, pCSI and VMAT were associated with comparable survival, disease control, and toxicity profiles. VMAT CSI was associated with lower odds of hospitalization compared with PS pCSI and a shorter median time to treatment initiation. Larger prospective studies are warranted to further optimize patient selection and delivery of CSI for patients with LMD.